USF Libraries
USF Digital Collections

Landscapes, memories and water : narratives, perceptions and policy-making on land and water in Monteverde, Costa Rica


Material Information

Landscapes, memories and water : narratives, perceptions and policy-making on land and water in Monteverde, Costa Rica
Translated Title:
Paisajes, memorias y agua : narrativas, percepciones y políticas sobre terrenos y agua en Monteverde, Costa Rica ( )
Physical Description:
Porras, Ina T


Subjects / Keywords:
Hydrological services
Servicios hidrológicos
Books / Reports / Directories   ( local )
Books / Reports / Directories   ( local )



Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
usfldc doi - M36-00417-ML-1023
usfldc handle - m36.417-ml-1023
System ID:

This item is only available as the following downloads:

Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader 00000nas 2200000Ka 4500
controlfield tag 008 s2005 xx |||| |||||| eng
datafield ind1 7 ind2 024
subfield code a M36-00417-ML-1023
1 100
Porras, Ina T.
0 242
Paisajes, memorias y agua : narrativas, percepciones y polticas sobre terrenos y agua en Monteverde, Costa Rica.
Landscapes, memories and water : narratives, perceptions and policy-making on land and water in Monteverde, Costa Rica.
g Abril 2005/April 2005.
Books / Reports / Directories
2 local
Hydrological services
Servicios hidrolgicos
Scanned by Monteverde Institute.
The State of Water in Monteverde, Costa Rica: A Resource Inventory.
4 856


Market opportunities associated w ith hydrological services in a tropical montane cloud forest Landscapes, memories and water Narratives, perceptions and policy-making on land and water in Monteverde, Costa Rica Ina T Porras (IIED-CLUWRR) Miriam Miranda (CINPE) April 2005 1


Table of Contents 1 Introduction ............................................................................................................1 1.1 Objective of this study.....................................................................................2 1.2 Methodology of analysis .................................................................................2 2 General description of the area ...............................................................................3 3 Reconstructing the history through narratives ........................................................8 3.1 Early settlers, before the 1950s .......................................................................9 3.2 Local forests transforme d into pasture, 1950-1985 .......................................12 3.3 The technological and ec otourism era: 1985-2000s ...................................... 15 4 Local perceptions of water resources ...................................................................19 4.1 It used to rain more Popular perceptions in Monteverde ........................19 4.2 Downstream perceptions ............................................................................... 24 4.3 The scientif ic evidence ..................................................................................25 4.4 Overlapping the results ..................................................................................26 5 Effects over policy ma king ...................................................................................28 5.1 Upstream/downstream compensations ..........................................................28 5.2 Markets for Watershed S ervices in Costa Rica .............................................29 5.3 Markets for Watershed Se rvices in Monteverde ...........................................32 6 Summary and recomendations .............................................................................. 36 6.1 Science and popular perceptions differ .........................................................36 6.2 Land use changes fast ....................................................................................38 6.3 Governme nt perception is rather poor ...........................................................39 6.4 A people-centered approach .......................................................................... 39 6.5 Setting up negotiations for m arkets has a long way to go .............................40 7 Bibliography .........................................................................................................43 8 Appendix ..............................................................................................................49 List of Figu res Figure 1. The Arenal Watershed ....................................................................................4 Figure 2. Upstream -Downstream relationships ............................................................5 Figure 3. P opulation centers in the upper catchment areas ........................................... 6 Figure 4. Land use changes in Rio Chi quito and Guacima l watersheds, 1960-97 ........8 Figure 5. Land use patterns until first half of the XX Century ....................................10 Figure 6. Land use patterns from 1950 to the mid 1980s ............................................14 Figure 7. L and use patterns from the mid 1980s to the 2000s ....................................18 Figure 8 P&B of the influence of cloud forests over rainfall, N=39 ...........................20 Figure 9 Markets for W atershed Servi ces: summa ry of global initiatives ...................28 Figure 8-1 Idealized soil section ..................................................................................55 Figure 8-2. Rainfall and Evaporation in the W atershed Context .................................57 List of Tables Table 1. Main stakeholders in the study area .................................................................6 Table 2. Overlapping science, people and experts .......................................................26 Table 3. Markets and Payments for H ydrological Services in Costa Rica ..................31 Table 8-1 Model results of relation of deforestation a nd rainfall (in large basins) ..... 51 Table 8-2 Soil water content (volume fraction ) ........................................................56 Table 8-3 Effects of methods of deforest ation and post-cleari ng soil managem ent on runoff and soil erosion* .......................................................................................61 2




1 Introduction Using the watershed as a unit to understa nd upstream-downst ream relationships is not new (Mourraille, Porras and Aylward, 1995). Wi thin the context of this study it means looking for optimal land use scenarios ups tream to improve on-site and off-site hydrological services. The trick is to be able to maximise the living conditions of all involved. And, potentially, genera te a new source of income from market initiatives, including Payments for Environmental Services. Changes in water flows and their quality, especially during the dry season, constitute a major problem all over the tropics. In th e particular case of cloud forests, it is generally thought that clearance of trees wi ll result in loss of the extra input of moisture from passing clouds, leading to potentially lowered gr oundwater tables and thus reduction of stream baseflow. There are however, very few cases when the scientific base feeds direc tly into policy-making. And even less often are poor people participants in the decision-making proce ss, thus limiting the potential livelihood and welfare gains Context shapes policies (Mayers and Bass 2004). Physical, cultural, political and environmental conditions, and decisions m ade in the past, will determine the way policy is determined and the effects it has. An important aspect is how scientific knowledge becomes part of the policy making process. This includes how the message is put across (and to whom), how it is understood, who benefits from the dominant belief structures, and how is it sh aped by local historical patterns and people's perceptions of "how things work". In relation to science and policy, it is possi ble to distinguish at least four possible situations that could affect the vi ability of land and water policies: 1) The scientific knowledge does not exist; it is incomplete, or imprecise. 2) The scientific knowledge exists, but it is not communicated to policy makers, and even less to communities. Even in cases where sufficient s cientific knowledge exists, poor communication of resu lts to stakeholders particularly low-income inhabitants of upper reaches of remote catchments and policymakers means that potential liveli hood and welfare gains are often not realized. 3) Policy-making and science fail to take into account local perceptions on land and water. If local stakeholders have a differe nt set of beliefs than those upon which polic y is made, then the long-term viability of land reforms could be seriously undermined by lack of comm itment. Added to this, marginalized groups might be further hindered if th ey have little democratic voice or influence in determining policy, as is often the case in rural areas in developing countries. 4) Policy-makers fail to take into acco unt the particular history of a region How has the area evolved and develope d through tim e? Who are the main stakeholders? What are their percepti ons of policy and the government? 1


1.1 Objective of this study This report focuses on situations displayed in (3) and (4), where policy-makers and science fail to take into account people's se ts of perceptions and the particular history of a region. The report draws on evidence from the Monteverde area in Costa Rica, where a large study is current ly being conducted to determ ine the links between cloud forest and water flows1, as well as the socio-economic impacts and market opportunities associated with changes in land use. While the combination of both studies will provide important base info rmation to inform a possible negotiation system among stakeholders downstream and upstream to improve watershed management, it remains unclear what thei r own perceptions are when it comes to understanding relations between land use and water, and what are the main drivers of land use changes according to local history. The report has three components: 1. A detailed review of the different stak eholders and economic activities in the study area; 2. A narrative analysis2, which collects information from local stakeholders, especially from the remaining pionee r settlers in the area and their descendants, and investig ates the historical settl ement pattern of land use changes and its relation to water resour ces. It also provides information about future trends of land use changes in the Monteverde area. 3. An analysis of the local perceptions a nd beliefs of the relation between land use and water, following the concept of "mother statements" suggested by Calder (1999) (see methodology below). The information from this study will be contribute to the design of land use scenarios dependent on the support of the different stakeholders within the watershed. 1.2 Methodology of analysis The study is based on an exhaustive review of existing documents and reports about the local history of the watershed, especia lly the Monteverde Area, and a literature review of land use and wa ter studies around the world. The main body of the study was obtained through collection of pr imary information that included: Inception workshop (August 2002) Guided observation through severa l exploratory field trips Informal meetings with remaining pioneers Semi-structured face-to-face interviews 10 focus groups3 carried out with local gr oups including dairy and coffee producers, municipality re presentatives, environmental organisations, tourist 1 For more information about this project please look at: ts/costarica/indexhtml 2 People organize their experience and th eir memories mainly in the form of narrative. For instance, a good story is the one that can be lived vicariously by others. The power of the narrative technique is to get trustworthy data about feelings and understanding of ke y people. The success of this tool depends on the responsiveness of the informants (Booth, et al, 2000). 3A focus group is a very interactive and participative da ta collection technique, and its use is highly recognized and valued by the international sc ientific community. A focus group is conformed by 6 to 12 homogeneous participants, invited to meet together in order to discuss their perceptions on a particular topic problematic under study. Focus groups must be carefully planned and impl emented by a very skilled pe rson. This document reports on 10 focus groups with an average participation of 7 pers ons per group. Previous contact with a key member of 2


board, water utilities, and women groups between January and February 2003. Communities visited include Las Nubes, La Cruz and San Luis, located on the Pacific side of the continental divide. The focus groups were used as a very impor tant tool for the discussion of narrative issues and the percep tions with respect to the "mother statements" presented by Calder (1999): Forests increase rainfall Forest increase runoff Forests regulate flows Forests reduce erosion Forest reduce floods Forests sterilize water supp lies, by im proving water quality Participants were asked what they thought we re the relations between land and water, which in mo st cases was immediately understood as forest (or its absence) and water. If necessary, participants were prom pted with a particular question ( "do you think that forest increase rainfall?") For the narrative study partic ipants were asked about three main time blocks: before 1950, around the 1980's and during the present time. Participants were asked to co mment on different issues, such as their reasons (or their family's) to come to the area, what they thought were the main attractions, limitations, role of the goverment, incentives, social serv ices, etc. In all situations participants were asked to consider how did the forest enter in the economic considerations of the time, and how this affected its value thr ough time. Participants were also asked how they thought that the hydrology of the place had changed and what were the possible causes of these changes. They were also asked what were, according to their own experience, the most pressing issues re lated to water in their communities. 2 General description of the area The Arenal watershed (41,332 ha, see Figure 1 ) lies on the Atlantic side of the Guanacaste and Tilarn mountain chains th at form the Continental Divide in Costa Rica. The upper parts of th is watershed can be subdivi ded intro three main microbasins: Ro Chiquito (9,136 ha), Aguas Gatas (2,724 ha) and Cao Negro (7,248 ha). This particular study concentrates on the upper part of the Arenal Watershed, and it includes two other sub-watersheds (Caas a nd Guacimal) that drain directly into the Pacific Ocean rather than into Lake Arenal. The economic importance of this watershed (with its extension into the Temp isque watershed area) in the northern part of Costa Rica is evident (see Figure 2 and Table 1 for description of stakeholders). the community or group was made with anticipation and a da te was set for the meeting. Most of these meetings took place during the evenings, and lasted between 60 and 90 minutes depending on the interest of the group. The groups consulted include: dairy communities of Las Nubes, San Luis and La Cruz, directive board and coffee producers in El Dos, directive board of the Monteverde Cheese Factory, municipal ity, local aqueduct, tourism board (CETAM), and women's group CASEM. 3


Figure 1. The Arenal Watershed Source: Aylward et al (1998) Weather characteristics in th e upper parts of the catchme nts4 are responsible for the existence of important areas of cloud fore st, and in turn the existence of a very important conservation community and tourism activities in the area. Private reserves cover approximately 33,300 hectares. Other economic activities in the area include livestock (dairy5 and meat), small areas of agriculture, ecotourism and small patches of reforestation. The middle parts of the watershed are mostly dedicated to exte nsive ranching and some agriculture. Farm s are mostly large and their owners live in the nearby town of Tilarn. Water is stored in the Arenal Reser voir, an inter-annual artificial lake created to feed into a system of three hydroelectri c plants arranged in cascade (known as the ARCOSA system, which provides over a thir d of the electricity produced in the country). From the hydroelectric power syst em, water flows through a private fish farm and an area of intensively irrigated farm s, mostly dedicated to rice and sugarcane plantations, before draining in to the Palo Verde National Park, an important wetland that hosts a large population of migratory birds. The wetland serves as a filter for water that drains into the Gulf of Ni coya, one of the most productive estuary ecosystems in the world, which accounts for approximately 20 percent of the total fisheries harvest in Costa Rica (Hazell et al 2001, Aylward et al 1998). 4 Climate in the upper parts of the watershed is transiti onal, where wind patterns from the Caribbean meet those from the Pacific and create a variety of microclimates. Weather conditions are a result of a combination of global phenomenon such as polar cold fronts, tropical storms and hurricanes and local phenomenon as topographic position and winds (Lawton and Dryer, 1980). Cloud forma tion is encouraged on the Caribbean cost by the westerly winds. These clouds climb the eastern slope of Costa Ricas mountains, coo ling as they travel, and arriving heavy with rain and mist by the time they reach the continental divide. 5 Dairy production is sold to a local cheese factory 4


Figure 2. Upstream-Downstream relationships LAND USEQuakers Livestock producers Coffee producers IDA settlements Private landowners Hydroelectric Projects ICE La Manguera HEP Irrigation Watershed Services Users Conservation Community Private reserves Tourism ACA Local municipality ARESEP MINAE Other Associations and groups Women groups Artisans Tourism Government Domestic Water Use Ecological Flows The lower basin is relatively isolated in terms of human habitation. There are a few landholdings, with small producers dedicated to diary farming6 and hiring out to large ranchers. Ro Chiquito, the most impo rtant population area, was a flourishing community several decades ago but was isolat ed with the construction of the Arenal dam. About 5 years ago the main source of employment an open cast gold mine was closed because of negative environmen tal impacts. A strong migration process has since occurred in the area, and at th e moment there is only approximately 100 inhabitants in the community. Services ar e limited to a small primary school and one local shop. The main water users in the lower part of the basins are the two Hydroelectric Projects: IC E-ARENAL and La Manguera (located on the Caribbean slope), a small private initiative. Water is then diverted from the Atlantic to the Pacific through an irrigati on canal managed by the SENAR A project, ending in an important area of wetlands (Palo Verde). 6 Milk is sold to Dos Pinos, a national cooperative for dairy products. 5


Figure 3. Population centers in the upper catchment areas The main stakeholders and a brie f description is presented in Table 1 For more information and details, please see P orras, Miranda and Hope (2005). Table 1. Main stakehold ers in the study area Name Character Activity Description Private Producers The quaker community Private Dairy farming, ecoturi sm Arrive in 1952 from Alabama, USA. Monteverde Producers Private Cheese/dairy products Supports sustainability by providing technica l assistance on soil conservation, awarding prizes for sustainability effors and refusal to accept new producers in areas that are not suited to dairy production. Livestock producers Private Dairy and met producers Small and large producers. Represented by the Livesto ck Producers Association Coffee farmers Private Agriculture Agricu lture (mostly coffee), associated to regional coop eratives: Coope Santa Elena and Coope El Dos. Institute for Agrarian Development Parastatal Agriculture Silvopastor al pr ogramme dealin g with the resettlement of landless peasants unto smallholdings. Foresters Private Forestry Small and med i um foresters (mostly for reforestation and wind-breaks), linked to a regional foresters association (AGUADEFOR) Main Cloud Forest Reserves Monteverde Cloud Forest Reserve NGO Forest Protection/ ecoturis m/research Monteverde Cloud Forest Reserve. Covers approximately 1 0,000 ha of cloud forests. Managed by TSC. Monteverde Conservation League + Childrens Eternal Forest NGO Forest protection/ ecoturis m/sale of WS/research Largest private reserve in CR (22,000 ha). Incentive programs for soil conservation and reforestation in the area adjacent to R Chiquito. Sales watershed services to La Esperanza Hydropower (incremental payments of $3 to$10 ha/yr over 5 years) see note apart Santa Elena Reserve Community Reserve Forest Protection/ ecotur i sm /research Private reserve covering 310 ha of cloud forest entrusted to the Santa Elena High School. Opened in 1992 as means to protect the forest and generate revenue for local people. Some Local Groups and Associations Monteverde Institute Private NGO Educati on Dealing mostly with education on 6


Name Character Activity Description environmental issues, the MVI is highly involved in local sustai nable development. Women Associations Local group Community issues Fuerza Femenina is a strong small group of local women of all backgrounds dealing with local issues of sustainable development from a household point of view. Association of Guides Local group Environmental education, guid e d tours Controls the quality of certified guides within the cloud forest. Members must be local. Tourism Chamber Local group Community issu es Groups together a wide variety of s takehold ers to tackle community issues such as overdevelopment, and to push for regulation on the establishment of tourist activities. Artisans Cooperative Cooperative Crafts Pr omotes new economic activities in the form of craf ts for its members (mostly women). International Scientific Com m unity Various Support Internationa l agencies d ealing with development, research, purchase of land, and new economic activities. Public Sector Ministry of Environment (MINAE) Government Regulatory Responsib le for approving/assigning water concessions for hydroelectricity. Assigns permits for forest cutting and oversees violations to laws. Arenal Conservation Area (ACA) Government Regulatory Local ad ministra tive unit of SINAC and MINAE, comprising 204,320 ha of Arenal National Park, four protected areas, a national wildlife refuge and a buffer zone in which sustainable development is promoted. Local municipalities Government Regulator y Managers of local aqueducts. The Municipality of T ilarn h as negotiated unsuccessfully with ICE arrangements for local people to share in the benefits of the Arenal hydropower system. Acudectos y Al c antarillados (AyA) Goverment Domestic Water Provider Provides domestic water in the Santa Elena community, and quality control advise for other local rural aqueducts. Regulatory Authority of Public Serv ices (ARESEP) Government Regulatory Defines pr ices for electricity, domestic water use, irrigation, park entrance fees and other basic service tariffs. Educational sector Public and privat e Education Community primary public schools, two secondary schools, and several language schools in the area. Stakholders: Main downstream water users Costa Rican Electricity Institute (ICE) Government Hydroelectricity Control the Arenal-Corobici-Sandillar (ARCOSA) Hy droelectric complex, that feeds from waters from R.Chiquito microbasin. Supplies a pproximately 50% of national electricity. La Esperanza Hydropower Co Private Hydroelectricity Located on the Atlantic slopes (outside Arenal Watershed) but receives water from cloud forests in the upper parts of the watershed owned by MCL. SENARA Government Irriga tion National Water and Irrigation System, feeds on water from th e ARCOSA project and supplies water for the PRAT irrigation project in Guanacaste (over 15,000 ha of agriculture (mostly rice and sugar cane)). Source: Porras, Miranda and Hope (forthcoming) 7


3 Reconstructing the history through narratives Land use has changed significantly in the study area. Forests were initially cleared to give way to pastures as the ma in economi c activity. And during the past 15 years a combination of environmental policy and ec onomic drivers seem to be reversing the trend albeit slightlyand bringing forest s back on the map (see Figure 4 ). This Section of the study tries to reconstr uct the changing pattern s of land use in the watershed through the collectiv e me mories of local inhabitants and the revision of existing documents. The methodology include d personal interviews with key local people and several focus groups in the area7. Section 1.2 presents a more detailed explanation of the me thodology. The second part of this study atte mpts to distill how they have perceived the changes in land us e and water resources in the area. This Section presents the historical settlement of the upper parts of the watersheds into three main periods of time: 1. The first period: from early settlers to the 1950s. 2. The second period: from the arriva l of the Quaker community to 1985. 3. The third period: technological advan ces in the dairy industry and the ecotourism boom Figure 4. Land use changes in Rio Chiquito and Guacimal watersheds, 1960-97 Source: J Calvo, ITCR 7 For transcripts of the focus groups please contact the authors. 8


93.1 Early settlers, before the 1950s Early inhabitants of Pre-Columbian Costa Rica were m ostly nomade tribes, hunters and gatherers. Around 500 B.C. they bega n to move from tribal societies to chiefdoms. The main indication of human activity in the study ar ea comes from the Malekus people. Very little is known of them, but pottery shards found near Santa Elena suggest that they possible crossed Tilarns range tr aveling from the Caribbean to the Pacific. There is not evidence of im portant settlements in the area, although some participants in the Focus Group in San Luis mentioned the existence of Indian burials in the area. Human settlements began to appear in the Arenal W atershed late in the XIX century. The discovery of gold in Guacimal and Aban gares attracted a ra nge of immigrants. Nowdays, it is possible to find descenda nts from Spanish, Portuguese, Italian, Ukrainian and Chinese ancestry in the area of Guacimal and Abangares. The Gold Rush was short-lived, and Guacimal begun to loose its importance and population with the collapse of the mine. However, some disappointed gold diggers decided to stay and become subsistence farmers in the area (Quirs8, 2003 pers. com). At the same time, pressure on lands in the Central Valley from the expansion of coffee and urban areas contributed to early migration processes. The upper parts of Rio Chiquito, Cao Negro, Guacimal and Caa s watersheds were settled during the early 1900s by early pioneers from the mi ddle parts of the country -mostly San Ramon and Naranjoand from Guacimal. Landscape occupation followed up the same patterns of the Costa Rica coloniza tion and the 'Family Providers' Law, 1934, which stated that land had to be cleared to prove ownership. Law and Land Conversion. The Family Providers Law, enacted in 1934, "would be the toot cause of massive destruction of forests". It awarded 20 hectares of unused public land to heads of families. This in turn led to the creation of the Land Tenancy Information Law (1941), which awarded up to 300 ha of unused forest public land to anyone who could prove effective ownership. This was done through land clearing, although not everyone had the intentions, or means, to cultivate the land afterwards. Watson et al 1998. During the 1940s, in an effort to redu ce the econom ic influence of the coffee oligarchy, the government targeted a series of new economic activities. Livestock activities were given new support. The na tionalisation of the bank system provided cheap credits, and the expansion of roads and communication systems opened new market possibilities. Additionally, low up-front costs and good international prices for meat resulted in a major commercial expans ion for cattle ranching at country level. By the 1970s, Costa Rica was the fourth larg est exporter of meat to the United States (Aylward et al, 1998). 8 Emilce Quirs, 80 years old, is the granddaughter of Batista Oliverio, one of the Italian miners who become a farmer and raised a big family in Guacimal.


10 Some Livestock Land Use: Mostly forest. Some livestock and pigs begin to be introduced. No roads, very difficult access Government policies for claiming land Subsistence agriculture Limited Land Use Impact Farmers ( colonos) Few and isolated Arrive from the Gold Rush inGuacimal(Late XIX), and Central Valley as coffee takes over land there High rainfall and permanent cloud cover perceived Management options Subsistence and mostly barter exchange economy. Forest clearing by hand therefore the impact was limited. Relatively little pressure over resources External variables Stakeholders Source: recreated from Focus Groups and personal interviews with descendents from early settlers. Figure 5. Land use patterns until first half of the XX Century


"My grandfather came to the area from San Ramn, around the years of 1850. He came first to Santa Elena but didn't like the soils, so he came down to the San Luis area and began what it is now the town. There used to be Indians here. You can still find some burials. It was a hard life. There were only traditional remedies w ith herbs. Only the strong ones survived" Focus Group, San Luis. "By the mid of the last century the price of the land was incredible cheap. My father bought 200 hectares of forested land from my grandfather for only US$0.25. Only a small part of that forest was cleared". Arguedas, J. pers.comm.2003 "It was so remote and isolated that people hiding from the authorities found a safe haven to begin a new life". Brenes9, pers.comm. All in all, the colonisation process in Monteverde and Guacimal area was very slow and isolated. Cattle ranching was done in relatively small scale before the 1950s. During this time only the middle parts of th e Guacimal watershed were linked to the rest of the Country's economy. The upper parts of the catchments remained covered by cloud forests, and the few families living there subsisted from small patches of agriculture, small pig farming and a few h eads of cattle. Distances were long and difficult, and with no bridges, roads and access only through horse trails, reaching out to external markets was a near impossible task. The problem was particularly acute during the rainy season because of the intensity of the rivers' flow (Suarez, 2003 pers. com). "Times were very hard. Santa Elena was hopeless. The grass was bad. Agriculture was bad. The only thing to do was dairy farming, and it didn't give much anyway. Milk was only sold to the cheese factory and coffee to Beto Len. We had to get the milk out on a horse. It can now be done on a truck and it takes one a half hours. Just imagine how it was before". ( Focus group, San Luis). "When I arrived to La Cruz 50 years ago life was very hard. There was a lot of forest, and a lot of poverty. Houses were built with whatever local materials we had. Even roofs were made of wood and only a handful of houses had metal roofs. The houses had earth floors. No electricity. There was only one radi o. There wasn't even a mill, and things were made using an axe. People didn't have much vision. Some made illegal rum, the farms were patches of grasses with two or three cows. Large patches of forests were slashed and burnt to grow a bit of maize for subsistence. People survived by barter exchange. There were no jobs, and if anyone needed a labourer, it was paid by retu rning the labour. Perhaps only three persons would hire properly, and the Quakers, but we didn't have Quakers here. You could make about 1,50 colones for a hard-working day, and anyone with a thousand colones in his pocket was considered a millionaire. Anything to sale had to be taken out in horses. You couldn't sell anything in Santa Elena, you had to go all the way to Las Juntas". Focus Group in La Cruz. "My sister Louvigina got married and went to live in Monteverde (before the 1950s). She used to say that the humidity was permanent, and everything was always wet. It rained all year round, it was very misty, and sometimes it was impossible to see anything at all. Sometimes the sun would creep out, a little bit, and everybody would be cheerySometimes the temporales (rainstorms) would last for days without stopping, four or five times a year, and the rivers would be so swollen that we wouldn't be able to cross them. The coldness was unbearableone had to go with several layers of clothing all the time". Fragment of an interview to Rosa, daughter of Bastista Oliv erio, an Italian immigrant who arrived to Guacimal in 1902. 9 Brenes is descendent from one of the La Cruz first sett ler. He arrived to thi area when he was 4 years old. 11


The general perception of desce ndents fr om early settlers in the area is that landscape was densely forested, cloud cover almost permanent throughout the year, and precipitation, in the form of rainfall or mi st, was stronger than in present times. The impacts that such isolated form of living had on the environment was relatively small. The landscape remained forested in the uppe r parts, and land use changes were small due to isolation. Figure 5 presents a summary of this period. When the Quaker community arrive d in 1950 there were probabl y 10 other families at most living in the area (Arguedas, 2003 pers com). 3.2 Local forests transformed into pasture, 1950-1985 Attracted by the generous land laws in Co sta Rica, the Quaker community arrive d to the Monteverde area during the early 1950s10. Arriving from Alabama, they bought 3,750 acres of (mostly) forested land from lo cal families. A new stage of the region's settlement process began with their arrival. The process, a combination of their enthusiasm, engagement with existing local families and an array of national level policies, resulted in larg e scale land conversion. "The Quakers cut a lot of forest s and turned them into pastures, but there was already a lot of deforestation. In the last part of the 1980s the pastures changed again. Many pasturelands became hotels, other people bega n reforestation, an d others abandoned plots that quickly became secondary forests. A large part of Sa n Luis was pastures, and now it is secondary forest". Focus Group, San Luis. Large areas of forests gave way to agricu lture and dairy farming. Better shelters and roads began to be built, and the production of cheese, instead of fr esh m ilk, facilitated the access to external markets. Because road access was easier on the Pacific, the Guacimal watershed was clea red earlier than the Rio Chiquito Watershed (see Figure 4). While livestock dominated the new land uses, other economi c activities such as coffee, begun to be introduced in the area as alternative crops. Coffee was introduced during the 1970s, at a time when the high international prices due to the frost in Brasil made the activity very profitable (see Porra s, Miranda and Hope (forthcoming), for indepth description of each activity). At the same time, as early as the 1950s, the scientific interest in the biota of the Monteverde cloud forests began its ever-i ncreasing trend (Koens, 2003). From the tim e when the first Monteverde scientific studies were published in magazines and bulletins in the USA and Europe the area quickly became the must visit place for cloud forest researchers, some of whic h later become active members of the community, playing a key role in Monteverde 's development. According to Nadkarni (2000), there are over 250 publications a bout Monteverde, written between 1966 and 1995. 10 Part of the Quaker faith is the belief that there is an inner light in everyone and that this inner light is in essence a piece of God. Most Quakers consider themselves pacifists. The Quakers founded the Cheese Factory, the Friends School and in an attempt to protect the areas waters hed purchased much of the land that now makes up the Monteverde Reserve. The Quakers have played a major role in the development of the community and this is one of the things that make Montever de a special place (Adapted from ) 12


13 Even during this process of drastic changes in the landscape, the Quakers recognized the im portance of the cloud forest for the provision of permanen t, stable and clean water (Pateman, 2002). They set aside forested areas near the headwaters of the Rio Guacimal, to remain undisturbed and safeguard the water source of their hydroelectric project. In 1972, and to pr event land use changes in the surrounding cloud forest, visiting scientists George and Harriet Powell joined forces with longtime resident Wilford Guindon to establish a 328-h ectare wildlife sanctuary. In 1975, the community watershed reserv e received a grant of US $40,000 from the World Wildlife Fund, forming the initial core of the Mont everde Cloud Forest Biological Reserve. The creation of private reserves for the c onservation of cloud forest became a magnet first for scientific tourism, and then exte nded to include a wide range of ecotourism that changed dramatically the econo mic drives of the upper watershed. The Cloud Forest area attracted scientists, especially drawn to the small and brilliantly colored Golden Toad, discovered in 1964 by the Orga nization for Tropical Studies. This specie was not only new but also endemic, and became the "l ogo" to encourage cloud forest preservation. Sadly, the Golden Toad seems to have disappeared. Downstream, the Costa Rican Electricity Institute (ICE) began the construction of the Arenal reservoir during the late 1970s. The project created an ar tificial lake that diver ts water from the Atlantic to the Pacific side of the continental divide. It currently has three hydrologica l plants that provide over a third of the countrys total capacity. An immediate effect of this projec t was the displacement of local towns and ranching activities, especially in Tronador a, onto the higher and steeper slopes of the upper areas of the watershed. Prior to the construction of the dam, ICE commissioned an environm ental impact assessment (CCT, 1980), which suggested the St ate to purchase lands in Rio Chiquito, where the terrain was more rugged and prone to high sediments. Unable to do so, partly because of the cost but also becau se of high opposition from local ranchers, ICE implemented a 50-m wide riparian buffer strip around the reservoir (FernandezGonzlez and Aylward, 1998). During the 1980s the Arenal Forest Reserve was created (later renamed the Arenal-Monteverde Protected Zone). This reserve included most of the land not yet occupied of the Cao Negro and Aguas Gatas microwatersheds on the Atlantic. Thes e areas became part of the Arenal Conservation Area (ACA), when the country was divided into Conservation Areas during the 1990s. Kauch and Tosi (1989) suggest that part of the colonisation of this period is the result of land speculation, wh ere farmers claimed land with prospects of selling it back to the government for protection.


14 Livestock Land Use: Forests give way to pastures Relatively small local population Environmental efforts to protect the cloud forest begin No roads, very difficult access No local municipality World meat markets encourage conversion to pastures in lower parts of watersheds Subsistence agriculture Early forests reserves Land Use Impact Farmers Quakers Early scientistsCloud Forest becomes a magnet for international scientists Management options Monopoly of the cheese factory checks expansion of the industry. Coffee is introduced Early attempts to protect cloud forests MCFP began in mid 1970s DOWNSTREAM ICE hydroelectric project Coffee Increased pressure over resources Rapid deforestation Water pollution from coffee and livestock Erosion and sedimentation Productivity losses in new soils mean that new forests are taken down Increased competition for water resources for downstream users High rainfall and permanent cloud cover perceived Stakeholders External variables Source: recreated from Focus Groups and personal interviews with descendents from early settlers. Figure 6. Land use patterns from 1950 to the mid 1980s


As for the previous period, participants in the focus groups and pe rsonal interviews in the area perceive that rainfall was stronger during this period than nowdays. However, several environmental impacts of the land use patterns com mence to emerge. According to the participants, the rapid deforestation that took place in the area produced increased erosion and sedimenta tion, and there was a need to expand the agricultural frontier of ten due to losses in soil productivity. The coffee and dairy industry ran unchecked and wastes were thro wn in rivers damaging the water quality. Downstream users began to increase with th e introduction of the Arenal hydroelectric project and the expansion of new populat ion centers in the mid parts of the watersheds. The isolation of the upper parts of the watershed contribut ed to the creation of an independent spirit and the ab ility to cope with situa tions on their own. While the governm ent provided other regions with elect ricity and water, Monteverde developed its own small hydroelectric project and water service. Other servi ces, such as civil registry, had to be done in the nearest city, Puntarenas, several hours away. By 1984 the national census only reported 400 pe rmanent inhabitants in the area. Overall, the dynamic spirit characteristic of the Monteverde inha bitants started during this period. Parallel to rapid changes in land use was the increasing awareness for water and biodiversity conservation. The la st one, alongside th e recognision of the m ajestic beauty of the cloud forests, became the central pillars of a next historical period led by of technology and tourism. 3.3 The technological and eco tourism era: 1985-2000s The next period in Monteverde is char acterised by important technological im provements in the dairy activities and the boom of ecotourism as an economic activity. The seeds for environmental awareness we re planted during the historic period described previously. Non-governme ntal orga nisations (NGOs) play ed a key role in the transformation process of local landscapes towards rehabili tation of degraded forests. At the same time dairy producer s -organised by the Monteverde Producers group and partly pressed by local NGOsse t in motion a set of changes aimed to increase land productivity and reduce pressure to convert more forest. The coffee industry also began to improve their pr oduction process and reduce environmental damage, while appealing for internationa l recognition and access to new niche proenvironment markets. Some of these technological breakth roughs included improvement in the herd (for dairy), innovative soil conserva tion practices like windbreaks, the introduction of new grass or crop varieties and grains to complement animal diet (see Porras, Mira nda and Hope (forthcoming). Most of the initial tourists in the area were scientists. Visitatio n increased as the area became better known. However, and perhaps luckily, the difficulty of access meant that it was only real "nature-lovers" who arrived. Until the 1990s there were very few public facilities in the area, but the si tuation changed cons iderably with the 15


strengthening of conservationist NGOs (suc h as the Tropical Science Centre and the Monteverde Conservationist League), whos e fundraising campaigns showed the area's beauty and biodiversity to the world, a nd with the introduction of the tourism incentives by the Costa Rican government11. The ecotourism booming era was kickstarted in Monteverde. A range of livelihood combinations bega n to emerge. Som e locals gambled and completely changed their main economic activity from dairy to ecotourism. Others decided to explore the new activity alongside their regular fa rm activities. But for all, isolation was a thing of the past. The tourism industry became the main source of income for the region. New jobs stimulated immigration to the area, m aking the population grow from 400 to over 6000 in less than twenty years (INEC, 1984, 2000). The new Monteverde comprises mini-communities within itself. Perman ent residents, both Costa Ricans and (mostly) USA citizens, include dairy farmers, hotel and restaurant owners and workers, scientists, volunteers, religious groups, etc. Seasonal visitors include the large amount of tourists and migratory work forces. The cultural life began to change and according to some locals, Monteverde has more the characteristics of a city than a rural area, with inte rnational cuisine, pubs, and jazz concerts if one can navigate through the heavy rain, mud, and bumpy roads!. In 2001 Monteverde received approximately one million tourists (personal communication, CETAM, 2002). Projections from CETAM indicate twice as ma ny by the year 2010. Projects like the paving of the main access roads seem doublededge, as it could increase even more the number of daily visitations to the area. Threats to the cloud forest. The cloud forests in Monteverde face serious environmental issues. One of the most urgent problems concerns the existence of the cloud forest itself; studies have shown that the cloud has been lifting, providing less mist an d rain to the environment beneath. There are two hypotheses explaining this lifting: global warming and deforestation. In a study published in NATURE in 1999, Pounds et al. asserted: The biological and climatic patterns (in Monteverde) suggest that atmospheric warming has raised the average altitude of the base of the orographic cloud bank. This conclusion is built on evid ence that evaporation from warm o cean surfaces released heat as it condensed, accelerating atmospheric warming and de creasing the difference in temperatures between the lowlands and the highlands. In response to this the cloud-formation height, which is dependent on relative humidity surfaces, has moved up. The study also showed that the increasing dry periods in Monteverde are associated with warm episodes of the El Nino/Southern Oscillation. This study provided much evidence to support the lifting-cl oud-base hypothesis and gl obal warming has been widely accepted as at least part of the cause. Another theory advanced to explain the lifting cloud is that of lowland deforestation. In a study published in Science, Lawton et al. (2001) advanced the theory that reduced evapotranspiration after deforestation in tropical lowlands decreases the moisture content of the air mass flowing up the slopes of the adjacent mountains. This increases the lifting condensation level and thus the elevation of the cloud deck. The model results thus suggest that deforestation in the lowland tropics of the trade wind zone tends to shift the cloud forest environment upward in adjacent downwind mountains. This theory does not contradict the global warming results; in f act, its authors maintain that the two theories are complementary. Both causes could have an effect on the moisture levels and thus height of the clouds, and both recognize the danger of the cloud deck disappearing altogether. 11 Since the 1990s tourism has received important incen tives from the government, for example preferential credit, tax exemptions, and educational programs oriented to tourism careers. 16


17 The fast and spontaneuous growth has had important economic, social, and environm ental consecuences. According to Koens (2003) there is a remarkable gap in social and environmental aspect although economic variables are positive. Even though ecotourism is the main source of inco me, issues of sustai nability are still weak. A recent study by the the Monteverde Institute (Kim et al, 2003) found that, although drinking water12 is of good quality, there are nevertheless significant levels of pollution in rivers resulting from the disposal of untreated do mestic and industrial waste waters. A meeting of local stakeholders held in 200213, confirmed during the focus groups that one of the main concerns related to water was availability and quality for drinking purposes. Nitrate c oncentrates and the potential harms of uncontrolled urban expansion were signalled as some of the main threats to water in the region. Little concern was expressed a bout sediments or soil erosion problems in the upper parts of the watershed. This, howev er, remains an issue for the downstream hydroelectric project. While the upper parts of the watershed w itnessed the ecotourism boom, further down in the watershed the conservation versus development debate was taking another direction. Most of the lives tock activity has been histor ically concentrated in Rio Chiquito, where land use maps show th at the area under pasture has increased significantly since 1960 (Aylward et al 1998). However, Bolaos (1995) suggests that there has been no significant change since the early 1980s, mostly because most of Rio Chiquito was already occupied by then. Ranchers received heavy critisism from conservation groups. In 1993 the Arenal Conservation Area (ACA)14 was created and began a watershed management plan. Livestock activity was made a target because of its shortterm private land use, and conversion from pasture to forest was st rongly advocated, to the almost general opposition of ranchers (Fernndez-Gonzlez and Aylward, 1998). This "antipastures" agenda was not supported by recen t hydrological studies conducted in the area. A series of extensive biophysical a nd socio-economic studies in Rio Chiquito showed that "pastures fare s better than forest in comp arison, and even in the upper watershed, cloud forest land, where water sources for the reservoir originate, the hydrological impact could be improved by interspersing pasture where forest now reigns" (Fernndez-Gon zlez and Aylward, 1998). However, scientific evidence does not always figure high in the defining factors for policy-m aking, and the popular belief that forests are good for water. These issues will be discussed in the next Section. 12 Drinking water in the area come s from water springs and wells. 13 The stakeholder meeting was held on August 29th 2002 during the Initiation Work shop of this Project. See Footnote 1 for more details. The cons ultation included producer and consumer gr oups: coffee, dairy, civil society, women groups, tourist board, cheese factory, plus more formal organisations and institutions, including local municipality, water board, ICE, FUNDECOR, FONAFIFO, PRAT, MINAE and others 14 ACA was created building on existing reserves and incl uding new forest areas with the economic help of WWF Canada. Additionally, ACA promotes su stainable development in the area in agroindustry, reforestation and ecotourism.


18 Livestock Land Use: Natural forests gain value as ecotourism takes over Coffee remains relatively small but it spins into organic and environmentally-friendly activity Massive migration to the area Subsistence agriculture Forests reserves and areas dedicated to Ecotourism Land Use Impact Farmers Quakers Scientists Management options Ecotourism takes over local economies. Technological innovations to improve productivity and reduce environmental effects of dairy and coffee industry. Reforestation begins Downstream : ICE and Irrigation Coffee TOURISM Roads begin to improve, but not much Government incentives for tourism, reforestation and forest protection Decline in world meat prices First local municipality in 2000 Rapid population growth Seasonal workforce for tourism and coffee harvest Large number of tourists Local groups Lower levels of rainfall and cloud cover perceived Reforestation Increased pressure over resources Deforestation halts, and secondary forests begin to grow. Water pollution from coffee and livestock begins to be controlled. Unchecked waste waters from households, hotels and restaurants become a major local problem. Higher demand for water reduces river flows, but this is mostly attributed to deforestation in earlier period. Sedimentation and nutrients into Lake Arenalcould present a problem for the dam. External variables Stakeholders Source: recreated from Focus Groups and personal interviews with descendents from early settlers. Figure 7. Land use patterns fr om t he mid 1980s to the 2000s


4 Local perceptions of water resources "There can never be water without trees. When I was a child there was forest, and water, and animals. During my youth I saw it all disappear with deforestation and forest fires that dried up everything. Nowdays, there is more protection, and water flowsare coming back again. I believe that if there are trees, there is water. CETAM Focus Group, Monteverde This section of the Study looks in more detail at how do local people perceive the relation of forest and water resources, accord ing to their own experiences. The information was collected through focus groups in the communities and personal interviews. Participants were asked what th ey thought were the functions of the forest with respect to water resources. If necessary, they were prompted with questions related to forests and their relation to rainfall, runoff, flow regulation, erosion, flood control, and water quality. See Section 1.2 for more methodological details. 4.1 It used to rain more Popular perceptions in Monteverde The general feeling from the pe rsonal interviews and focus groups is that local people feel that it u sed to rain more in the past. The "past can be described as 20 years, 50 years, or more, depending on the ag e of whoever tells the story. "Nowdays, it rains less in summers than 30 years ago. But now the government* has bought and kept more forest, there is a tendency of more clouds now, and it gets colder again". Focus Group in Las Nubes. *Note: there seemed to be confusio n about the private reserves and the government. "When I was little the creeks used to fill up a lot more, especially during the rainy season when it was impossible to cross it. From about 15 years to now it hasn't fill up that much". Focus Group in Las Nubes. According to participants, the main functions of the forests are: Forests increase rainfall (and produce water) Forests capture mo isture from the clouds; Forests reduce evaporation fr om rivers, springs and ponds; Forests regulate f lows after rain events; Forests increase inf iltration of water, and regulate water during dry events; Agroforestry increases productivity of soils 4.1.1 Forests increase rainfall and produce w ater In the particular case of Monteverde, Cost a Rica, participants of the focus groups are adamant in their po int of view that forests have a definite effect on rainfall. According to participants in the focus groups, the ways by which forest affect rainfall are through recycled evaporation, allowing the formation of clouds that become rain again ( "it's not only the sea that produces rain") and by trapping the clouds and fog coming from the Atlantic. 19


"It's a fact that it rains less nowdays, especially during the dr y season. But the protection and increase of forests in the upper lands means that there are more clouds now, it gets cooler and rainfall is beginning to increase again. And this affects pastures. It's not good for milk production, pastures are not good quality and agriculture is almost out of the question because it rains too much". Rancher in Las Nubes "The forest reserves in the upper lands help with the moisture through maintaining a cloud base and increasing rainfall". Rancher in La Cruz "One of the main functions of our forests is to capture the clouds coming from the Atlantic". Monteverde Producers Executive Board Focus Group "Forest maintain moisture and convert evapora tion back into precipitation, like a sponge". "I believe that there must be a positive relation between forest and rainfall. Even if altitude and life-zones might affect, it used to rain more before, when the country had more forest as a whole" Independent policy advisor, San Jos. For the majority of participants in the focu s groups there is a direct relation between forest cover and rainfall (" the re is more rainfall when there is more forest "). Over half of the survey respondents (56%) belief that cloud forest produce more rainfall than any other type of land use, including other types of forests, forest plantations and agriculture (see Figure 8 ). Figure 8 P&B of the influence of cl oud forests over rainfall, N=39 28% 56% Same More All participants in the focus groups perceived that in the past sunny days were rare, as clouds were constantly over the region a nd there was a lot of horizontal precipitation even if there was m ore pasture than toda y. Some participants believe that while global warming have affected the current (lower) precipitation le vels, reduced water levels today are also a consequence of pa st deforestation. The positive effects from current reforestation and forest protection m easures take some time to make effect and will be seen in the future. There is, however, a growing perception (esp ecially outside the cl oud forest area) that forests do not necess arily have a strong influence over rainfall, which is defined by other environmental conditions, lik e the existence of seasons. 20


4.1.2 Forests reduce evaporation of water bodies According to the focus groups, one of the functions of the forest by which they help ma intain water flows in the system is through the prevention of evaporation from springs; by providing shade and shelter and maintaining the humidity in the ground. Deforestation exposes water bodies to the sunlight and increases evaporation, and therefore reducing water flows. "Forests keep the moisture in the soil and rivers by providing shade to water springs and preventing evaporation". Tourist board member "Even if forests don't increase water levels, they provide shade and help reduce fires which have effects on the water. The lower parts of the watershed don't have trees and you see fires all the time, and it is very dry". Women's group CASEM "Forest keep moisture and therefore allow water to exist. There are three springs in our family farm. Until some years ago our father would cu t down trees without thinking, but now we know that it will reduce water. We have started reforesting with trees and banana plants. Women's group CASEM. 4.1.3 Forests regu late flows after rain events General perception in the upper part of the watershed is th at forests help regulate and soften the impact of rainfall, com pared to other types of land use and reduce immediate runoff. Participants argued that rive r levels after events of high intensity of rain take longer to peak when there is forest when compared to open pastures: "It's easy to see the effect of fo rest over runoff just by looking at the rivers coming from the Reserves after heavy rainfall. When it rains the rivers recharge slowly but when they are full they take longer to go down to their normal levels. In Rio Chiquito, on the other hand, where there is less forest, rivers rise and fall again very quickly after the storm". Rancher, Las Nubes de Monteverde "Forests absorb rainfall and reduce the speed of water runoff. But also improved-agriculture could get similar results" Independent policy advisor, San Jos "Forest plantations with "thirsty" species, such as conifers, will help to dry-up the soil and reduce surface runoff during an d after rainfalls. You can se e this happening when you compare pine plantations with those areas with pastures or, even worse, urban areas with pavements and roads. School teacher, San Jos 4.1.4 Forest increase infiltration of w ater The sponge effect, where it is believed that forests' roots will soak up the rain and slowly release it back to the rivers, or "guide" it through the soil to the underground reserves where it com es back again as spri ngs, is strongly intertwined into people's set of beliefs: "Forests soils infiltrate water" "Forests act as an interception filter: collect s water and inserts it back to the rivers". 21


"Forests are a sponge. Forests capture water from the clouds, infiltrate it, and the water comes back up in the lowlands. Water springs have their origin in upland forests". "Forests keep and maintain water flows. Where th ere is no forest, everything else is dry and barren". Quotes from Focus Groups on infiltration There were some reports on experience of tree species actually decreasing water flows, although this was more of a "lone cr y" than general feeling. One woman in the CASEM focus group explained that some trees "suck-up" the wate r, and other types of trees help infiltrate and increase water in the springs: "Trees like guachipeln will reduce water. Other trees, like Higuern, with all the broad leaves, protect against the sun. You know the popular saying: where there's an higuern there's water". Native species are good. Women's group CASEM. There are some local studies that put forward the sponge e ffect, although th is is still debated by the international scientific community. Ortiz (2002), presents results from experiments which support the sponge effect theory, arguing that forested lands in the exercise have greater underground water storage capacity than pasture areas. Nevertheless, the study has recei ved significant cri tique about the experiment design which puts the results in doubt. These i ssues are later discussed in Section 4.3 and the Appendix 4.1.5 Forests help regulate wa ter in dry events According to a representative from the Aqueduct, deforestation will reduce the quantity and quality of water, with particul ar effects during the summer. According to the comm unity aqueduct, Monteverde has sufficient resources to supply water for its population, even with a 14% population grow th rate (CETAM, focus group). While the authorities do not expect problems s upplying water for private homes and hotels, they foresee a possible reducti on in streams due to an incr ease in water use, but also from deforestation. Most of the water shortages could be exp ected during ex treme events, such as El Nio. The community already experienced some problems during 1994 and 1998, and according to the Aqueduct there might be some shortages during the dry season of 2004. Most of these problems tend to happe n in the middle to lower parts of the watershed, where river levels drop to about 25% of their normal levels. It is interesting to note that nobody from the uppe r parts of the watershed mentioned water shortage during the focus groups. 4.1.6 Agroforestry systems increase prod uctivity Agroforestry systems began to be intr oduced in Monteverde during the mid 1980s, mostly as windbreakers15. Due to altitude and the wind patterns, in December and January the area suffers from very strong winds coming from the Atlantic, with negative impacts on wind erosion and dairy productivity. The introduction of windbreak reforestation projects while initially rejected ( La Cruz Focus Group) is 15 See Porras, Miranda and Hope (2005, forthcoming) for more information about windbreaks in the area. 22


now regarded as highly beneficial. Accord ing to farmers in the focus groups, it is estimated that dairy productiv ity has increased approximate ly 20% per hectare. While farmers perceive a reduction in usable past ureland due to the in troduction of trees, the benefits in increase productiv ity far outstrip the costs. Tr ees as windbreakers are seen as providing a series of benefits, incl uding reduction of runoff, increase in soil productivity, protection of pastures agai nst wind erosion and generation of other benefits such as fruits, medicine plants, bi odiversity corridors, and, very important for local farmers, improvements in landscape beauty. In the past 15 years the Coopera tive has tried to help with the reforestation process, with the help of incentives from the Government*. We have planted about 35 to 40 thousand trees per year since 1989. We help farmer s obtain government incentives for this, and work with them to install agroforestry system, such as shade coffee and windbreaks, using native species such as por, avocado, citrics, and trees that provide households with timber. There have been important effects, such as reduction of wind and runoff erosion, increased soil productivity, and better landscape beauty. Ther e is less need for herbicides. El Dos Focus Group note: incentives from the Government refers to payments for environmental services. Not everybody agrees with this. Some partic ipants in Las Nubes believe that, because they are located in the upper parts of the m ountains, trees enter in direct competition for light, and the resulting qua lity of pastures is very poor and this has negative economic for the farmers: we know that trees are good for water, but we wish we could cut them and improve th e productivity of our land (dairy farmer in Las Nubes) An interesting aspect is the dual perception of s oil quality within the forest. For some participants, forests usually have good quali ty of soil because of all the layers of organic matter from the trees (San Luis Focus Group), making newly converted lands into highly productive areas. Th is concept was expressed as tierra nueva (new soils). Other participants expressed that while it is true that erosion is nearly nonexistent in forests, all the organic matter th at makes new soils so desirable actually comes from the canopies, and if the area is de forested it would quick ly lose its fertility and become barren, and produce a lot of sedi mentation because of the topography and sharp slopes of the area ( Municipality Focus Group) This is the moment when, according to participants in San Luis, these areas would turn into tierras viejas (old soils) and it became necessary to deforest new areas (volcar montaa otra vez ). Some agroforestry systems are good, but some fast-growing plantations like cypresses are not considered beneficial b ecause they dry up the soil. Localized efforts are being made by the Coffee Cooperative El Dos to in troduce reforestation with native species, as learning experience taught them that so me exotic species re duce water supplies. It is important to rescue the credibility in reforestation projects. Some years ago it was the boom about eucalyptus and pine, and later on we were told that it was actually not good for water. We need to learn more about the tree species. The problem is that even if there is research done into this, it doesnt come to the communities but stays in offices and desks. (El Dos Focus Group. For some participants in the focus groups living in the mi ddle or lower parts of the watershed, it is difficult to engage in refo restation because it is expensive, but also because trees will dry up for lack of water and pest attacks. 23


One member of the Aqueduct expressed that although the effect of reforestation was possibly imperceptible or non-ex istent in terms of water levels, its importance was in terms of reduction of erosion and acting as a barrier in water springs. 4.2 Downstream perceptions 4.2.1 Hydroelectricity Due to the inter-annual nature of the Aren al reservoir, ICE pr oduction engineers have advocated the view that forest cov er is irre levant for water flows and in fact, the area in the watershed "might as well be paved", without negative consequences for the hydroelectric services (Fernndez-Gonzlez and Aylward, 1998). This argument has had support from local ranchers who argue that well managed pastures can promote runoff and prevent erosion and sedimentation. For obvious reasons, conservationist groups oppose this view, and substantial te nsion exists in the area. Supprting this view, an in-depth study coordinated by the In ternational Institute for Environment and Development, the Tropical Science Centre and the Universidad Nacional (Aylward et al 1998), indicates that: "pasture fares better th an forest in comparison, and even in the upper watershed, cloud forest land, where wate r sources for the reservoir originate, the hydrological impact could be improved by interspersing pastur e where forests now reign". NEED TO UPDATE THIS WITH THE INFORMATION THAT MIRIAM IS GETTING FOR US. 4.2.2 Irrigation Water from the reservoir feeds into the largest irrigation project of the country (SENARA). Largely, land use in the upper part s of the watershed is not perceived as a major threat to the supp ly of water flows, and their main concerns are related to infrastructure and flows management. According to representants from the irrigation project16, while the institution does not have a defined environmental policy, it is in terested in making sure they are able to receive (and then supply) the required water flows. The main environmental threats in the watershed SENARA perceive are: 1. The absence of an integral watershed ma nagem ent plan, which results in water pollution. Local municipalities (Caas and Tilarn) do not have regulation plans, or political will, to protect water resources. Pollution comes from direct discharge from septic tanks, and wastewaters from domes tic, industry, and dairy farms. 2. Pollution fr om wind-driven garbage from landfills located nearby the irrigation canals. 16 Interviews with Nora Pine da. Regente Ambiental. DRAT. and Roberto Spesny. Departamento de planificacin Sena ra. December 2004. For more details on the irrigation project see Porras, Miranda and Hope (forthcoming). 24


3. Excess of water in the drainages resulti ng from inefficient use of water. For example, La Mula creek, which cuts through the Palo Verde Wetland National Park has lost the seasonality characteristics (dry in summer, wet in winter) required for the wetland. The excessive use of water means that the creek is always full. 4. Deforestation and illegal logging in the upper parts of the watershed result in losses of soil fertility and increa sed sedime nts in the drainages. 5. Water pollution with pesticides from ir rigated farm s could get in the Wetland and have negative environmental impacts. The main environmental services SENARA expects are not land-use related. Better water quality is m ostly related to the aut horities either establishing new guidelines for wastewaters, or making sure that ex isting regulations are followed. SENARA also needs constant water supply from the reservoir throughout the year. Water for irrigation depends on the energy producti on from ICE. Currently SENARA needs between 42 to 70 m3 of water, de pending on the season. SENARAs water requirements during the dry season are eas ily met because the hydroplants operate every day. Their problem is the wet season, when hydroelectricity production is not constant in the Arenal Reservoir and water is not passed through the canals. Charges for irrigation are extremely subsidised (approximately $45/ha/yr) and the organisation is not able to generate the funds it woul d need to make water storage tanks and regulate flows. A recent study by Pineda, Environmental Regent of DRAT, suggests that the tariff should be approximately $65/ha/yr. The main environmental service that SENARA receives is water, and the institu tion is not particularly interested in other envi ronmental services (s uch as biodiversity, landscape beauty or carbon seque stration). Environmental issues are not a priority for the institution. Their efforts are concentr ate on irrigation systems, cleaning existing canals, and saving water flows. However, the organisation takes ac tive part in local and national water-related discussions. It has an environmental regent who gives support in environmental issues. The ideal water use in the upper parts of the Wa tershed is forestry cover in riparian areas and high slopes, improved pastures and organic agri culture with soil management. In the lower parts of the watershed the Institu tion would like more efficient water use and organic production. 4.3 The scientific evidence The Appendix presents a review of the scien ce behind land use, forest and water resources. In short, the main effects from forests with respect to watershed environmental services are (Calder, 2002): Experiments from catchments indicate lo wer runoff from forest compared to other areas under shorter vegetation. Howeve r, particular characteristics of tree species and soil types will affect the degree by which evaporation and transpiration affect runoff. Competing effects from increased infiltration and higher evaporation and transp iration from trees could result in higher or lower dry-season flows, and 25


the effects are likely to be specific. Afforestation will most probably not lead to higher dry season flows. While natural undisturbed forests might ha ve lower rates of erosion, the effect from disturbed forests or forest pl antations could be the opposite. It all depends, at the end, on m anagement techni ques and tree species that minimize soil impact. A quick internet-based consultation about the linkages between forests and water done by the author using the Rim anchik networ k based in Per produced 15 responses. Answers came from forest engineers a nd NGO advisers. The results indicate that, according to this group of experts, the links of forests and water are: o Forests do not necessarily increase rainfall, unless cloud forests by increasing horizontal precipitation. o Surface runoff is reduced to a certain extend th rough increases in infiltration rates. However, only the existence of organic matter will help improve soil porosity. Big conifer forests, for example, that do not allow for other vegetation to grow, will not stop runoff and possibly increase it. o Forests reduce erosion and sedime ntati on, especially through the existence of different layers of vegetation that brea k down the intensity of rainfall drops, and soil erosion through heavy winds. o Forest control small-scale flood events. o Forests increase water qu ality through re duced levels of sedimentation and by lowering water temperature with shade, therefore reducing formation of microorganisms. o Forest cont rol soil structure with their roots, theref ore soil quality is better. o Most people me ntioned that forests in crease infiltration rates making more water available duri ng the dry season ( sponge effect) o Water flows in rivers increases with fo rests. Only one response indicated that fl ows will decrease because trees capture water before it reaches water bodies. 4.4 Overlapping the results If land use policy should be based in scie nce, then the question is whether or not science overlaps with people's perceptions on forests and water. Comparing the results from local perceptions in Section 4 with that of scientific research in the Appendix it seems at first glance that there is a cons iderable gap between experts' opinion, general public, and science. However, what seem s to be widely known in the expert community is not necessarily reaching dow n to land stewards in upper watersheds. It is important to stress that while there is a significant amount of cloud forest in Monteverd e, and science indicates that in th at particular case forest could increase precipitation through the capture of fog, most of the perceptio ns of inhabitants of the upper and middle parts of the watershed rega rding forest effects on water extended to other types of forests as well. Table 2. Overlapping science, people and experts 26


People's P&B Expected Effect (2)Science Experts' opinion Forests increase rainfall. Deforestation decreases rainfall. Especially in the case of cloud forest s, the expected added benefits from horizontal precipitation contribute to higher streamflows, with added importance during the dry season. Rainfall is not likely to be infl uenced by forest, unless at the very large scale. Even cloud forests contribution, through horizontal precipitation, is relatively small. Similar to science. However, cloud forests ar e perceived as contributing to water flows, especially during the dry season. Forests reduce surface runoff, but increase infiltration. The root network and low soil compaction fro m forests creates more permeability in the soil and prevents water reaching the streams and leaving the watershed too quickly. Forests reduce runoff. Infiltration rates affected by types of soil and the existence of organic layer. Same as science. Forests increase dry-season flows by soaking-up water in rainy season and slowly releasing it back in the system. The more even distribution of water throughou t seasons is very important for run-of-river facilities or those with small reservoir storage. Prices of energy generated during dryseason are usually higher. But not for inter-annual facilities where total annual flows are more important. Dry-season effect uncertain. Depe nds on the difference between losses through higher evapotranspiration and gains through infiltration. Same as people's P&B No information about floods was collected locally during the focus groups because they do not occur in the upper/middle parts of the watersheds. Downstream hydroelectric projects in the area suffer from extreme events every 4-5 years. These events are linked to higher yields of sediments. It is perceived that change from forest cover will increase the frequency of the events (Rojas and Aylward, 2001). Flood benefits can be felt onl y in small events and small watersheds. Similar to science. Forests have lower levels of erosion and sedimentation. Increased siltation of reservoirs has direc t costs in terms of dragging and stopping production to so that. The machinery is also affected by sand and suspended particles. Sediments can reduce the live storage of a reservoir, but could also have a positive effect if they land in the dead storage area (Aylward et al, 1998) Natural forests have little erosion and sedimentation. Properly managed forests could have reduced levels of sedimentation compared to unmanaged pastures. Soil management and natural fore sts have lower levels of erosion and sedimentation. Forests increase water quality. (see above) Most natural forests will provide goo d water quality. Adverse effects are likely to come from bad management techniques rather than forest themselves. Same as P&B (2) Based on Rojas and Aylward (2002). See Porras (2005) for more information on hydroelectric projects. 27


5 Effects over policy making How do peoples beliefs affect policy maki ng? This section concentrates on the ongoing developm ent of Market-Mechanisms for Watershed Protection and how the findings from the Montever de study fit into this. 5.1 Upstream/downstream compensations Market-based mechanisms are currently being heralded as an alternative to manage ment of environmental goods and servic es at watershed level. It is expected that markets will encourage not only environmental protection, increase economic efficiency and save public funds. A considerable number of initiatives around the world has been identified by Landell-Mi lls and Porras (2002), where 61 cases of markets were found in 22 countries (see Figure 9 ), most of them marketing to water quality and regulation. In mo st of these cases, the private sector seems to dominate the supply and demand (in the form of pr ivate landholders for the former and large projects for the latter), while intermediaries have mostly taken the form of government, local municipalities and NGOs. Figure 9 Markets for Watershed Services: summary of global initiatives 1 5 17 18 20 Europe (including Russia) Africa Asia/pacific Latin America & Caribbean North America Source: Landell-Mills and Porras (2002) There is however, little information as to what do watershed markets mean for welfare and poverty alleviation. While econom ic bene fits could take the form of income generation for suppliers, new jobs, cost sa vings in relation to command and control and source pollution control, increased effi ciency in hydroelectric and water supply systems and other positive spin-offs for other water-user activities, there could be significant economic costs to watershed ma nagement in the form of provision of watershed protection, transaction costs asso ciated with the ma rket, and opportunity costs of forgone land uses. Social benefi ts highlighted in th e literature review included health benefits, environmental e ducation, training in improved land uses, improved recreational opport unities, and reduced sound and smell pollution. Other 28


benefits include social institution strengthening, improved scientific knowledge and land title clarification. It is worth noting that the literature review did not present any information as to social costs of watershed markets, and nearly in general, little or nothing was said as to what markets might mean for poor households. In most cases, it is simply assumed that people will benefit, and no especial measures are taken to understand the impact of markets in th eir livelihoods, and how to maximize their potential participation in them. The main constrains identified in market development were especially related to (Landell-Mills and Porras, 2002): High transaction cos ts, in the case of multiple-stakeholder transactions, lack of costeffective intermediaries, poorly defi ne property rights (for land tenure and service rights), and the lack of clear a nd comprenhensive regulatory framework. On the dema nd side: lack of scientific evidence of the rela tion of land use and water, lack of participation of key stak eholders, and lack of willingness to pay. On the supply side: low awareness of ma rket opportunities and the capacity to exploit these, lack of credibility in service delivery, and cultural resistance. One issue that the development of wate rshed ma rkets must overcome is property rights. While land resources are not that pr oblematic and land has 'owners' (private property and large reserves), water owners hip must be determined. In Costa Rica, water rights belong to the Stat e, and the Ministry of Envi ronment has control over it, granting exclusive rights to particular us ers but not allowing us er right transfers, therefore eliminating the possibility of creating water markets. Until now, implicit water prices are obtained through the value of land nearby water works (rivers, springs, lake, and water canals of the irriga tion system), and licences to use water for recreation. Prices for domestic water use, irrigation, and entrance fees are decided by an independend authority: ASES EP (Celis and Segnestam 2001). 5.2 Markets for Watershed S ervices in Costa Rica The effects of land use, and particularly changes on fore st cover, on wa ter quantity and quality have been an on-going debate in Costa Rica, particularly during the past years with the introduction of the Payments for Environmental Services and the recent involvement of private groups as demande rs of better and more reliable water resources. The Law states that owners of forests could claim compensation for the environmental services their forests produce, in the form of biodiversity conservation, carbon sequestration, landscape beauty and water conservation. From the beginning these four services have been bundled together for simplicity sake, and the only difference allowed within types of fore st is for conservation (US$200/ha/over 5 years), sustainable forest management (U S$320/ha/over 5 years), and reforestation projects (US$450/ha/over 5 years). The amount of payment initially establishe d tried to consider different aspects, including the opportunity cost of land17. Additionally, the law was introduced at the 17 This remains one of the flaws of the system, as the oppor tunity cost was selected in terms of pasture for the whole country, not allowing for variation within the country. For example, Miranda, Porras and Moreno (2003) present an analysis of the impact of PES within the central region of Costa Rica and suggest that reforestation projects are not likely to take place in the amount initiall y expected because the opportunity cost of land in the area is much higher than the suggested payment. 29


time when carbon markets were being presen ted as a glamorous opportunity in the international markets, and this is reflected in the payment levels allowed for reforestation (i.e conservation projects receive considerably less than reforestation as the amount of carbon to sequester in th e former is smaller). Since that, new developments have taken place at local, nati onal and international level that question the way that the Law is being applied. Wh ile international cons ensus has not been reached in terms of carbon markets, local in itiatives for watershed conservation have been put forward as a more reachable targ et for marketing environmental services, with a packed produce of improved wate r quality, quantity and improved dry season flows. While the scientific evid ence of the physical links between water and land use (especially forest cover) appear to be te nuous, and in some cases non-existing or even counterproductive, local initiatives are al ready underway and payments are being collected and allocated within different wa tersheds. There is not common consensus as to what is being sold and bought, a nd it could be argued that while current initiatives may have evolved based on a willingness to improved public relations on the part of companies or even on the precau tionary principle of risk reduction if land use changed, it is not likely that long term initiatives will survive unl ess it is clear that a tangible service is really taking place. Watershed services provided by forests have been recognized in Costa Rica for a long time. As early as 1888 a decree was passed declaring a 2-km wide strip of the sides of Barva Volcano as State-owned land, with the objective of protecting the streams and springs that supplied drinking water to the towns of Alajuela and Heredia (Watson et al. 1998). Nevertheless, the first case of an incipient case of market for watershed services took place in 1997, when the National Company of Po wer and Electricity (CNFL) agreed to pay landowne rs located in the Virilla watershed in order to ensure conservation and reforestation of existing forest on their land. While the scientific evidence of the physic al links between wate r and forest cover (conservation or reforestation) appear to be tenuous, and in som e cases non-existing or even counterproductive, local initiativ es are already underway and payments are being collected and allocated within different watersheds ( Table 3 ). There is not common consensus as to what is being so ld and bought, and it could be argued that while current initiatives m ay have evolve d based on a willingness to improved public relations on the part of co mpanies or even on the prec autionary principle of risk reduction if land use changed (Calvo, 2000; Rojas and Aylward 2003, Pagiola 2002, J Kellenberg per.comm. 2001), it is not likely that long term initiatives will survive unless it is clear that a tangible service is really taking place. 30


Table 3. Markets and Payments for Hy drological Services in Costa Rica Service/Mechanism/Case Status Summary 1. Hydrological Services to Hydropower Production ( A) Transfer Payments: FONAFIFO and Hydropower Companies (i) Energia Global: Don Pedro and Rio Volcan Hydroelectric plant Implemented and comi ng to a close, likely to be renewed Company pays $10/ha/yr and FONAFIF O pays the remaining $30/ha/yr. FUNDECO R acts as intermediary. Over $43000 were allocated during the first year. Contracts are for 5 years. (ii) Hidroelectrics Platanar (1) Ongoing implementation Company pays $15/ha/yr an d FONAFIFO the remaining $25/ha/y r. For landholders without land titles the Company pays $30/ha/yr. FUNDECOR and CODEFORSA are intermediaries. Contracts are for 5 years. (iii) Compaia Nacional de F u erza y Luz (3) Aranjuez, Balsa and Cote Ongoing implementation Company covers the full amount of the payment ($40/ha/yr) plus expenses for FONAFIFO ($13/ha during the first year and $7/ha for the remaining years. Contracts are for 10 years. There is no other intermediary between the company and FONAFIFO. (B) Voluntary Contracts (i) Esperanza HEP and Monteverde Con s ervation League Ongoing implementation The agreement settles a dispute over some land where the hy droelectric plant is to be built, granting the right to the company to build and use the water during 99 years, after which infrastructure and land will be the property of MCL. Payments are made gr adually starting with $3/ha during the first year, to $10/ha during the fourth year. After the amount of payment is variable and depends on production and sale price. 2. Hydrological Services to Water Supply (A) Transfer Payments: FONAFIFO and Industry (i) Cos ta Rican Brewery Agreed The comp an y (FLORIDA ICE & FARM) agreed to pay US$45/ha/y r for 1000 ha located in the watershed where their water originates. It al so pays additional money to FONAFIFO and FUNDECOR to administer and monitor the programme. More recently it liased with EHSP (see below) to pay for environmental services in overlapping areas. (i) Melia Playa C onchal Hotel Proposal The company is exploring the option of developing a management p l an for the watershed of the Nimboyores River in order to ensure the protection of the water source in the long term. This water will be key for the development of the hotel s expansion projects. (B) Water Use Charges (i) Heredia Public Water Supply Compan y Charges levied to water consumers, pay ments to forest owners pending Company collects 1.90/m3 in 1999 to help protect the company s catchment areas (Ciruelas, Segundo, Bermudez, and Tibas rivers). Payments to landowners have not begun yet. Source: Adapted from Rojas and Aylward, 2003 Despite the large number of initiatives with in the country, ma rkets are incipient and are constantly changing. The national context is very dynamic and evolves quickly, therefore allowing for improvements and adjustments on the go. Additionally, very little attention is being put onto the social effects of the Payments for Environmental Services. Although it is clear that mark ets for environmental services are not a 31


poverty alleviation tool, the question of how the PES is altering the rural landscape and what are their effects on peoples live lihoods has not been put forward strongly enough. For example, Miranda, Porras and More no (2003) suggest that the use of PES in the central valley of Costa Rica has not necessarily changed significantly the landscape since most payments have been allocated on relativel y wealthy landholders who maintain their forest on their own in terest, and most are not interested in reforestation because it does not pay enough to compete with other existing land uses (i.e coffee, dairy farms or possible urban de velopments). Nevertheless, in other areas of the country the situation might vary, sm all landholders might feel forced to enter into long-term reforestation projects because they lack alternatives for their land and might decide to abort the programme if ma rket situations cha nged. In these cases, it may be wiser to introduce other land use systems that improve watershed management and provide short/medium term livelihoods for small landholders. While it could be argued that for the Cost a Rican case the ma tter of land use change has become largely academic now that defo restation in Costa Ri ca has virtually come to a halt in the last few years (from 16,400 ha/ year in 1986-1997 to 3,300 ha/year in 1997/2000; Snchez-Azofeifa and Calvo, 2002) (J. Fallas, personal communication, November 2001), the question of diminished streamflows following forest removal is as acute as ever elsewhere in Central America (Kaimow itz, 2002) where upland forest protection is much less secure (IUCN-O RMA, 2001). Even further, the continued pressure to undertake revegetation activitie s, particularly reforestation, and the environmental services payments that promote such efforts is strong in Costa Rica, as elsewhere in the world. The need to better understand not just the hydrology but also the economics of reforestation or watershe d management efforts is tremendous as demonstrated by Kaimowitz (2002). However, social issues have often been tertiary in this process due to top-down and centraliz ed approaches to watershed management. 5.3 Markets for Watershed Services in Monteverde This section outlines the local feelings with respe ct to their forests and the externalities they might create. This section reviews the potential room for upstream/downstream negotiations to improve land use. 5.3.1 Our forests provide many benefits The concept of forests and their role in the provision of environmental services is strongly rooted in the farm er s and producers of the Monteverde area. The recognision of benefits also brings the issue of fair ness of compensation, but it is not clear who should compensate for what. It is also rema rkable that, despite the existence of the Payments for Environmental Services programme in Costa Rica for several years, Monteverde is still lagging be hind in joining the programme. o Forests provide many benefitsa nd trade-offs Participants in the Focus Groups recogni se the importance of forests for the maintenance of environm ental services, espe cially biodiversity, espiritual values, and water. 32


As Windbreakers: Trees in windbreaks are important to protect the cattle in areas very exposed to strong winds. Those are the places where reforestation is important. I dont think that its to increase water, because theres alr eady plenty of it. Farmer in Las Nubes Focus Group. We wish we could just cut all the trees down and get as much light as possible for pastures. But we know that if we do that well loose all our water. Farmer in Las Nubes. Trees are life. Yes. But we wish we could cu t them down, and get more milk so we can sell it. Farmer in Las Nubes. o Compensation is fair Farmers believe that if their forests are producing environmental services and som eone else is benefiting, then its is only fa ir that they should be paid for what they already generate in environm ental services, but also to change land uses upstream. Is it fair to limit production in order to plant more forest?. Yes, I think so. If we continue deforesting well end up badly. But there should be some kind of help to reforest. For us is better to have less trees and more pastures. Farmer in Las Nubes. We used to receive payments from the PES, but not any more. We have a small reserve of 10 manzanas, and were getting payments of 10,000colones/hectare. But theyve suspended the payments, arguing that we need to put them in a separate title deed. I think that the money is just sitting with MINAE, and not with the people who are really protecting the nature. The PES could be very important. We are all small producers. If I have 20 ha with half under conservation, that would give funds to complement income for several families. If timber was well paid then wed live of that, but thats not the case. And then, if we are producing $1350/ha in clear oxygen, then why shouldnt we receive that? There should be justice. I dont think that its feasible to have a charge for water downstream. Besides, the damage is reciprocal. If we reforest, we will loose the water upstream, too. Fa rmer in La Cruz. We receive many benefits from the forest. But ICE also benefits a lot from this forest. How much would they benefit if all this area was forest?. Municipality Focus Group. o Who should pay Finance for reforestation should be through incentives, such as PES (El Dos Focus Group). However, several participants pointed out other groups (ICE, and water users downstream) should also pay directly. However, most participants in the focus groups agree that it would be very difficult to get people from downs tream communities to pay for watershed services ( capaz que nos matan!Focus Group in La Cruz) Most of the water from here goes to the Arenal Lake and it is ICE and the Irrigation project who use them. They should pay. And some people from towns, like Cabeceras. But I dont think that we could ask them to pay.. Las Nubes Focus Group. However, local perceptions about ICE are not very positive. We conserve forest just to let the wate r come to ICE. I oppose their policy of expropriationthey can just come and take your land away, without consultation. 33


Its not fair that we have to keep the forest only for ICE. There is a new proposal to let water resources under the control of MINAE, but I do not agree. What is then the purpose of the Water Utilities? To pass on monthly bills to the houses, or should it be to regulate water?. What is the benefit for ICE of forests up in the watershed? Monteverde should receive compensation from ICE for the forests it keeps. How much would that be? If a farmer stops dairy farming to reforest, how much is he losing? ICE should be paying for that. When ICE created the lake, they didnt compensa te everybody. They diverted rivers that other towns were using. Even the municipality of Tilaran lost when the area was inundated and the towns and fincas were lost. The municipality doesnt even receive compensation for the space that the lake takeswhy do we have to pay property taxes then? Municipality Focus Group o Solutions Patches of forest and pasture would be the most beneficial land us e. The forest will guarantee the provision of water, an d the pastue areas will provide the light that we need for the cattle. Farmer in Las Nubes Focus Group. 5.3.2 Why are they not engaging? While the Payment for Environmental Servic es (PES) has been imple mented in Costa Rica for some years already, its involvement in Monteverde has been more limited and the most significant beneficiaries are th e private reserves ra ther than individual private landowners Several reasons were mentioned during th e Focus Groups for the lack of active participation in the PES scheme: o Land Titles: One landowner in La Cruz received paym ents for 10 hectares, but later on had problems related to p roperty titles and th e payments were cancelled. There are many cases in which landowners do not have clea r property titles, and the process to actually obtain them could be a long, tedious, and expensive proce ss that they are not willing to face. o Opportunity cost: While some landowners in the Monteverde area have significantly increased the value of their forests through ecot ourism a ctivities, there is st ill a large amount of people who depend on other land uses for their livi ng. Another problem that was evident in the area is the dependency of cattle in the forest. Several farmers pointed out that while cattle prefers open, light areas of pastures, sometimes they need to take refuge in forested parts (for example, during stor ms or high winds), and to complement their diet. This has been a traditional activity th at would not be allowed under the PES, and would directly affect them negatively. o Perceived negative effects of more forests: 34


There are some landowners, li ving in the upper parts of the watershed, that perceive that increasing the area under fo rest will result in more hum idity and cloudiness in the area, and this will affect negatively their agriculture activities. o Too much regulation: Farmers are reluctant to loose the freedom from Governm ent intervension they have enjoyed for a long time. They fear that th e current levels of regulation are too high, and signing in to the PES would just give a green light for further controls. Not long ago I had some arguments with [M INAE] because they wont let me use my trees. They live out of regulation, but I dont. If I need to use my forest, I have to do it in hiding, and I feel as I was robbing myself . Farmer in Las Nubes. o Lack of information It because evident during the Focus Groups that perhaps the single mo st important reason to explain the lack of participation in the PES pr ogramme is actual lack of information. Many farmers had not even heard of the PES, assuming that it was just another set of [ever changing] incentives to reforest. Others had many strongly rooted ideas of expropriation. Others felt that this was something out of their reach and only for the big private reserves. Much of this reserve to participate could be overcome, slowly, with information oriented to the farmers. A random survey of over 100 landowners in the area showed that not one individual was receiving PES, and less than half of them were even aware of the prog ramme (Hope 2004, see below for more detail on the survey). PES perception in Monteverde. The impact of the PES on livelihoods is evaluated across a range of qualitative responses to respondents knowledge and perceptions of the policy (see table below). Less than half of each of the three livelihood groups we re aware of the PES policy. Coffee farmers reported the highest leve l of awareness (46%), followed by livestock farmers (34%) and then tourism (22%). Only one percent of livestock farmers had applied for the payment and this farmer (n=1) had been unsuccessful in receiving the payment. Openended questions in the survey instrument generated a range of responses to why respondents had not applied for the PES. Four categories emerged from the responses: information, low returns (US$ per ha), land title and commitment. Lack of information was the dominant reason (61%) why livestock farmers had not applied for the payments. One in three coffee farmers cited this reason, whilst a similar pr oportion (32%) identified the low returns of the payment level. The opportunity cost of payments compared to other productive land uses is identified as a constraint to wider adoption of PES policy in the Varilla watershed in Costa Rica (Miranda et al., 2002). Livestock farmers also identified low returns but given that the majority of the sample knew little to nothing of the PES policy this proportionately lower percentage of responses to a second-order constraint is consistent with their lack of information. Coffee farmers increased knowledge of the qualifi cation criteria for PES stated that lack of a title deed to the property was a another limitation (28% of responses) to uptake of the policy. Seven percent of livestock farmer s recorded title deed ownership as a constraint, also. Finally, there was a lower proportion of respondents who described a reluctance to enter into land contracts with the government. Though this represents a minority of responses here (9% of coffee and 13% of livestock), this theme surfaced regul arly in discussions with farmers based on widespread distrust of government land management. 35


6 Summary and recomendations The main findings and recommendations of this study are: Science and popular perceptions differ, and this can affect both policy-ma king, and the possible success and uptak e of new land use policies. Land use changes fast. Urbanisation, not deforestat ion, is the ma in threat. A people-centered a pproach is needed. Governme nt perception is poor, mostly because of lack of communication. Uptake of payments for environm ental se rvices by small private landholders has still some way to go. 6.1 Science and popular perceptions differ Different groups perceptions into knowledge, attitudes and beliefs on land/water relationships vary substantia lly in humi d tropics. Perhaps the strongest debate focuses on the effects of forest (or deforestation) on water qualit y, quantity, and the effect on low flows. There has been a historical tre nd of predictions (many times alarming), of the relations between change of forest cover, and soil and water degradation. According to Saberwal (1997), the majority of these predictions are characterized by: an absence of em pirical data to support particular scenarios of degradation, an absence of long-term data to enable the detection of directional points, a failure to separate naturally occurr ing events from those induced by human activities; and a failure to distinguish seasonal from pe rm anent changes in vegetation cover. Saberwal (1997) suggests that the desiccating infl uence of deforestation was amply discussed in Europe since the 17th century, as an input used by European foresters to press for the establishment of forest reserves in France and Germany, and the institutionalisation of forest conservation. European forestry held a notion that forests played a crucial role in influencing climate through increa sed precipitation and moderation of temperature extremes. The European influence can be traced in America and India until the beginning of the 20th century, period in which most of their forestry experts were educated in Europe. The connection between forests and rainfall was useful for foresters to gain greater control over forestlands, although empirical meteorological da ta proved otherwise. The larger debate of forests and water is pr incipally fuelled by the collision of viewpoints and economic needs of land settlers and colonise rs, engineers, scientists, conservationists and environmentalist, united with existing power institutions. While scientific ideas may have provided material for the formulation of the discourse, the institutional context in whic h it has taken place has provided its shape and direction (Saberwal, 1997). Calder (2002), suggests th at the disparity between scientific research and policy agenda, especially when linked to dry-season flows, has arisen through the extensive promotion of certain land uses and en gineering interventions by vested interest groups in th e absence of any effective di ssemination of the scientific evidence which may allow a contrary view . The economic imp lications of such approach are found in the form of wastag e of development funds on unachievable 36


targets and the unwarranted blame of upland communities whose practices have generally had only marginal impacts on downstream flooding. A new study by World Bank-WWF Alliance for Forest Conservation and Sustainab le Use18 shows that protecting forest areas pr ovides a cost-effectiv e means of supplying many of the worlds biggest cities with high quality drinking water, providing significant health and economi c benefits to urban populations. Well-managed natural forests can minimize the risk of small la ndslides, erosion, and sedimentation. They contribute to filtering pollutants, such as pe sticides. According to the report, adopting a forest protection strategy can result in massive savings. The report argues that, for example, it is much cheaper to protect forests than to build water treatment plants. Common perception in Central America is that fo rests increase rainfall, captures water from precipitation and slowly releases into the ground, protect water springs during summer, protect against intense flooding events, and re duce sedimentation (Kaimowitz, 2001). This perception is f ound nearly all over the world. A quick email about local perceptions conducted in Vietnam (IIED, no date) indicates that people belief almost in general, that forest increase flows in rivers and streams, increase dry-season flows, increase rainfall, control fl oods, and reduce sedimentation (Elaine Morrison, personal communication 2003). Similar quotes abound around the world. The effects of the Mitch Hurricane in Central America in 1998 were largely blamed on deforestation by common consent from politicians, media, environmental groups, and international agencies. A quick se arch in the Internet shows at least 55 references to the aggravating effects of deforestation: "Mitch Hurricane calls attention on environmental degradation": "Damages in watersheds due to deforesta tion: the Mitch case": PRISMA, Guatemala. docs/prisma/03391html "Deforestation, poverty and global warming aggravated the Mitch effects": ms/elan/may99/msg01118html The devastating effects of deforestation ar e often cited around th e world, most of the time without strong scientific data to back up these statem ents (Saberwal, 1998). In 1986, referring to deforestat ion in the Hima layas, the environmentalist Norman Myers, (winner of the 2001 Blue Planet Prize), indicates that: "Primarily because of deforesation in thei r headwater regions, the river systems are increasingly subject to disruption, leading to floods followed by droughtsThe Himalayan forests normally exert a sponge effect soaking up abundant rainfall and storing it before releasing it in regular amounts over an extended period. When the forest is cleared, rivers turn muddy, and swollen during the wet season, before shrinking during drier periodsFlood disasters are becoming more frequent and more severe" ( cited in Saberwal, 1998) The link between land use and water have b een a concern for scientists during many years and in more recen t times decision makers are becoming more aware of the economic implications of bad watershed ma nagement. Changes in land use could not only affect water resources (l iability of the ri ver to flood, the magnitude of such flood, 18 For more information see news/artir eport07ing.php ). 37


sediment loads and the dry season flows), but soil degradation could also be a result of overgrazing, poor irrigation and land ma nagement, and over-exploitation of vegetative cover ( Calder, 1999 ; Russell, 1981 ). A clear understanding of these links is key to design appr opriate policy me asures to improve watershed management. Myth-based policies could result not only in a waste of economic resources, and time, but could also have potential negative consequences that would be felt in the medium and long term. While for a long time it has been the task of the goverment to ensure the provisi on of watershed services, an emergence of new alternatives that could prove more effi cient and cost-effectiv e, in the form of market-based mechanisms for environmenta l services, in which the consumer pays principle becomes the main driver, but it is also expected, and demanded, that a true service be delivered. 6.2 Land use changes fast Land use has important economic values in the upper and mi ddle parts of the watershed. In the upper parts, the intr oduction of new technologies, such as windbreaks, has increased the productivity of dairy farming. Coffee industry is managing to access niche pro-environment markets and obtaining premiums. Ecotourism has given a whole new value to na tural forests, and the local population is rapidly capitalising on these valu es through activities such as hotels and guided tours. Rapid urbanisation -not defore statio nis the problem. The general perception in the study area is that water is under threat fr om population expansion. Deforestation has had major impacts, but it is not the only factor a ffecting water resources: Water flows decrease because of higher competition: According to most participants, the lo wer parts of the watershed have less water because of deforestation: (" Up in the mountain forest we have waterfalls and water is plentiful. However, as you go down the watershed everything has been deforested and is dry "). Water is abundant in winter, but lower parts of the watershed, like Guacimal, suffer from serious water shortages during the dry season. However, while there seem s to be a widespread belief that deforestation in the watershed has contri bute to lower water levels, there are some who indicate that water levels downstream are decreasing because of higher competition for water for pipe lines. Some groups, like the Aqueduct, worry that the increasing demand of wa ter for new pipes will reduce water available downstream, especially duri ng the summer. Improving watershed management and smooth provision of water flows throughout the year will imply more than just engaging in wa tershed conservation, but looking into water distribution among different users. Water quality deteriorates for lack of control: Reduced water quality in rivers and wa terways is considered as one of the most serious problems that the area f aces. Participants in the focus groups indicated that tourism must be cont rolled to avoid negative impacts on the environment, especially in terms on unr estricted growth of tourism-industry 38


and the lack of controls on basic serv ices such as wastewater disposal and traffic control. 6.3 Government perception is rather poor Thank God the government has stayed away from here (CETAM focus group) We havent had many government incentives here. We havent had municipality (until recently), or hospital. There arent any politicos with land in this areait is only the pioneer families and their descendants, with our new vision for conservation (CETAM focus group). Monteverde and its surrounding areas have evolved almost free from governmental intervention. In fact, it is only a year ago that the region has its own local municipality located in Santa Elena. Most people are hi ghly suspicious of the government and its institutions, and recent the high level of re strictions imposed on the use of forest within their own properties (" I need to use my own forest, but have to do it sneaking around and feeling as if I'm robbing myself" farmer in Las Nubes). Government institutions have limited control or actual authority in the area. Despite the existence of laws, there is little m onitoring and most people do as they want. Control over land use has come primarily fr om the dairy and the coffee Cooperatives, who try to put environmental re gulations to their providers. There are many problems with the government and the local authorities. They have many laws but nobody follows them, and instead of helping they just interfere with everything. Laws should be more localisedwhat is good for San Jos is not necessa rily good for this parts. Focus Group in El Dos. This situation means that, if negotiation for watershed ma nagement is to take place, government intervention will be seen with suspicion. 6.4 A people-centered approach Protection, People, and Progress Sustainable developm ent, as explained by a farmer during the focus group in La Cruz, has to de al with protection, people, and progress. If forests provide important environmental se rvices to local, regional and international users, it is only fair that land stewards receive compensation from either engaging in forest protection or abstaining themselves from forgone economic benefits of other land uses. It is important to create ne w sources of employment for young people to prevent migration to other areas. Historic isolation. Another key eleme nt for the success or failure of any development proposal is the historic isolat ion of the area, which has c ontributed to the independent spirit of the inhabitants. The upper part s of the watershed, in particular those surrounding the cloud forests, have had a history of isolation from the rest of the country. This has contributed to forge the independent character of its inhabitants. Until very recently the area had very few economic activities, and land tenure is largely in the hands of descendants from the first settlers, the Quakers, or the Reserves. 39


Roads: a double-edge sword. Roads are always a conversat ion topic in Monteverde. Roads are unpaved. During the rainy season they are thick with m ud and in the dry season they are covered with fine dust th at permeates into clothes and lungs. A 4x4 vehicle is strongly recommended during both seasons as the higher clearance and the traction can be essential. There is controversy surrounding the decisi ons for paving the road. A mi nority of people, mostly living in what is actually Monteverde, so me of them Quakers, oppose the paving of the roads for a number of valid reasons. Part of the reason being the influx of people it could bring and all of th e associated problems that come with large numbers of people, pollution, social problem s, etc Some people would like to see Monteverde stay the same. It is hard to stand in the way of Progress and a bill was signed to pave parts of the road. That was 2 years ago and so far there is no pavement and people are not holding thei r breath. Things can be a sl ow process with the Costa Rican government. Continuous learning from research. It is important for landowners to learn from experience about the types of sp ecies used for reforestation (" in some cases there are trees close by to water sources, not because they help keep the water but because they need it, we need to learn about t hat" ( coffee producer in El Dos)) Communities have lost credibility in previous reforestati on programmes that used massive amounts of eucalyptus and conifers, as th ey were later on discovered to be harmful for water. One coffee producer complains of the large amount s of research done, but the little of it that goes back to the communities where it was originated. 6.5 Setting up negotiations for ma rkets has a long way to go PES is perceived as a threat. The way that th e current PES programme in Costa Rica is seen by some participants of the focus groups as another way to expropriate lands for them. It doesn't work out for me. I have 60 ha of forest. I could be r eceiving $3000 per year, but it will be as if I'm selling it off to the government, b ecause I cannot even touch my own forest after that, even if I need it" (dairy farmer in Las Nubes). Another restriction is the lack of land titles, and the high costs of obtaining one to take part in the programme. It is quite likely that landowners will refuse to be part of the gover nment PES, and any market will probably have to be highly local in order to gain support. The importance of watershed protection a nd conservation has been on the political agenda of governme nts and international agencies for a while. In more recent years, the emergence of the private sector in response to lack of effectiveness in governmental measures has cropped up in the form of markets for watershed services. Landell-Mills and Porras (2002) presente d a review of 61 cases of market mechanisms in watershed protection around the world, and the number of initiatives and enthusiasm by supporters is steadily increasing. The basi c premise is that improved land use in the upper part of the watershed, normally assumed as forest protection or reforestation, has positive e ffects on water services, and downstream water users will compensate upland land stewards for the protection of such services. It is worth to note that while the development of similar markets for carbon sequestration have strong scientific cons iderations, the expansion of markets for 40


watershed services is characterised by a nearly total absence of strong, consistent, defendable scientific evidence th at support the claims made. Does it really matter what people belief? If the ov eral l objective of integrated watershed management is to improve environmental conditions upstream, and improve livelihoods through payments for wate rshed services, surely it does help if people belief that trees and forest improve water flows. Upstream forests will be improved, upstream people will receive payments, and downstream people will willingly pay for forest conservation. In many watershed conservation initiatives, putting science in place is costly, and many times impossible due to lack of information. So why does it matter? It matters because funds are limited, and decisions are costly. Tw o extreme situations in which policy does not take into acc ount science are presented below: Situation: "Reforest large areas of watershed to reduce sedimentation". If the project does not cons ider possible effects on hi gher evapotranspiration the overall effect of the project would be si gnificant reductions in water flows. For example, the initial Law 21 in Panama pr oposed large reforestation of the Panam a Canal Watershed, but initial studies revealed that water flows c ould be significantly reduced in dry seasons causing potentially im portant losses for the passage of ships through the canal (Aylwa rd et al 2001). Situation: "Hydroelectric companies paying upstream reforestation projects to improve their local P R image". There are cases in which hydroelectric companies might be willing to engage in payments for watershed services, not so much because they are convinced of the hydrological service they r eceive but to improve local acceptance and image. However, situations like this can be frag ile and prone to be dropped by the company at any time. Only when there is a real threat to the company's input, long-term commitments are likely to succeed. In the same way, it is important to unders tand what people living in uplands think of the eff ects on water arising from their la nd management decisions. The knowledge of which land use measures must be taken to provide a service for which they could receive a payment is one key aspect to ensure the success of the market. Misinformation, or erroneous beliefs, might bring land stewards to embark in costly land use activities that will derive no se rvice in the medium or long term, and therefore no compensation. Likewise, sceptical landowners might not be interested in engaging in new watershed management programmes despite the existence of payments, if they have suspicions about the seriousness of the programme. While the protection of forests is worthwh ile in environmental, cultural, social, econom ic and hydrological terms, it is important to understand the rightful linkages between forests and water in order to desi gn the appropriate measures of watershed management. Taking the wrong land use meas ures, like assuming that reforestation with conifers or fast growing species will increase water flows, could potentially have serious economic effects downstream if flow s are actually reduced. Understanding 41


what the linkages are will help design th e appropriate measures to ensure a more effective integrated watershed management. 42


7 Bibliography Anderson, J. M, Spencer T. 1991. "Natural Forest Hydrology," Carbon, Nutrient and Water Balances of Tropical Rain Forest Ecosystems Subject to Disturbance: Management Implications and Resear ch Proposals. France: UNESCO, 25-33. Andreini, M.; Van de Giesen, N.; Van Edi g, A.; Fosu, M.; and Andah, W. Volta Basin Water Balance. ZEF-Discussion Papers on Development Policy, No. 21. Aylward, B. and Echeverra, J. 2000. Syne rgies between Livestock Production and Hydrological Function in Arenal, Costa Rica. Accepted for publication in the Environment and Develo pment Economics, Cambridge University Press June 2000. Aylward, B. forthcoming. Land-use, Hydrol ogical Function and Economic Valuation. In Forests Water Peop le in the Humid Tropics, edited by M. Bonnell and L. A. Bruijnzeel. Cambridge: Cambridge University Press. Aylward, B., and Tognetti, S. 2001. Valuation of Hydrological Externalities of Land Use Change: Lake Arenal Case Study, Costa Rica. Land-Water Li nkages in Rural Watersheds Case Study Series. FAO, Rome 2001. Aylward, B., Echeverria, J., Fernandez Gonzal es, A., Porras, I., Allen, K., and Mejias, R. (1998) Economic Incentives for Watershed P rotection: A Case Study of Lake Arenal, Costa Rica. Collaborative Research in the Economics of Environment and Development (CREED), IIED, London. Bands, D. P., Bosch, J. M., Lamb, A. J., Richardson, D. M., Van Wilgen, B. W Van Wyk D. B., and Versfeld, D. B. 1987. Jonkershoek Forestry Resarch Centre Pamphlet 384. Pretoria, South Africa, De partment of Environmental Affairs. Bolaos, R. 1995. Mapa de Uso Actual de las Subcuencas Ro Chiquito, Cao Negro y Aguas Gatas. San Jos: Centro Cientfico Tropical. Bonell, M. 2001 (unpublished) Possible Impacts of Land Use Conversion on the Spatial and Temporal Orga nization of Precipitation. Bonell, M.; Hufschm idt, M.M. and Glad well, J.S (Editors) 1993. Hydrology and Water Manageme nt in the Humid Tropics. Cambridge University Press. Booth, et al, 2000. On narrative MIRIAM CHECK Bosch, J.M. and Hewlett, J.D. 1982. A Review of Catchment Experiments to Determine the Ef fect of Vegeta tion Changes on Water yield and Evapotranspiration. Hydrology 55:3-23. Brocklesby, E. D. and Holland, J. (1998) Participatory Poverty Assessments and Public Services: Key Messages from the Poor. London: DFID, Social Developm ent Division. Bruijnzeel, L. A., Proctor, J. 1995. "H ydrology and Biogeochemi stry of Tropical Montane Cloud Forests: What Do We Really Know?," Tropical Montane Cloud Forests, 38-78. 43


Bruinjzeel, S. 1992. Managing Tropical Forest Watershe ds for Production: Where Contradictory Theory and Practice Co-Exist. In Wise Management of Tropical Forests, Oxford Conference on Tropical Forests 30 March-1 April 1992. Bulte, E., van Soest, D. P., Cornelius van Kooten, G, and Schipper, R. A. (2002) Forest Conservation in Costa Rica When Non-Use Benefits Are Uncertain But Rising. Am erican Journal of Agricultural Economics 84(1): 150-160. Calder, I. And Dye, P. 2000. Hydrological Impacts of Invasive Alien Plants. In Management Practices: Alien Invasive Species Calder, I.R. 1999. The Blue Revolution: La nd Use and Integrated Wa ter Resources Management. Earthscan Publications. Calder, I.R. 2002. Integrated Watershed and Flood Mitigation Management. FAO report. Calder, I.R. 2002. Forests and Hydrological Services: Reconciling public and science perceptions. Land Use and Wa ter Resources Research 2 (2002) 2.1.12. Newcastle, UK. Calvo, J. (2000). Case study: payments for watershed services in Costa Rica. In: M.F. Price & N. Butt (Eds.), Forests in Su stainable Mountain Development. A State-of-Knowledge Report for 2000 pp. 428-429, CABI, Wallingford, U.K. Calvo, J. 1996. Efecto del Uso de la Tierra y Estudios de B alances Hdricos para Cuencas Tropicales: con particular re ferencia al Embalse Arenal. Report prepared for CREED Costa Rica. San Jos, CCT/CINPE/IIED. Calvo, J. and Quirs, O. 1996. Modelos de Prediccin de Carga Media Anual de Sedime ntos en Suspensin en Cuencas Rurales de Costa Rica. CREED Costa Rica, notas Tcnicas #4. San Jos. CCT/CINPE/IIED CCT. 1980. Estudio Ecolgico Integral de las Zonas de Afectacin del Proyecto Arenal. Report prep ared for ICE. Sa n Jos, Centro Cientfico Tropical. Celis and Segnestam, 2001. Institutional Framework for a Resource Monitoring System in The Arenal-Tem pisque Watershed. Working paper. CT Energia S.A. (2000). Analisis del Impact o de la Conservacin de Bosques en la Generacin HidroDFID (2002) Poverty Reduction and Sustai nable Development the Challenge for Johannesburg. A speech by the Rt Hon Clare Sh ort MP, Secretary of State for International Development. Development Policy Forum, QEII Conference Centre, London, 20th June, 2002. At: Echeverria, J., Hanrahan, M. and Solor zano, R. (1995) Valuation of non-priced amenities provided by th e biological resources within Monteverde Cloud Forest Preserve, Costa Rica. Ecological Economics 13: 43-52. Fallas, J. 1996. Cuantificacin de la In tercepcin en un Bosque Nuboso, Monte de los Olivos, Cuenca del Ro Chiquito, Guanacaste, Costa Rica. CREED Costa Rica, Notas Tcnicas #6. FAO (2000) Inter-agency Experiences and Lessons. From the Forum on Operationalizing Sustainable Livelihoods Approaches. Pontigano (Siena) 7-11 March 2000. FAO, Rome, 2000. 44


Fernndez-Gonzlez, A., and Aylward, B. 1998. Incentives and Institutional Arrangements for Participatory Watershed Management: The Case of Arenal, Costa Rica. The World Bank/WBI's CBNRM Initiative. http://srdis.ciesin.columb Grandin, B. E. (1988) Wealth Ranking in Smallholder Comm unties: A Field Manual. Intermediate Technology Publications Ltd, London. Hamilton, L. S., Pearce, A. J. 1988. "Biophysical Aspects in Watershed Management," Easter, K W., Dixon, J. A., Hulschmidt, M. M., Watershed Resources Management. An integrated Fr amework with Studies from Asia and the Pacific. Westviewer Pr ess / Boulder and London, 33-53. Hazell, P., Chakravorty, U., Dixon, J., and Celis, R. 2001. Monitoring Systems for Managing Na tural Resources: Economic Indicators and Environmental Externalities in a Costa Rican Watershed. EPTD Discussion Paper No. 73, International Food Policy Research Institute and The World Bank. Hewlett, J.D. 1982. Forests and Floods in the light of Recent Investigation. Proc. Canadian Hydrological Symposium 14-15 June 1982, Fredericton, 543-60. Hicks,B.J., Beschta,R.L., and Harr,R.D. 1991. L ong-Term Changes in Streamflow Following Logging in Western Oregon and Associated Fisheries Implications. Water Resources Bulleting 27 (2): 217-226. Hough, J. 1986. Management Alternatives for Increasing Dry Season Base Flow in the Miombo W oodlands of Southern Africa. Ambio 15 (6): 341-46. Hsia, Y.S., and Koh, C.C. 1983. Water Yield Resulting from Clearcutting a Small Hardwood Basin in Central Taiwan. In Hydrology of Humi d Tropical Regions, with Particul ar Reference to the Hydrolog ical Effects of Agriculture and Forestry Practice. Proceedings of the Hamburg Sympsium August 1983. pp 215-220. IAHS Publ. No.140. Hope, R. 2005. Livelihoods report. Hydroconsult, A.B. 1993. El Embalse Aren al: Condiciones Generales, Problemas y Propuesta para un Program a de Estudio. Report prepared for ICE. Uppsala: Hydroconsult. Instituto Nacional de Estadsticas y Censos (INEC) census data, 1984 and 2000. IUCN-ORM A (2001). Estado de la Conservaci n de los Bosques Nubosos en MesoAmerica. IU CNIWMI (2000). Improving Water and Land Resources Management for Food, Livelihoods and Nature. Colombo, Sri Lanka, IWMI : 72. Jansson, M. 1996. Investigations in the Aren al Dr ainage Basin: Subject Report No. 1 of the Arenal Reservoir Project. Re port prepared for SIDA, Uppasala: AB Hydroconsult. Jermar, M. K. Water Resources and Water Management. In Developments in Wa ter Science 28 Elsevier, 1987 45


Jorgensen, J.R. and Wells, C.G. 1986. Tree Nu trition and Fast-Growing Plantations in Developing Countries. International Tree Crops Journal 3(1986):225-44. Kaimowitz, D. (2002). Useful myths and intr actable truths: the politics of the link between forests and water in Centra l America. I n: M. Bonell & L.A. Bruijnzeel (Eds .), Forests Water People in the Humid Tropics. Cambridge University Press, Cambridge (in press). Kaimowitz, D. 2001. Cuatro medio verdades sobre la relacin entre bosques y agua en Centroamrica.RUTA en Centroam erica. Boletn Trimestral No 17, Abril. Kauck, D. and Tosi, J. Jr. 1989. Public Poli cy Decisions and Private Sector Resource Use:Changing Patterns of Deforestation and Land Use in the Lake Arenal Basin, Costa Rica. SanJos, Costa Rica:Tropical Science Center. Kim, E., Rhodes, A., Katchpole, S., Wells, A., Scheffe, B., Dallas, S, and Pufall, A. W ater Quality Study Of A Cloudforest Watershed In Monteverde, Costa Rica. Smith College, Northampton, USA and the Monteverde Institute, Costa Rica. Koens. 2003. Monteverde. Krammer,R.A.; Richter,D.D.; Pattanayak,S .; and Sharma ,N.P. 1997. Ecological and Economic Analysis of Watershed Pr otection in Eastern Madagascar. Journal of Environmental Management 49 (3): 227-295 Lal, R. 1983. Soil Erosion in the Humid Tr opics with Particular Reference to Agricultural Land Development and Soil Managem ent. In Hydrology of Humid Tropical Regions, with Particul ar Reference to the Hydrological Effects of Agriculture and Forestry Practice. Proceedings of the Hamburg Sympsium August 1983. pp 221-239. IAHS Publ. No.140. Lal, R. and Russell, E.W. (Editors) 1981. Tropical Agricultural Hydrology: Watershed Manageme nt and Land Use. Landell-Mills, N. and Porras, I.T. 2002. S ilver Bullet or Fool s Gold?: A global review of markets for forest env ironm ental services and their impacts on the poor. London. Institute for Environment and Development. Lawton, R.O., Nair, U.S., Pielke Sr., R.A. & Welch, R.M. (2001). Climatic impact of tropical lowland deforestation on nearby montane cloud forests. Science 294: 584-587. Mayers, J. and Bass, S. 2004. Miranda, M., Porras, I, and Moreno, M.L 2002. A quantitative analysis of the social effect of paym ents for environmental se rvices in the Virilla Watershed, Costa Rica. IIED/CINPE. Update this reference Mourraille, C., Porras, I.T. and Aylw ard, B. 1995. La Proteccin de Cuencas Hidrogrficas: Una Bibliografa Anotada de Hidrologa, Valorizacin Econm ica e Incentivos Econmicos. IIED, Centro Cientfico Tropical, and CINTERPEDS. San Jos, Costa Rica. Nadkarni, ?. 2000. On Monteverde Nicol, A. (2000). Adopting a Sustainable Livelihoods Approach to Wa ter Projects: Implications for Policy and Practi ce. London, UK, Overseas Development Institute: 31. 46


Ortiz, E. 2002. Pagiola, S. 2002. Paying for Water Services in Central America: Learning from Costa Rica. In Selling Forest Environmental Se rvices: Market-based Mechanisms for Conservation edited by S. Pagiola, J. Bishop and N. Landell-Mills. Pateman, 2002. Monteverde. Porras, I. 2005. Valuing cloud forests contri bution to water flows in hydroelectricity. An application in Arenal and Peas Blancas, Costa Rica. CLUWRR, IIED. Porras, I., Miranda, M. and Hope, R. 2005. Exploratory Paper??? ARE WE GOING TO PUBLISH THIS ONE TOO? Pounds, J.A., Fogden, M.P.A. & Campbell, J.H. (1999). Biological response to climate change on a tropical m ountain. Nature 398: 611-615. Rietbergen, S. 1993. The Easthscan Reader in Tropical Forestry. Earthscan Publications Ltd, London. Rojas, M. and Aylward, B. 2002. Rojas, M., and Aylward, B. 2002. What ar e we Learning from Experiences with Markets fo r Environmental Services in Costa Rica? A Review and Critique of the Literature. A report for the International Institute for Environment and Development. Russell, E. W. 1981. "Role of Watershed Ma nageme nt for Arable Land Use in the Tropics," Lal, R., Russell, E. W., Tropical Agricultural Hydrology. Great Britain: Pitman Press, 11-16. Saberwal, V. K. 1998. Science and the Desicca tionist Discourse of the 20th Century. Environment and History 3, 309-343. Cambridge, UK, The White Horse Press. Saboro, J. and Aylward, B. 1997. Anlisis Espacial de Erosin y el Transporte de Sedime ntos en Tres Micro-Cuencas de Arenal, Costa Rica. CREED Costa Rica. Notas Tcnicas #7. San Jos: CCT/CINPE/IIED. Saleti, et al. 1979. Snchez-Azofeifa, A & J. Calvo. Estudi o de Cobertura Forestal de Costa Rica empleando im agines Landsat 2002. Albe rta University, Edmonton y Centro Cientfico Tropical, San Jos, Costa Rica. 30 pp. Scoones, I. (1998) Sustainable Rural Live lihoods: A Framework for Analysis, IDS Working Paper 72. Sen, A. (1990). Cooperative Conflicts. In, Persistent Inequalities (Ed) Tinker, I. Oxford, Oxford University Press. Shaw, E.M. 1988. Hydrology in Practice. Third Edition. Chapman & Hall, London. Singh, K.D. et al (1990). A M odel Approach to St udies of Deforestation'. DEFR 3, FAO, Rome. Turton, C. (2000). Enhancing Livelihoods Through Participatory Watershed Development in Ind ia. London, UK, Overseas Development Institute : 28. 47


Vsquez A.; and Rodrguez, C. 1995. Anlisis de Erosin en la Cuenca del Ro Chiquito. Report prepared for CREED-CR. San Jos, Centro Cientfico Tropical. Viessman, W. Jr, Knapp, J. W., Lewis, G. L., Harbaugh, T. E. 1977. Introduction to Hydrology. Harper & Row. Watson, V., Cervantes, S ., Castro, C., Mora, L, Solis, M., Porras, I., and Cornejo, B. 1998. Making Space for Better Forestry. Policy that Work for Forest and People series No. 6, Costa Rica. Centro Cientfico Tropical, San Jos and IIED, London. Wilson, M. F and Henderson-Sellers, A. 1983. Deforestation Impact Assessm ent: the Problems Involved. Hydrology of Humid Tropical Regions, with Particular Reference to the Hydrological Effects of Agriculture and Forestry Practice. IAHS 140. Proceedings of the Hamburg Sympsium. Zadroga, F. (1981). The hydrological importan ce of a montane cloud forest area of Costa Rica. In: R. Lal & E. W. Russell (Eds.), Tropical Agricultural Hydrology J. Wiley, New York, pp. 59-73. 48


8 Appendix A review of linkages of land use and hydrology The hydrological cycle is basically driven by the energy received from the Sun. The Earth reflects part of the energy (34% on av erage), distributing the rest between air, water, soil and geological formations (Jermar 1987). The albedo (coefficient of reflection) depends essentially on the surf ace type, state and quality of atmosphere above and the angle of the sun rays. For exam ple, on average stretches of water reflect 10% of the energy, lawns 15%, forests 20% deserts 30% and snow 80% (Jermar 1987). Fluctuations in the soil and water te mperature and evaporat ion are the result of the acceptance of effective radiation. This hydrological cycle is an uninterrupted process of water motions. Most of the Earths water is stored in the ocean s (94.2%), 4.13% is stored as groundwater, and 1.25% in ice sheets and gl aciers. Only a very small fraction (0.019%) corresponds to surface water on la nd and soil moisture (0.0055%), while rivers carry only 0.00008% of the total wate r of the planet. Fi nally, a tiny proportion of water is kept as atmospheric vapo ur (0.00096%) (M.I. Lvovich, 1979, quoted by Shawn 1983). This total amount of water is indestructible (Viessman et al 1977) and the hydrological budget can be considered a closed system. The hydrological cycle could be summ arized as a continual cycle of: 1. Precipitation 2. Interception (evaporation and evapotranspiration) 3. Surface runoff 4. Percolation into the subsoil (groundwater) Heating of the water su rfaces (the ocean being the m ost important one) causes evaporation defined as the transfer of water from liquid to gaseous state. According to Shawn, this vapour remains stored in the atmosphere for an average of 10 days, when through a process known as condensation the water vapour changes back to the liquid state and forms clouds. With favorable conditions, precipitation (in form of rainfall or snow) is produced either retu rning to the ocean surface or to the land surface. In the second case, before precipit ation reaches land surface, it can be intercepted by vegetation from which water might return b ack to the atmosphere as evaporation. Rainfall reaching the ground may go down as run-off and join creeks and rivers, or it might infiltrate into the ground. Water in the soil percolates through the unsaturated layers to reach the water tab le, where the ground becomes saturated, or it mi ght be taken back up by the vegetation to be transpired back into the atmosphere. The surface run-off and the groundwater flow together and join in surface streams and rivers, and might be temporarily held in lakes, but finally flow into the ocean. 49


It has been long discussed the effect th at land use would have on the hydrology of a region. In tropical areas, th is discussion falls immediat ely on the grounds of tropical forests and alternative land uses. These fo rests provide abundant environmental and socio-economic benefits and can be used in different ways ( Bruinjzeel (1992) ): 1. Maintain fo rest with little or no disturbance, for nature reserves, steep headwater areas of strategic catchmen ts, or geologically unstable areas. 2. Manageme nt of natural forests for co ntinuous production of timber and NT products. 3. Clearing of forests and subsequent use of land for grazing, farm ing, tree plantations, mining, settlements, etc. In most developing countries, land use changes are a daily occurrence. While it m ight be desired to maintain strategic natural fo rests, few countries can afford to do so. Existing natural forests are threatened with requirements for agricultural lands, firewood, timber and pulp demands. At the same time, water requirements downstream are increasing: hydr oelectric and irrigation proj ects, transport, industry and growing cities all demand higher amounts of water. All actions taken at different stages of the watershed could have important effects on quantity and quality of water, with serious externalities over other users downstream. Rainfall and Precipitation The distribution of forest is a consequence of climate and soil conditions not the reverse. Bands et al, 1987 Water is present in the atmosphere in th e form of water drop lets, vapour and ice crystals or snow. Pr ecip itation occurs as a result of a balanced process. When the air is pure and it becomes greatly saturated, then water vapour become water droplets. The presence of small particles, or aero sols, provide the nuclei around which water droplets are formed. Moist air must be cooled to near its de w point, a process that could happen through: the rising of air by an impeding mounta in range that causes a reduction in pressure and lowering of tem perat ure without transference of heat, the meeting of two very different air masses, the contact between a mo ist air mass and a cold object, such as the ground. Once cloud droplets are formed, their growth depends on other tension forces such as the humidity of the air, rates of transfer of vapour to water droplets and the latent heat of condensation released. Condensed wate r appears in various forms of clouds. There are major categories of precipitation types: convective, orographic and cyclonic (Viessman et al 1977). Convective precipitation is typical of the tropics, brought about by the heating of the air at the inte rface with the ground. They might take the form of light showers or high intensity stor ms. The tropical zones located in the trade belts in latitudes 5-25o north and south of the Equator present irregular wind patterns and the development of tropical maritime air masses. The cloud-forming activities here are forceful and subsequent rainfall can be considerable, with up to 300 mm falling in 24 h. Orographic precipitation results from the mechanical lifting of moist 50


horizontal air currents above natural barriers, such as mountain ranges. Finally, cyclonic precipitation is associated with the moveme nt of air masses from high to low pressure regions, created by the unequa l heating of the earths surface. Nevertheless, although the main driver fo r precipitation is lin ked to geographical situations, it has been lo ng debated what the effect of land use change (i.e. def orestation) would be. Theory states that the height of trees will increase the orographic effect which will, in turn, lead to an increase in rainfall. The effect, however, is likely to be only slightly ( Calder 1999 ), as the possible increase in rainfall will be most of the time captured by the canopy and evaporated again. There are some claims that land use could affect precipitation pa tterns and forestation will increase rainfall, or conversely, that de forestation will reduce rainfall. Most of the time this assertion is related to the positiv e effect, or feedback loop, between forests and local or recycled precipitation formed by the evaporation of water trapped in the tree canopy. There is the notion that variability of land surface at mesoscale can influence the amount of precip itation and its spatial distribu tion. It is claimed that any local wind circulation that c oncentrate water vapour from transpiration or wet canopy evaporation, favour the formation of clouds ( Bonell, unpublished ). Rainfall is said to increased by a combination of factors su ch as energy budget, frictional effects, changes in horizontal convergence and vert ical velocities; but also more rapid evaporation from intercep ted rainfall over vegetation. Regional models have shown that, at a mesos cale level, total conversion of land use in the Amazon will lead to decrea ses in rainfall. The magnitude of this change is not a common figure and it reflects the degree of uncertainty involved. It might range from estimates as high as 50% ( Saleti et al 1979 ), later dismissed because of ambiguous estimation techniques, to a low value of 6% estimated by the Institute of Hydrology in 1994 ( Calder 1999 ) (see Table 8-1 ). Table 8-1 Model results of relation of deforestation and rainfall (in large basins) Effect on preci pi tation Description Source 50% reduction Following total land conversion in the Amazon ba sin. Estimation techniques and assumptions used do not sustain the assertion. Saleti et al (1979) (cited in Bonell 2001 ) 25-35% reduction Following total land conversion in the Amazon ba sin. Minimum of <10% near the Amazon river estuary and >50% in the SW at the foothills of the Andes. Eltahir & Bras (1994) (cited in Bonell 2001 ) >20% reduction 34% reduction Following land conversion for the Southern part of the Amazon basin (modeling resolution of 500 km). Amazon basin, length scale of 2750 km. Trenberth (1999) (cited in Bonell 2001 ) 6% Following total removal of the Amazon basin. The effect will be higher in the drier northeast of the continent, at about 0.5 mm per day. Institute of Hydrology 199 4 (cited by Calder 1999 ) Not changed Studies of historical rainfall records in Calder 1999 51


Southern India failed to show any decrease despite the large-scale conversion of the drydeciduous forest to agriculture. 27% Estimated percentage of local rainfall for th e West African region. Gong & Eltahir (1996) (cited in B onell 2001 ) significant effect Role and effe ct of meridional conditions of land surface (vegetation cover and soil moisture) in the dynamics of the west African monsoon and rainfall variability. Zheng & Eltahir (1998) (cited in B onell 2001 ) significant reduction Reductions in rainfall in Florida, USA, where mu ch of the summer rainfall depends on local evaporation from the everglades, and much of these areas have been subject of continuous land use changes. Pielke et al (1999) (cited in Bonell 2001 ) 30% increase Estimated from a complete forest cover in a 400x 400 km by 30% compared to a 3-D mesoscale model simulation for bare soil in south-west France. The effect is strengthened by frictional effects associated with the passage of weather fronts from the sea to land along coastal areas. Blyth et al (1994) (cited in B onell 2001 ) Perturbation associated with land clearan ce include (Wilson and Henderson-Sellers 1983): increased surface albedo (reflection capacity of solar radiation, with white clouds and snow reflecting about 90% of radiat ion and dark tropical ocean absorbing nearly all of it); perturbation of the carbon cycle and greenhouse effects; local changes in the water balance; addition of particulates to the troposphere, both directly from com bustion and by increasing the wind-blown dust, and perturbation of the hydrological and turbul ence characteristics over areas where tall forest stands are replaced by low crops of cleared land. During rainless periods, tropical montane cl oud forests (TMCF), such as the existing forest in the upper part of M onteverde, Costa Rica, play an essential role in providing water downstream through th e capture of water from the clouds by capturing and condensing cloud droplets through vegetation surfaces, a process known as horizontal precipitation (HP) ( Bruinjzeel and Proctor 1995 ). The quantity of HP depends both on vegetation factors (height of vegetation, canopy size and structure, biomass, type of leaves and epiphytes) a nd climatic conditions (moisture content, drop sizes, velocity and direc tion of passing air, orientati on of forest ridge or slope, and duration of the process). Total reco rded amounts of HP could vary from 70 mm for an elfin forest in a 3100 m in Ven ezuela to 940 mm in a 1300 LMF in eastern Mexico (cited by Bruinjzeel and Proctor 1995 ). Through a combination of horizontal precipitation and low evapotranspiration rates, values of annual streamflow for TMCF (expressed as a ratio of incident rainfall ) are among the highest reported for any tropical forest. It is still uncertain the effect of forest clearing on total wa ter yield, as the final effect will partly d epend on the relative magnitudes of HP and evapotranspiration of original 52


and new vegetation cover. If contributions of HP are very high (and evaporation very low) then streamflow will probably decrea se after conversion. However, if HP contribution is relatively low, it is almost ce rtain that forest cleari ng will cause soils to become wetter and water yield to increas e (Steinhardt 1979 and Blackie 1979, cited by Bruinjzeel and Proctor 1995 ). Nevertheless, seasonal flows might be affected differently, particularly if the infiltrati on capacity of the soil is affected (i.e. decreased) by land conversion. Reaching the soil surface: interception and runoff Evapotranspiration. Evaporation is probably the most difficult phase to calculate in the hydrol ogical cycle, yet it accounts fo r a considerable part of the water in circulation. This process can happen in two ways: evaporation from water surfaces, like rivers, la k es, reservoirs or ponds. It is relatively easier to estimat e this variable if the wa ter body capacity is known and there is no leakage. in the form of transpiration from vegetation (also known as evapotranspiration ), as water is lost through interception of precipitation by vegetation leaves and transpired water from the plants. It is very difficult to estimate, and final amounts of eva potranspiration depend on the type of vegetation, its ability to transpire and th e availability of water in the soil. Evaporation depends crucially on water avai lable. If the body water disappears, then open water evaporation stops, while plants could keep drawing their water from the soil where moisture is held under tension. The rate of transpiration is finally affected y the stom ata in the leaves, or their capacity to transpire, the soil moisture content, and by meteorological factors. A series of physical factors affect both processes. These factors are (Shaw 1994): a) Latent heat required to change from l iquid into gaseous form. In nature it is provided by the energy of the Sun, in the form of solar (short-wave) and terrestrial (long-wave) radiation. b) The temperat ure of the air, affected also by the Sun. As the temperature of the air increases so does the amount of water vapour it can hold and it can vaporize faster. Evaporation is high in tropical regions and lower in polar regions. c) The saturation deficit o f the air or the amount of water that can be taken up by the air before it becomes saturated. This variable corresponds to the difference between the saturation vapour pressure at the air temperature and the actual vapour pressure of the air. Therefore, more evaporation occurs in inland areas where the air is usually drie r than coastal regions with damp air from the sea. d) Wind speed also affects evaporation. As wa ter evaporates, the air above the evaporating surface becomes more humid until saturation point and it no longer holds more vapour. If it is windy, drier air will substitute the humid air and evaporation will increase. Therefore, evaporation is greater in areas with plenty of air movement and lower in sh eltered localities wher e air tends to be stagnated. 53


Wind speed and air temperature could ha ve conflictive effects. Wi ndy areas (which increases evapora tion) are usually cooler (which should imply lower evaporation), and sheltered areas are of ten warmer. Over large catchment areas it is the general characteristics that will have the most important final effect. e) The atmospheric pressure or weather pattern. Low atmospheric pressure is usually associated with damp unsettled weather, with air well charged with water vapour and conditions are not conducive to aid evaporation. f) The nature of the evaporating surface, as it m odifies the wind pattern. Friction originated by a rough, irregular surface reduces wind speed, but tends to cause turbulence and the subsequent increase in the vertical component in the wind will enhance evaporation. On the other hand, strong winds over a flat, open water surface can cause waves whic h provides increased surface for evaporation and turbulence. There is little friction from wind passing through smooth surfaces, and evaporation here is mostly affected by horizontal velocity. If there exists a continual supply of wate r, then evapotranspi ration is regulated by meteorological cond itions. The evaporation plus transpiration from a vegetated surface with continuous water supply is known as potential evaporation or the maximum potential loss rate due to the prevailing meteorological conditions. If there exists a continual supply of wate r, then evapotranspi ration is regulated by meteorological cond itions. The evaporation plus transpiration from a vegetated surface with continuous water supply is known as potential evaporation or the maximum potential loss rate due to th e prevailing meteorological conditions. Diverse studies (summarized by Calder 1999 ) indicate that in both very wet and very dry conditions, evaporation from forests is likel y to be higher than from other shorter crops, due to higher atmospheric transport of water vapour from the rough surfaces in the former, and to the deep rooting of trees that allows them to obtain water from deeper soils as compared to shorter crops. The reduction of forest canopy through tree cutting decreases the evapotranspiration losses, resulting in increased water y ield in streams from the harvested area ( Hamilton and Pearce 1988 ). The magnitude of the effect de creases as regrowth of natural vegetation takes places, which happens mo re rapidly in high rainfall areas (6-10 years). The effect is similar if trees ar e replaced with short season arable crops, especially in areas with good rains and pronounced dry season ( Russell 1981 ). Hibbert (1967) and Pereira (1973) (both cited by Lal 1983 ) observed that cutting down natural forests increases streamflow at a rate ge nerally proportional to the reduction in forest cover over the basi n. Land management studies conducted in western Nigeria (a 44-ha ba sin) indicate that defore station increases runoff and interflow component significantly. During th e 3 year study, results show that at the beginning of the dry season in January, the baseflow increased from nearly zero to 3.2 mm/month. Possible explanations include: Direct storm runoff increases because of gradual d eterioration of surface soil structure and infiltration; 54


Baseflow increases because of gradua l decrease in bush regrowth and the no utilization of subsoil water by shallow-rooted seasonal crops; Storage cap acity is limited by the decr ease in organic matter content and in the relative proportion of retent ion pores in the soil profile. In summary, the increase in direct ru noff and base flow is associated with corresponding decreases in soil water storage, evapot ranspiration and surface detention (Lal 1983). Agronomic practices (crop and soil management) that causes frequent and prolonged exposure of the soil to raindrop impact permit much more surface runoff. Lal (1983) suggest several possi ble measures to avoid soil exposure in an agricultural context: Replacing natural vegetation with plantati on crops, such as rubber, oil palm, coffee, and cocoa, will eventually rest ore the soil-vegetati on equilibrium (Lal 1983, Russell 1981, Pereira (1973, cited by La nd (1983)) reported that deforestation followed by tea planting would cause lower runoff and er osion that if followed by re-planting. Agroforestry ma ximises output without increasing the risk of soil erosion. A combination of deep-rooted annuals can maximize water use and should decrease baseflow. Woody perennials on terrace banks can stabi lise the back slope and decrease the risk of their breakage and eventual failure. According to Russell (1981) planting faster-growing production forests to replace slow-growing indigenous forests, need not cause any disturbance to the river flow compared with the original natural forest, so there is no conflict between their role for production and watershed production. Reaching the soil. The soil is conforme d by different layers (see Figure 8-1 below). The top layer consist usually of vegetation litter and partly deco mposed debris lie on the surface above the A horizon, which is a la yer generally friable and rich in hum us. The B horizon is mainly composed of we ll weathered parent material, with its structure modified by roots and living crea tures. The C horizon is unconsolidated rock material with a wide range of particles and stone sizes. Below this layer there is usually consolidated bedrock. The thickness of these layers depend on geological structure and geomorphology of surface features Figure 8-1 Idealized soil section 55


Vegetation Litter partly decomposed A horizon B horizon C horizon Consolidated bedrock Vegetation Litter partly decomposed A horizon B horizon C horizon Consolidated bedrock Note: based in Shawn (1983), p.81 The function of the soil as wa ter store depends on the p acking of the clay or sand particles and the am ount of space between th e solids. Most of the water comes from melting snow or from rainfall, and it infilt rates the soil layers by gravity and surface tension through the pores of the soil, until it reaches the saturated layer of the soil where all the pore space is occupied by water. The surface over which the water pressure equals atmospheric pressure is defined as the water table. The wetness of the soil can be assessed in the form of volume fraction ( ), equivalent to the depth ratio of soil water (i.e. the equi valent depth of free water relative to the depth of soil for a unit plan area). Soil m oisture can be related to precipitation and evaporation depths. Table 8-2 shows that there is less water available at field capacity in a sandy soil as it drains quickly, and the retention capability of the soil increases with the clay content of the soil. At the pe rmanent wilting po int, a clay soil contains a significant amount of water. Table 8-2 Soil water content (volume fraction ) Type of soil Clay content (%) Saturation Field capacity Permanent wiltin g point Sand 3 0.40 0.06 0.02 Loam 22 0.50 0.29 0.05 Clay 47 0.60 0.41 0.20 From Shawn (1983), reproduced from Marshall and Holmes (1979) Soil Physics Cambridge University Natural forests usually help increase the cap acity of the soil to retain and infiltrate water down by providing a layer of mulch, leaf cover and humus. This rich debris is the most important factor contributing to wa ter infiltration, rather than the existence of trees per se For example, forestry plantatio ns with removed litter and heavy machinery passing through might compact the soil and actu ally reduce infiltration. According to Russell (1981), most perennial ri vers in the tropics arise in highlands with an excess of rainfall over transpirati on (R>T), and run to lower sections in the watershed wh ere the reverse usually occurs and lack of water could prove a potential problem during dry spell. The water leavi ng the area and the seasonal flows of the river depend on land use management in the upper part of the watershed, by keeping up the infiltration rate of rainwater into th e soil at least as a high as the normal 56


maximum intensity of rainfall as m easured over a suitable time period. This usually happens where the natural vegetation has b een little disturbed, for this normally produces a soil surface capable of absorbi ng the rainfall and of allowing it to percolate into the deeper subsoil and seep out into the river as springs (Russell 1981; ). Figure 8-2. Rainfall and Evaporation in the Watershed Context Upper Part of WS Lower Part of WS W = f ( R ET ) ET = f ( V, D, WH, SRO) Usually R > ET Usually R < ETwhere: W: Amount of Water R: Rainfall ET: Evapotranspiration V: Vegetation D: Depth WH: Water holding capacity of the soil SRO: Surface Runoff Upper Part of WS Lower Part of WS W = f ( R ET ) ET = f ( V, D, WH, SRO) Usually R > ET Usually R < ETwhere: W: Amount of Water R: Rainfall ET: Evapotranspiration V: Vegetation D: Depth WH: Water holding capacity of the soil SRO: Surface Runoff Source: Based on Russell, E.W. 1981 It might be profitable for the river basin as a whole to develop land use systems for the upland s that minimize transpiration dema nds there, therefore increasing the water available for use in the lowlands. Seasonality and Forests Low-flow effects: the sponge effect According to Hamilton and Pearce (1988) the so called 'sponge effect': has no scientific basis, and all controlled experiments have shown increased streamflow throughout the year after reducing forest cover, with the largest proportional increases in the dry season. One exception would be the case of severe soil erosion or soil compactation, where the ability of the soil to absorb water is greatly reduced and dry season baseflow is not maintained. This, however, is not a scientific report based on controlled experiment from any catchment. Afforestation or reforestation over large parts of the watershed will normally reduce streamflow during dry season. One possibility is to have continuous cuttings (thinnings) or final harvests in mosaic patterns over the reforested watershed. If properly done, reforestation could reduce erosion . Hamilton and Pearce (1988) re port that most of the case studies reported by Bosch and Hewlett (1982) the greatest percentage of increase in yield occurs in the low-flow period, as Some streams that ceased to flo w during the dry season remained perennial following cutting (Gilm our 1977 reported by Hamilton and Pearce 1988 ). This result can be explained because the ve ry de ep roots of natural forest and many perennial grasses will dry the soils to depths of four to five meters during dry spells without their transpira tion rated being largely affected, whilst short rooted crops tend to use water from the top (one to one and-a-half meters deep). ( Russell 1981 ) 57


This is a particular important result where dry season flows are particularly important. Cutting could be beneficial, provided that it was repeated when water yields have declined due to regrowth (Hamilton and Pearce 1988). As Hicks et al (1991) report, the increase in dry-season flow is likely to be short lived, and coul d even be reverted to lower levels than the original vegetation, if thirstier species take over following logging (see following case st udy). Since the eighties, Hough (1986) has given management suggestions involving tree rem oval in the semiarid miomboo woodlands of Southern Africa in order to increase dry season flows. In conclusion, the competing effects of ev aporatranspiration versus infiltration will either result in increases or decreases in dry-season flows ( Calder 1999 ). These effects are likely to be site specific and related to the type of so il in the area, as well as the type of forest and the precipitation amount. It is not possible, however, to affirm that reforestation (or afforestation) programmes will definitely result in increase dryseason flows. Aquifers Recharge A fairly event of water flow is obtained when the system of upland use maintains the infiltration r ate of rainwater into the soil at least as high as the normal maximum intensity of rainfall. These conditions tend to prevail when natural vegetation has been little disturbed, producing a soil surface capable of absorbing the rainfall and allowing it to percolate into deeper subsoils an d seep out into the river as springs ( Russell 1981 ). Tree-cutting activities decrease the evapotra nspi ration rate, which in turns normally results in higher stored soil moisture and th erefore more water available to recharge groundwater, springs, and wells ( Hamilton and Pearce 1988 ). The reduction in interception allows a greater percentage of rainfall to reach the forest floor. The additional precipitation would either cont ribute to higher runoff, evaporate or infiltrate the ground. If the understorey vegeta tion, litter and forest root mat are kept after tree-cutting, infilt ration rates need not be affected and groundwater recharge would be maximised. On the opposite, the use of heavy logging machinery would compact the soil and decrease infiltration rates, resulting in larger runoff amounts and lesser groundwater recharge. Flooding and storm protection "fuelwood cutting in the Middle Hills of Nepal is 'deforesting' the hills, decreasing forest area, and initiating a series of erosional and hydrologic effects that lead to destruction and death in the lower Ganges" (presented by Hamilton and Pearce 1988). The precedent assertion is one of many views widely held by foresters an d the media, making stormflow protection is a majo r concern for downstream communities affected by deforestation and/or degradat ion in the upper parts of the watershed. Floods are a natural phenomenon, in which ri vers discharge any excess water arising from occasional large rainfall events. Their eff ect is likely to be interpreted as good or bad, depending on the actors involved and th e intensity of the event. For example ( Calder 2002 ): Agricultural and fishing activities in the lowlands could benefit from midintensity floods that carry sediments and nutrients from the uplands, however, 58


as the flood intensity increases the ri sks and hazards of destruction also increase. Forestry activities in the uplands co ul d benefit from political support and funding for their activities in exchange of the perceived benefits that forests will have on reducing floods. However, for big events, tress can fall during the storm and block waterways. Wetlands located in the lowlands can be nefit from the seasonal effects of floods. Engineers benefit from t he creation of costly structures to alter the drain system of the watershed. Scientists (hydrologist, agronomists, so il scientists, econom ists, and social scientists) benefit from funding for their research in the area (as long as there is a problem there is a need for research). The media benefit from coverage and possi ble se nsationalist stories related to the floods. Politicians will respond to where votes lie. Development organ isations (local, national and international) will try to accommodate for a solution that can be easily defended. Floods are partly linked to local land use syst ems, but generally this is only true for sm all rainfall events and small scale wate rsheds. There is however a great temptation to blame floods on agricultural systems in the upper parts of the watershed, especially if deforestation is involved, and in more recent years, large floods have been linked to global warming. As explained in previous sections, all forests tend to have higher evaporation rates than other types of ve getation, and natural fo rests exhibit higher infiltration rates, due to porous soils a nd the existence of understorey and humus layers. The combination of th ese two factors generally cont ributes to lower runoff and lower soil erosion rates. This is not nece ssarily the case for forest plantations, where infiltrations rates could be reduced because: Natural understorey, mu lch, or humus layer usually does not exist (for example, in teak, eucalyptus and pine plantations); Lack of proper ma nagement activities involved with the preparation of the plantation and logging activities (des ign of drainage systems, road construction, road use, use of heavy machinery that compacts the soil, etc). Trees can have a positive effect in stabilizi ng slopes due to the binding effect of their roots. Stormflow, peakflow volumes and dur ation are usually increased by harvesting ( Ham ilton and Pearce 1988 ). Downstream effects are intensified if roads, skid trains and log landings have not been well prep ared and maintain. Results from Bosch and Hewlett (1982)'s review of 94 catchment expe riments indicate that in almost all the cases tree cutting without mechan ical logging resulted in increases in peak flows, and in many cases an increase in stormflow volumes ( Halmilton and Pearce 1988 ). Forest operations (not conversion to agriculture) in upstream catchments have not been shown to increase flood flows seriously in major streams. ( Hewlett 1982 ). However, the relative magnitude of th ese events is inverse to the magnitude, intensity or duration of the storms. For large storm even ts there is little impact of land use changes. 59


The role of forests as stream-flow regulator s has been extensively discussed since the beginnings of the 20th century, when during 1908-1911 foresters made use of this connection to advocate for the Weeks Act ( Saverwal, 1997 ). The Act, which proposed to federally buy and manage watershed fo rests in the interest of protecting commercial interests linked to navigati on of inland waterways, was opposed by ranching and timber interests but also engine ers who demanded clear evidence of such relationship. Several assertions like the one below were not based on any scientific evidence but rather on political grounds a nd were later on dismissed by the broader scientific community. Gifford Pinchot, Head of the Forest Service, talking on the Ohio floods in 1907: The great flood which has wrought devastation and ruin in the Upper Ohio Valley is due fundamentally to the cutting away of the forests on the watersheds of the Allegheny and Monongahela Rivers (Saberwal 1998). Land use and soil erosion Forests have long been attributed with lowe ring soil erosion. In fact, there are several cases when natural forest s are indeed beneficial ( Calder 1999 ): Natural fo rests have a high infiltration rate that reduces surface runoff and erosion transport; Tree roots increase soil stability, which together with the reduction in soil water pressure tend to reduce erosion; Forestry or agroforestry system might be preferred on steep slopes to retain mass movement of soil. Soils in the humid tropic are typically unstable. Quick de siccation following intense storms causes a surface crust th at d rastically reduces soil infiltration, especially where soils have lost their protective cover vegetation ( Lal 1983 ). Rapid deterioration in soil structure is partly because of low soil organic matter content. Non-wood products harvesting is unlikely to have ma jor detrimental impacts on water quality, with the ex ception of removal of forest litter ( Hamilton and Pearce 1988 ), which has been used traditionally for mosquito repellent coils (Philippines), livestock bedding and cooking fuel (Himalayas). If over-harvested, leaf litter removal could have serious effects on the hydrological regime Forest litter is not only important in nutrient cycle but has an extreme importan ce in protecting the soil from raindrop impact, minimizing slash erosion. In this sens e, the litter is considered in many cases more important than the high forest ca nopy. In fact, high forest canopy of largeleaved species can increase raindrop impact. Deforestation, method of la nd clearing and developmen t, and tillage system s significantly increase soil er osion. A monitoring exercise of 3-4 ha basins at IITA, Ibadan reported the following results (see Table 8-3 ) (Lal 1983): Soil erosion under dense na tural perhumid and seas onally hum id forests is usually low, resulting in low sediment lo ads of rivers draining tropical forested river basins; Traditional farm ing with incomplete clearing have minimal runoff and soil loss; 60


Complete manually clearing followed by m echanised farm operations reported 48 mm runoff and 5 ton/ha of soil loss over 3 years. Mechanically cleared land followed by mechanised far m operations reported 201 mm runoff and 15 ton/ha of soil loss. The most effective soil conservation system of land clearing and management was ma nual clearing followed by no-tillage. Soil er osion and runoff loss from shear blade clearing was within acceptable limits. Sediments from machine cleared plots was greater than that from manually cleared plots. Table 8-3 Effects of methods of deforestation and post-clearin g soil management on runoff and soil erosion* Treatment Basin area (ha ) Ru noff mm: Soil erosion (t/ha) 1979 1979-81 1979 1979-81 Forest 15 T T T T Traditional farming 2.6 3 6.6 0.01 0.02 Manual clearing+no-tillage 3.1 16 16.1 0.4 0.4 Manual clearing+conv. tillage 3.2 54 79.7 5 9.8 Shear blade clearing+no tillage 2.7 86 105 4 4.8 Tree pusher-root rake/no tillage 3.2 153 170 15 16 Tree pusher-root rake-conv.tillage 4 250 331 20 24.3 T=unmeasurable trace. Source: Lal (1983). from an alfisol for maize-cassava-maize-cowpea rotation from 1979 to 1981. Land was cleared in 1979 Soil erosion is usually most severe in the first year after the clear ing, but after the soil has stabilized, erosion depends more on postclearing soil m anagement than on the methods of land clearing (Lal 1983). The heralded adverse effects of deforesta tion and soil e rosion are mostly related to unsuitable forest management techniques, rather than the removal of trees per se. Some of them include (Calder 1999): bad logging techniques that increase the compaction of the soil; drainage activities th at might initiate gully formation; road construction can mobilize sediments; excessive subsequent grazing by animals that lead to soil compaction; rem oval of understorey and greater erosion risk; and splashinduced erosion from drops falling from the leaves of forests unto an unprotected soil. A study by Drs Edwards and Blackie (reported by Lal and Russell 1981) of the analysis of 16 years of data show that a com plete commercial tea estate with roads, houses, factory, offices and workshops can be developed in a stream source of tall forest without long-term damage to soil stability to the amount and regulation of streamflow. Areas prone to shallow debris slides because of steep slopes and soils with low or no cohesion benefit from stab ility provided by tree roots. In these situations, intensive wood products extrac tion can accelerate the landslide activity. Roads, skidding, or other ground disturban ces can also increa se this risk. According to Hamilton and Pearce (1988) increased stream sediment is not necessarily a consequence of logging, when careful logging and management practices are followed. However, this is not generally the case in most tropical countries, where sediments following tree -cutting are potentia lly distressing on 61


aquatic life, reservoir siltation rates, a ltered stream channels, and reduced water quality for domestic and industrial use. The use of streamside buffer strips and selective logging are the mo st important measures to control sediment delivery into the water. Litter extraction, and fuelwood harvesting temporarily break s the nutrient cycle process and part of the nutrient budget can be lost through leaching and water movement. Removal of foliage with wood re presents a further lo ss in the nutrients. Fast-growing fuelwood plantations which ar e totally harvested can cause declining productivity on many tropical soils (Jorgenson and Wells, cited in Hamilton and Pearce 1988). 62