|USFLDC Home | Tropical Ecology Collection [Monteverde Institute]||| RSS|
This item is only available as the following downloads:
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xlink http:www.w3.org1999xlink xmlns:xsi http:www.w3.org2001XMLSchema-instance
leader 00000nas 2200000Ka 4500
controlfield tag 008 000000c19749999pautr p s 0 0eng d
datafield ind1 8 ind2 024
subfield code a M39-00226
El uso de macroinvertebrados como indicadores de calidad del agua a lo largo de dos Corrientes en un paisaje alterado en Caitas, Costa Rica
The use of macroinvertebrates as water quality indicators along two streams in a disturbed landscape in Caitas, Costa Rica
Human alteration of natural landscapes often has significant impacts on nearby aquatic ecosystems, affecting both abiotic and biotic factors. Water quality testing has primarily depended on the monitoring of abiotic factors; however, the use of biotic indicators has emerged as a valuable tool. This study looks at the water quality of two
joined streams in Caitas, Costa Rica, a region dominated by coffee and dairy production. The abiotic indicators measured included dissolved oxygen, temperature, nitrogen, phosphorous, pH, and turbidity. Macroinvertebrates were sampled as the biotic indicators. Principal Components Analysis was used to determine the similarity between sites. Based on BMWP-CR scores, one site had bad, very contaminated water quality; six sites had bad quality
contaminated water; and three sites had water with regular quality, eutrophic, medium contamination. There were no significant correlations between the biotic and abiotic indicators of water quality. BMWP-CR scores were significantly correlated with S and H; indicating that as S and H increased, so did BMWP-CR. This study indicates that human alteration of natural landscapes is likely affecting aquatic ecosystems and the macroinvertebrates that inhabit them; however, because there are no correlations between biotic and abiotic factors, it cannot conclude how exactly these alterations are affecting them.
La alteracin humana de los paisajes naturales a menudo tiene un impacto significativo sobre los ecosistemas acuticos cercanos, afectando tanto los factores abiticos y biticos. Pruebas de calidad del agua ha dependido principalmente en el seguimiento de los factores abiticos, sin embargo, el uso de indicadores biticos se ha convertido en una herramienta valiosa. Este estudio analiza la calidad del agua de dos corrientes se unieron en Caitas, Costa Rica, una regin dominada por el caf y la produccin de leche. Los indicadores abiticos medidos incluyen oxgeno disuelto, temperatura, nitrgeno, fsforo, pH y turbidez. Se tomaron muestras de macroinvertebrados como indicadores biticos. Anlisis de Componentes Principales se utiliz para determinar la similitud entre los sitios. Sobre la base de las puntuaciones BMWP'-CR, tena un sitio de mala calidad, muy contaminado el agua, seis centros contaban con mala calidad del agua contaminada, y tres sitios tenan el agua con una calidad regular, eutrficos, la contaminacin del medio. No hubo correlaciones significativas entre los indicadores biticos y abiticos de la calidad del agua. CR-BMWP' resultados se correlacionaron significativamente con S y H ', lo que indica que como S y H' aumenta, tambin aumenta BMWP'-CR. Este estudio indica que las alteraciones humanas de los paisajes naturales es probable que afectan a los ecosistemas acuticos y los macroinvertebrados que habitan en ellos, sin embargo, porque no hay correlaciones entre los factores biticos y abiticos, que no se puede concluir exactamente cmo estas alteraciones son las mismas.
Text in English.
Aquatic organisms--Effect of water pollution on
Costa Rica--Puntarenas--Monteverde Zone
Organismos acuticos--Efecto de la contaminacin del agua en
Costa Rica--Puntarenas--Zona de Monteverde
Tropical Ecology Summer 2010
Ecologa Tropical Verano 2010
t Monteverde Institute : Tropical Ecology
The Use of Macroinvertebrates as Water Quality Indicators along Two Streams in a Disturbed Landscape in Caitas, Costa Rica Emily Thorpe Environmental Studies Program, Salisbury University ABSTRACT Human alteration of natural landscapes often has sign ificant impacts on nearby aquatic ecosystems, affecting both abiotic and biotic factors. Water quality testing has primarily depended on the monitoring of abiotic factors; however, the use of biotic indicators has emerged as a valuable tool. This study lo oks at the water quality of two joined streams in Caitas, Costa Rica, a region dominated by coffee and dairy production. The abiotic indicators measured included dissolved oxygen, temperature, nitrogen, phosphorous, pH, and turbidity. Macroinvertebrates were sampled as the biotic indicators. Principal Components Analysis was used to determine the similarity between CR scores, one site had bad, very contaminated water quality; six sites had bad quality contaminated water; and three sites had water with regular quality, eutrophic, medium contamination. There were CR scores were CR. This study indicates that human alteration of natural landscapes is likely affecting aquatic ecosystems and the macroinvertebrates that inhabit them; however, because there are no correlations between biotic and abiotic f actors, it cannot conclude how exactly these alterations are affecting them. INTRODUCTION Human alteration of natural landscapes, including deforestation, agricultural uses, and urbanization, affects aquatic ecosystems and can be detrimental to aquatic s pecies (Karr et al 1985). Removal of forest within a watershed can lead to increased surface runoff, stream velocity, and debris avalanche erosion depending on precipitation, slope steepness, and soil type. Erosion of gravel roads produces fine sediment s, which can be the most harmful to stream water quality. Removal of riparian forest is also a common consequence of agriculture and intensifies erosion, often delivering increased loads of sediment to streams and causing increased turbidity levels. This sediment often carries pesticides, herbicides, and nutrients from fertilizers, such as nitrogen (N) and phosphorous (P), applied to agricultural fields within the watershed. Loss or reduction of near stream vegetation can also lead to increased stream te mperatures, altered channel structure, reduced inputs of leaf litter and woody debris and less retentiveness (Allan 1995). When the riparian zone is maintained or crops are planted on prairie rather than forested land, these changes in stream characterist ics may be less affected (Gregory et al 1991).
The Monteverde region of Costa Rica spans several life zones and contains many small communities including the town of Caitas, classified as a premontane moist forest, where this study was performed. The region is characterized by three seasons based on cloud and precipitation types: the wet season, the transition/windy misty season, and the dry season. Much of the Monteverde community, including Caitas, was deforested between 1920 and 1950, releasing nu trients from the organic matter and helping to spur early dairy production. Since the 1970s, dairy farmers have found it necessary to adopt new management practices that include rotational grazing, which helped to spread manure around pastures and reduce soil compaction; but also included the application of chemical fertilizers and pesticides (Griffith et al. 2000). Since then, coffee production in the region has also intensified and many farmers have switch from shade grown coffee to sun tolerant dwarf v arieties, resulting in the removal of shade trees and requiring heavy use of nitrogen fertilizer and pesticides. Aggressive weed control and the removal of shade trees expose the soil, intensifying erosion rates on coffee plantations. However, terracing can help to reduce this erosion (Griffith et al 2000). The aforementioned human disturbance of multifaceted riparian landscapes not only affects abiotic factors, such as DO, pH, temperature, turbidity, N, P, and velocity; but can also have detrimental eff ects on aquatic life such as macroinvertebrates, causing changes in their abundance, diversity, and composition (Allan 1995). Solely depending on chemical and physical measurements when evaluating water quality only provides us with instantaneous informat ion about water quality conditions. By conducting biological monitoring, we can get an idea of both past and current conditions because macroinvertebrates are sensitive to intermittent pollution that may not be present at the time of sampling. Benthic ma croinvertebrates are the most commonly recommended organisms for biological monitoring of stream water quality (Resh et al. 1996). These organisms are useful bioindicators because of their long life spans, relatively sedentary behavior, quick reaction to most pollutants, and similar abundance in primarily flowing aquatic systems around the world (Maue & Springer 2008). Previous studies have identified certain taxonomic groups as characteristic of clean water and others as characteristic of polluted water a nd assigned pollution tolerance scores to different taxa (Resh et al. 1996). This study uses both chemical and biological indicators to investigate water quality in two streams flowing through a previously disturbed landscape, primarily consisting of pas ture and coffee fields, in Caitas, Costa Rica. I hypothesize that water quality of the stream in more forested areas that are farther removed from the disturbed landscapes of pasture and coffee fields will be better than sections of the stream in closer c ontact with these agricultural uses. I also expect to see that sites primarily in contact with coffee fields have lower water quality according CR than sites in contact with primarily pastured areas because of the previously described higher inpu ts to and erosion from coffee fields. This research is important for providing evidence of the impact of deforestation and other landscape disturbances, as well as for determining at what point aquatic ecosystems are less affected or can recover from distu rbances. Further research on aquatic macroinvertebrates also provides evidence for the inclusion of macroinvertebrate monitoring in standard water quality testing required by law.
MATERIALS AND METHODS Study Site My study sites are located in premontan e moist forest in the town of Caitas near Monteverde, Costa Rica. I identified two streams that flowed through human altered landscapes, coffee fields, cattle pastures, and roads. At none of the sites did coffee or pasture directly touch the stream; all sites had a forest buffer of at least 5 meters on one side, with some sites having forest buffers on both sides or greater buffers than others. One stream was temporal and about two thirds of its total length was flowing. The other stream, Quebrada Berr os, is larger and has constant flow throughout the year. Discharge from a nearby mechanic station empties into the beginning of the temporal stream; however, there is a gap in stream flow from where these inputs are discharged and where temporal stream fl ow begins. I took 4 samples along this stream and 1 sample where it met with the larger stream. Along the larger of the two streams, I took 3 samples before it met with the temporal stream and 2 samples after it met with the temporal stream. Site 6 was where the larger stream began from two springs on a cattle farm. Site 5 was sampled just downstream from the road near Trapiche on the farm of Victor Torres. Site 7 and 8 were taken even farther downstream near other cattle pastures. All other sites (1 4, 9, 10) were samples on the coffee farm of Guillermo Vargas (Table1; Fig. 1). All samples were taken during the month of July, 2010. TABLE 1. Ten sample sites with latitude, longitude, and elevation (m) in Caitas, Costa Rica. SITE GPS Coordinates E levation (m) SITE 1 1256 SITE 2 1190 SITE 3 1200 SITE 4 1220 SITE 5 1296 SITE 6 13 44 SITE 7 1241 SITE 8 1238 SITE 9 1264 SITE 10 1265
FIGURE 1. Map of the nearby roads, as well as the forest, streams and agricultural areas, indicate the 10 sites that were sampled during the month of July, 2010. Abiotic Water Quality Measurements I measured water quality using the following factors : temperature, turbidity, pH, dissolved oxygen (DO), and nutrient levels of phosphorous (P) and nitrogen (N). I took water samples to test for turbidity, N and P using the LaMotte colorimeter. I used an Oakton membrane electrode to measure DO and tempera ture at the sample sites. I used Fermont pH strips to test pH. Biotic Water Quality Measurements I collected aquatic macroinvertebrates as my biological indicator of water quality using a 12 inch benthic macroinvertebrate collection sieve, a white tray, and a pair of tweezers. I positioned the sieve in the water and dislocated and scooped up the rocks, sand, sticks, or leaf litter in front of it to dislodge the macroinvertebrates. I emptied the contents of the sieve into the white tray and
used the twee zers to remove the macroinvertebrates and placed them in a vial of 96% alcohol. I sampled each site for 90 minutes. In the lab, I identified the preserved macroinvertebrates to family using a stereoscope and macroinvertebrate identification guide. I det ermined water quality using macroinvertebrate family richness and the Biological Monitoring Working Party CR). The BMWP was originally created by Hellawell (1978), modified by Alba Tercedor & Snchez Ortega (198 8) and adapted to include CR is a system used to calculate a biotic index score based on the identification of macroinvertebrates to the family level (Resh et al 1996; Stein et al. CR assigns a value ranging from 1 to 10 to each family of macroinvertebrates (Table 2) (Springer et al 2007). Organisms that could not be identified to the family level were assigned a 0 and therefore did not contribute to CR index. For each site, I also Weiner diversity index and the number of families (S). Regression analyses were used to compare the biotic and abiotic indicators in order to det ermine any significant correlations. I also used Principal Component Analysis (PCA) to determine how similar the sites were. CR values according to Alba Tercedor (1996). Water quality BMWP Associated Color Waters with excellent quality > 120 dark blue Waters with good quality, no contaminations or obvious distortions 101 120 light blue Waters with regular quality, eutrophic, medium contamination 61 100 green Waters with bad quality, contaminat ed 36 60 yellow Waters with bad quality, very contaminated 16 35 orange Waters with very bad quality, extremely contaminated < 15 red RESULTS I collected a total of 1,005 macroinvertebrates from 36 families, excluding those that could only be iden tified to the level of order (Appendix, Table S1). Diptera was the most abundant order with 356 individuals identified, with the family Simuliidae comprising 237 of these. Iso (Isopoda) and Hydropsychidae (Tricoptera) were present at every site; however, Iso is not CR scores for the 10 sites ranged from 21 to CR occurred at site 2 and the CR and S w ere significantly correlated ( r = 0.91, PCA was performed to separate the 10 sample sites based on how similar they were. The PCA factor 1 of macroinvertebrate families explained 20.4% of the variance, while the second factor explained 16.9% (Fig. 4). The macroinvertebrate families of Baetidae, Leptophlebiidae, and
Leptoh yphidae made the largest positive contribution to the factor 1. The macroinvertebrate families of Chironomidae and Tipulidae made the largest negative contribution to the factor 1. The factor 2 distributed sites based on other differences in family comp ositions. The families of Gyrinidae, Gerridae and Simuliidae made the largest positive contribution to the factor 2. The families of Belosomatidae, Odontoceridae, Hebridae, Tabanidae, and Naucoridae made the largest negative contribution to the factor 2. The PCA results are fairly consistent with the CR scores and S; grouping sites with similar scores together and placing sites that had especially high scores (site 2) or especially low scores (site 1) apart from each other. The PCA res ults also reflect the differences in family composition at each site. Site 6 was located where the stream began and had a differing composition from the rest of the sites, placing it away from the rest of the sites in the PCA (Table S1 in Appendix). TA BLE 3: The abiotic water quality factors measured and the biotic factors of macroinvertebrate CR measured along two streams in Caitas, Costa Rica during the month of July, 2010. SITE DO (mg/L) Temp. pH Nitrogen (mg/L) Phosphate H (ppm) Phosphate L (ppm) Turbidity (FTU) S H' CR SITE 1 8.37 17 6.5 1 1 0.06 2 8 1.61 21 SITE 2 9.17 16.6 6 4 2 0.07 22 20 2.54 96 SITE 3 9.38 16.9 6 8 4.4 0.23 20 9 1.78 52 SITE 4 9.55 16.4 8 6 2.5 0.17 21 11 1.94 56 SITE 5 9 .53 16.8 8 4 2.3 0.09 22 11 1.69 51 SITE 6 8.67 16.1 6.5 2 1.5 0.45 16 12 2.16 40 SITE 7 9.6 16 8 3 1.4 0.06 13 18 2.25 86 SITE 8 9.45 17.1 8 1 0.8 0.08 11 13 2.04 62 SITE 9 7.67 17.1 6.5 1 1 0.11 3 11 1.83 39 SITE 10 9.21 16.8 8 2 1.3 0.13 10 11 1.15 54 CR score and number of macroinvertebrate families (S) per site based on 10 sample sites along two streams in Caitas, Costa Rica during the month of July, 2010 ; (r = 0.91, p < 0.001)
FIGURE 3: Correlation of div based on 10 sample sites along two streams in Caitas, Costa Rica during the month of July, 2010 ; (r = 0.76, p = 0.01). FIGURE 4. Plot of factor 1 and factor 2 of the Principal Component Analysis (PCA) numbered 1 10 based on sites samples for macroinvertebrates in Caitas, Costa Rica during July, 2010. Component 1 explains 20.4% and Factor 2 explains 16.9% of the variance between sites.
CR were significantly co rrelated with any of the abiotic CR and DO (mg/L) were the most closely correlated (R 2 = 0.34, p = 0.07 CR was second most closely related to turbidity (FTU) (R 2 = 0.28, p = 0.11, n = 10; Table 2). FIGURE 5: CR score as a function of dissolved oxygen (mg/L) at the 10 sample sites along two streams in Caitas, Costa Rica during the month of July, 2010; (p = 0.07, n = 10) Of the ab iotic factors, P (ppm) and N (mg/L) were the most strongly correlated ( R = 0.96, p < 0.001, n = 10) (Fig. 6). Turbidity and DO were also significantly correlated (R = 0.71, p = 0.02, n = 10) (Fig. 7). FIGURE 6: Correlation of phosphate high range and nitrogen levels at the 10 sample sites along two streams in Caitas, Costa Rica during the month of July, 2010 ; (r = 0.96, p < 0.001).
FIGURE 7: Correlation of dissolved oxygen (mg/L) and turbidity (FTU) based on 10 sample sites along two streams in Ca itas, Costa Rica during the month of July, 2010; (r = 0.71, p = 0.02). DISCUSSION CR for the 10 sample sites ranged from 21 96, indicating a wide range of water qualities within a fairly small sample area primarily dominated by pasture and coff ee fields in Caitas. Site 1 was the only site with water of bad quality and very contaminated. Sites 3, 5, 6, 9, and 10 had contaminated bad quality water. Sites 2, 7, and 8 had waters of regular quality with medium contamination and/or eutrophism. No sites had scores less than 15, indicating very bad quality, extremely contaminated water. No sites had scores between 101 and 120 or above 120, meaning that no sites had good or excellent quality water (Springer et al. 2007). I found representatives of families with the lowest pollution tolerance (BMWP scores of 10) at 6 sites (sites 2, 4, 5, 7, 8, 10) (families Heptageniidae, Hydrobiosidae, Perlidae, and Polythoridae). At sites 2 and 7, I found representatives of two families with the lowest tolerance (families Perlidae, Polythoridae and Heptageniidae, Perlidae respectively), while at the other sites I only found one family with this score. Perlidae was the most common of the lowest tolerance families with 67 individuals found at a total of 5 sites (si tes 2, 4, 5, 7, and 8). Perlidae belongs to the order Plecoptera. Compared with temperate regions, Costa Rican Plecoptera are less diverse and contain only one genus, Anacroneuria (Hanson 2000). Hydropsychidae was the only family that C R found at every site and are characteristic of good or regular quality water. CR, indicating that these are also good indicators of water quality since as water quality declines, so does fa mily richness and diversity.
CR in parts of the stream that were the most forested and the most removed from human disturbances such as coffee fields and pastures. The PCA explains the correlati on between sites by separating them based on CR score (96 regular quality, eutrophic, medium contamination) and was located in the most forested region of coffee fields encroached into the riparian zone (Fig. 1). Just CR score of 52, placing it in the bad water quality with contamination category. Site 3 is located near a coffee fie ld that has been planted within the riparian zone. Three other sites (1, 9, and 10) with bad CR scores of 21, 39, and 54 respectively) were located near coffee fields within the riparian zone as well. Site 1 had th CR score (21) and was located at the point where water washes down from a coffee field and into the stream. Ditches had been dug to try and keep sediment and nutrients from flowing down into the stream as well, but the findings suggest that this is not an effective way of protecting macroinvertebrates from coffee production. I also expected to find that cattle pastures had less impact than coffee fields on water CR. The sites that were predominately exposed to pa sture were sites 4 CR score (40 bad quality, contaminated). Site 6 was also located far away from any other sites, at the headwaters of the stream, and had a dissimilar species compositio n. Sites 4 and 5 were closer to the road CR scores were low as well (56 and 51 respectively bad quality, contaminated). Sites 7 and 8 were located much farther downst ream from the road, and although they were not heavily forested (both contained clearings and banana trees on one side and forest followed by pasture on the CR scores (86 and 62 respectively regular quality, eutrophic, m edium contamination) than the other sites located near pastures. Sites 4, 7, and 8 CR scores higher than the sites located near coffee fields (1, 3, 9, and 10). s consistent with a previous study conducted by Burnett (2009). This study focused on the effect of land use, specifically forest, coffee, and pasture, on stream water quality according to CR in San Luis and Caitas, Costa Rica. Burnett sampled 6 s ites three times each for a total of 18 samples so when looking at the results, we must take into account the more intensive CR scores of 107 (good quality, no contaminants or obvious distortions) a nd 75 (waters with regular quality, eutrophic, medium contamination). The pastures received scores almost identical to those of the forests. Coffee fields received scores of 70 and 90 (waters with regular quality, eutrophic, medium contamination). How ever, both of these studies work together to support that forest cover is important in the maintenance of water quality according to biotic indicators. CR and abiotic indicators, a finding consisten t with Burnett (2009). This helps to support the idea that macroinvertebrates
physicochemical factors (Resh et al 1996). This also helps to support the use of macroinve rtebrates as bioindicators because their populations were clearly affected by human landscape disturbance, but were not correlated with the investigated abiotic factors. The only CR was DO (mg/L). In CR and therefore aquatic macroinvertebrates were impacted by human disturbance at all of the 10 sample sites. The finding that none of the abiotic factors were significantly correlated to B CR suggests that abiotic factors do not do a sufficient job in water quality testing. Macroinvertebrate sampling should become an integral part of water quality monitoring protocol because biotic changes can be observed when abiotic changes, possibly caused by intermittent pollution such as runoff, cannot be detected. Site 2 was in the most heavily forested area and had the highest CR score. Even though this site did not achieve a rating of good or excellent water CR sc ore indicates that there is a necessary forest buffer that must be maintained or restored in order to protect aquatic ecosystems from human disturbances such as agriculture and roads. It is important to protect natural landscapes as well as the species th at inhabit them, especially in places such as Monteverde, which provides areas of restricted, isolated highland habitat that is important to many organisms (Wheelwright 2000). Future studies should further investigate the size of riparian buffers necessar y to maintain high water quality in streams, as well as further investigate possible correlations between abiotic and biotic factors. ACKNOWLEDGMENTS I would like to thank Pablo Allen for his guidance throughout this project, for helping me identify o ver 1,000 bugs, Zoidberg like and poop covered. I would also like to thank Karen Masters for suggesting this study site to me in the fi rst place. I am absolutely grateful to Guillermo Vargas for allowing me to do this study on his farm and for showing me around; I would have been and was absolutely lost out there without him. Gracias a mi familia tica, Nery, Victor, and Mauricio Torres, for their tolerance of my mediocre Spanish and muddy boots, as well as for letting me experience their beautiful farm for two weeks. A special thanks to my boyfriend, Brandon Keese, for all of his help, support, and useful map reading skills. A big than k you to my Mom and Dad, as well as my advisor, Dr. Lewis, for giving me the opportunity to study abroad. Thanks to my roommates, Alex and Ellie, for putting up with my procrastination and late nights. Finally, I would like to acknowledge Charles, for he is the cheese to my macaroni, and this trip has made me realize the wonders of his stress research papers are due. LITERATURE CITED Allan, D.J. 1995. Stream Ecology: Structure and Function of Running Waters. Chapm an & Hall, New York. Burnett, A.S. 2009. The Effect of Land Use on Stream Water Quality in San Luis and Caitas. CIEE Fall 2009, Monteverde, Costa Rica.
Fernndez, L. and M. Springer. 2008. El efecto del beneficiado del caf sobre los insectos acuticos en tres ros del Valle Central (Alajuela) de Costa Rica. Revista de Biologa Tropical, 56: 237 56. Gregory, S.V., F.J. Swanson, W.A. McKee, and K.W. Cummins. 1991. An Ecosystem Perspective of Riparian Zones. Bioscience, 41: 540 51. Griffith, K., D. Peck and J. Stuckey. 2000. Agriculture in Monteverde: Moving Toward Sustainability In: Monteverde: Ecology and Conservation of a Tropical Cloud Forest, N. Nadkarni & N. Wheelwright. Oxford University Press, New York, pp. 394 407. Hanson, P. 2000. Insects and Spide rs In: Monteverde: Ecology and Conservation of a Tropical Cloud Forest, N. Nadkarni & N. Wheelwright. Oxford University Press, New York, pp. 95 147. Karr, J.R., L.A. Toth, and D.R. Dudley. 1985. Fish Communities of Midwestern Rivers: A History of Degradat ion. Bioscience, 35: 90 5. LaMotte. 2007. SMART2 Colorimeter Reagent Systems Test Instructions. Available in PDF on LaMotte website. Maue, T. and M. Springer. 2008. Effect of Methodology and Sampling Time on the Taxa Richness of Aquatic Macroinvertebrates and Subsequent Changes in the Water Quality Index from Three Tropical Rivers, Costa Rica. Revista de Biologa Tropical, 56: 257 71. Resh, V., M. Myers, and M. Hannaford. 1996. Macroinvertebrates as Biotic Indicators of Environmental Quality In: Methods in Stream Ecology. F. Richard Hauer & G. Lamberti. Academic Press, Inc., San Diego, pp. 647 55. Springer, M., D. Vsquez, A. Castro, B. Kohlmann. CR de la calidad del agua. s.1.: University EARTH. Stein, H., M. Spring, and B. Kohlm ann. 2008. Comparison of Two Sampling Methods for Biomonitoring Using Aquatic Macroinvertebrates in the Dos Novillos River, Costa Rica. Ecological Engineering, 34: 267 75. Vannote, F.L., W.G. Minchall, K.W. Cummings, J.R. Sedell, and C.E. Cushing. 1980. Th e River Continuum Concept. Can. J. Fish. Aquat. Sci., 37: 130 7. Wheelwright, N.T. 2000. Conservation Biology In: Monteverde: Ecology and Conservation of a Tropical Cloud Forest, N. Nadkarni & N. Wheelwright. Oxford University Press, New York, pp. 418 56.
APPENDIX Table S1: Total number and distribution of macroinvertebrates identified in each of the 10 study sites in Caitas, Costa Rica. Order /Family SITE 1 SITE 2 SITE 3 SITE 4 SITE 5 SITE 6 SITE 7 SITE 8 SITE 9 SITE 10 Totals Blattodea 1 1 Cu ca 1 1 Coleoptera 7 5 2 4 12 4 15 12 6 4 71 Carabidae 3 3 Chrysomelidae 2 2 Coleoptera 1 3 3 Curculionidae 1 1 Dryopidae 2 8 10 Elmidae 1 1 1 3 1 7 Gyrinidae 2 2 3 1 3 11 Hydrophilidae 2 2 Ptilodactylidae 2 2 8 5 2 19 Staphylinidae 3 1 1 1 2 2 2 1 13 Diptera 25 15 11 33 47 7 48 21 51 98 356 Chironomidae 19 4 6 16 3 8 23 13 92 Culicidae 1 1 Diptera 1 1 1 Simuliidae 9 11 27 31 38 21 18 82 237 Tabanidae 1 1 2 Tipulidae 5 1 3 2 9 3 23 Ephemeroptera 27 7 10 1 18 14 4 81 Baetidae 13 6 1 3 10 2 35 Heptageniidae 2 2 Leptohyphidae 6 9 8 4 27 Leptophlebiidae 8 1 1 5 2 17 Hemiptera 2 4 6 1 13 3 3 1 33 Bel ostomatidae 1 1 Gerridae 1 1 2 Hebridae 1 6 5 1 1 14 Naucoridae 2 1 5 2 10 Veliidae 3 2 1 6 Isopods 60 8 29 5 3 58 1 2 23 1 190 Iso 60 8 29 5 3 58 1 2 23 1 190 Odonata 2 1 1 1 5
Table S1. Continued. Calopter ygidae 1 1 Cordulegastridae 1 1 1 3 Polythoridae 1 1 Plecoptera 6 16 3 24 18 67 Perlidae 6 16 3 24 18 67 Psuedoscorpion 1 1 Pscorpion 1 1 Tricoptera 5 21 16 25 31 18 30 19 20 9 194 Calamoceratida e 3 7 1 4 2 1 18 Hydrobiosidae 1 1 Hydropsychidae 5 17 9 22 31 13 26 19 16 7 165 Odontoceridae 4 4 Philopotamidae 1 3 4 Tricoptera Unknown 2 2 Turbellaria 2 1 2 1 6 Planaria 2 1 2 1 6 Grand Total 9 9 91 67 100 100 100 139 92 100 117 1005