Madeleine Wojack Amanda Lueke Carlos Guindon Environmental Sustainability Monteverde Institute Goucher/Mount Holyoke Program April 14 , 2014 Comparative Impacts of American Imported Cheese and Queso Blanco Abstract: The purpose of our project was to compare the impacts of American and Costa Rican dairy production. Although Costa Rican dairy production is pasture based and uses less resources, it is also done closer to tropical forests and can directly impact the biodi versity of these important ecosystems. We used previous studies on American and Costa Rican dairy production to conclude that American dairy production is so unsustainable that the potential loss of biodiversity in Costa Rica is preferable. We concluded th at local, pasture systems with good land management practices are better for all communities involved in dairy production. Resumen: El proposito de nuestra proyecto fue para comparar los impactos de la industria lechera en los Estados Unidos y en Costa R ica. Aunque la industria lechera de Costa Rica usa los apacentaderos y no usa muchos rescoursos, tambien las fincas son muy cerca de los bosques tropicales y pueden afectar la biodiversidad de estes ecosistemas importantes. Us Ã amos estudios anteriores sobr e la industria lechera de los Estados Unidos y Costa Rica para concluir que sistemas locales con apacentaderos y direcci Ã³ n buena de la tierra son mejores para todos las comunidades de lechera. Introduction: Animal agriculture is incredibly resource intensive compared to plant agriculture. Cows are particularly bad offenders, as they require more water, land, and feed per calorie of food produced than most domesticated animals. They also release large amount of methane, a particularly potent greenhouse gas Koneswar a n and Nierenberg 2008. Despite all these drawbacks to farming cows, global beef and dairy consumption is on the rise. Therefore it is important to consider the impacts of these products, and to explore which cow originated product s are the most sustainable. Many facets of dairy production contribute to greenhouse gas emissions and other forms of environmental destruction. Important factors to consider when examining the effects of dairy production are land use changes, water use a nd watershed impact, waste management, impacts on biodiversity, and feed production and transportation Koneswar a n and Nierenberg 2008. Land use
Wojack and Lueke 2 changes and impacts on biodiversity are particularly relevant in Costa Rica, home to 2.5% of the Earth s biodi versity, contained in just 0.03% of its surface area INBio website. Commercial dairy production in the United States has mostly been done on large scale, industrial farms. Cows are treated with anti biotics and hormones and are not given much room for p asture. Commerical cow feed made from corn and soy is the norm Benson 2008. The cows are mechanically milked, and the milk is transported and processed far away from the source. The cheese is then sold in the United States or exported. Dairy farming in Costa Rica has traditionally been all pastured cows and the cows have been hand milked, but we observed on a field trip that mechanical milking is becoming much more commonplace. The milk is generally transported shorter distances and when used to make che ese is less processed than American cheese. In the case of homemade cheese, the milk can travel almost no miles and undergoes very minimal processing. However, Costa Rica is home to one of the highest levels of biodiversity on the planet. Clear cutting of tropical forest for pasture land is one of the largest threats to biodiversity in both Costa Rica and the tropics generally INBio website. Biodiversity plays a crucial role in ecosystem resilience and other vital environmental services Myers 1996. Amon g its most important functions, biodiversity regulates climate, biogeochemical cycles, and hydrological functions and promotes soil protection, crop pollination and pest control Myers 1996. It is also very important for the eco tourism industry, one of t he two biggest employers in the Monteverde area along with Productores de Monteverde S.A., the local cheese factory Myers 1966; Porras 2005. Converting tropical forest to pasture has devastating impacts on biodiversity, particularly species extinction Buschbacher 1986. Even after letting pastureland lie fallow for many years, secondary forest does not immediately r eturn to mature forest structure and soil will not regain comparable nutrient levels for many years after, both of which discourage native, endangered animals who require forest interior as opposed to edge habitat from returning, such as quetzals Stokstad 2008.
Wojack and Lueke 3 Deforestation also can change the climate, which has serious impacts on ecosystem health and resilience Lawton et al. 2001. In the Amazon, 50% of all precipitation comes from evapotranspiration, making the forest one of the most important compo nents of the water cycle Buschbacher 1986. In Costa Rica, evapotranspiration decreased by 30% after mature rainforest in the northeastern region was cut down Buschbacher 1986. Evapotranspiration and leaf surface area are vital elements of the water cyc le in the cloud forest climate, with the majority of all precipitation cycling in the form of clouds and fog, captured by vegetation through condensation on leaves and other plant matter Lawton et al. 2001. Deforestation of the cloud forest, and even of areas below the cloud forest, has dire impacts on the rest of the region Lawton et al. 2001. Soil protection and health is another concern. Mature forests nurture and anchor soil, protecting it from erosion and adding nutrients through a rich leaf litt er. While tropical soil tends to be far nutrient poorer than temperate soils, this still holds true. Tanner et al . 2014 observed a prominent and distinguished sandy A horizon 10 cm deep below the O horizon, which they speculated must be caused by soil er osion, compared to mature forest soil which had a deeper O horizon and less sand. Tanner also found lower levels on carbon and lower carbon flux in pastureland than in forests, both mature and secondary Tanner et al . 2014. However, pasture and forest soi ls tend to have comparable levels of nitrogen, bulk density and root biomass Powers 2004, Tanner et al. 2014. Regardless, while there is a slight increase in soil nutrients directly following slash and burn, in the following years there is a gradual, sig nificant decrease Buschbacher 1986. The change can also affect soil pH, which disturbs soil microorganisms, vital to organic matter decomposition and nutrient cycling Potthast 2011. Commercial dairy production has been vital to the economic health o f Monteverde since the arrival of the Quakers. It wasn t until the arrival of ecotourism that the Monteverde community became less dependent on the dairy industry Porras et al. 2005 . Global trade has also cut into the profits from dairy farming, as impor ted processed American cheese is much cheaper than Costa Rican produced
Wojack and Lueke 4 cheeses due in part to agricultural subsidies in the United States. However, it is still very common for Monteverde families to make their own queso blanco using raw milk from local cows. Dairy products are an important part of the diets of Monteverde residents, as we have experienced with our host families. It is not unusual to be given a few slices of fried queso blanco for any meal of the day. In our experiences processed cheese is also the most commonly used cheese in restaurants. For these reasons, we chose these two cheeses for our analysis. Dairy production in Monteverde is important to consider because it is often done in close proximity to the cloud forest reserves. Life c ycle analysis takes every stage of a product s production and consumption into account. It gives a complete picture of a product s ecological impact, and is considered the best way to measure how green an object is. To analyze dairy products, we will mak e estimates of the energetic input needed to feed and water the cattle, the distance the dairy and finished products need to travel to get to market, and impact farms have on the local ecosystems. We also look at other environmental factors that relate to biodiversity and other aspects of environmental health. Materials and Methods: To start our research, we used various academic databases to find previous studies on the impacts of both concentrated animal feeding operations CAFOs and pastured animal farming in the United States. We used these studies to create a holistic picture of CAFO produced cheese and homemade cheese from pastured cows. We then gathered information to make estimates of how far the feed and milk would travel at each step, based on previous literature and our own personal experiences with how homemade cheese is p roduced in Monteverde. Results and Discussion :
Wojack and Lueke 5 While decreasing deforestation is generally preferable to slash and burn to make pasture Buschbacher 1986 , pasture is preferable and more sustainable than planting annual crops Buschbacher 1986. Even converting pastureland to cropland degrades soil. Powers 2004 found that the soil nutrient decreases from changing forest land to banana plantations was a 37% decrease in carbon and a 29.3% decrease in nitrogen Table 1 . The change from pasture to c ropland was a decrease of 34.6% for carbon and a 35.8% decrease in nitrogen Table 1 . This suggests that even with fertilization, nutrients are still not taken up permanently by the soil, possibly due to soil microbial losses, soil erosion and rain run of f Powers 2004, Potthast et al. 2011. In addition, pasturing livestock represents a far more closed loop system than agricultural crops. Livestock, including dairy cows, eat in the same place they defecate, ensuring that nutrients return to the soil and can naturally regenerate Barnes 2011. This is not only better for the soil and environment as far as raising livestock goes, it is better for the animals as well, producing far less trauma and waste, and decreasing the need for additional inputs such as fertilizers and feed Barnes 2011. Livestock in confinement operations produce approximately 500 million tons of waste annually in the US, causing a huge waste problem and generating far more methane and other greenhouse gases that would have been reduced in a pasture system due to diet and sustainable organic waste management Koneswaran and Nierenberg, 2008. The minimized need for additional fertilization is one benefit of pasture raised dairy due to the recycling of organic waste. It also reduces if not entirely removes the need for producing feed for livestock. In confinement style operations, cows eat an average of 50 lbs of concentrate a day. Eighty percent of all soybeans and over 50% of all corn grown in the US , the two largest cash crops in the country, go into livestock feed Koneswaran and Nierenberg, 2008. Feed constitutes an enormous share of t he greenhouse gas emissions , with 41 million metric tons of CO 2 produced every year from just producing fertilizer for growing feed alone Koneswaran and Nierenberg, 2008. Most fertilizers are
Wojack and Lueke 6 produced using the energy intensive Haber Bosch process of con verting atmospheric nitrogen, N 2 , to readily accessible ammonia, NH 3 Koneswaran and Nierenberg, 2008. This has a huge effect on the biogeochemical nitrogen cycle, particularly with regards to nitrogen run off and eutrophication Koneswaran and Nierenberg , 2008. Nitrogen run off and subsequent eutrophication in the Mississippi River Watershed from agriculture, particularly of corn and soybeans, the principal components of livestock feed, causes the Dead Zone in the Gulf of Mexico, a hypoxic area the size of Connecticut in which no life can survive Bruckner 2012 . These figures do not include the environmental costs of transporting the feed where an estimated 0.8 million metric tons of CO 2 are emitted annually just for the transportation of livestock feed and products across the world Konewsaran and Nierenberg, 2008. However, transportation emissions and effects are dwarfed when compared to the environmental costs of producing feed for confinement operations Koneswaran and Nierenberg, 2008. Even accou nting for the harmful effects of monocultures, including soil degradation , release of sequestered carbon, and erosion, a UN study by the Food and Agriculture Organization FAO found that North America as a region produced fewer greenhouse gas emissions G HG emissions than did Central and South America FAO 2010. The FAO found that North America produced approximately 15.5% of the world production of milk, but produced only 8% of world GHG emissions related to milk production, while Central and South Amer ica produced only about 11.5% of the world s milk but 16% of the related worldwide GHG emissions 2010 , Figure 1 . These findings can be attributed largely to climate and yield per hectare, which relates especially to the specific regional breed of cow , wi th most GHG emissions and fluctuations between regions occurring from CH 4 emissions Figure 2 . The more cows necessary to produce the same amount of milk, the higher the CH 4 emissions are Figure 1, 2. The study found that the yield in North America was an average of 8900 kg/year/cow, and in Central and South America it was 1500 kg/year/cow FAO 2010 , necessitating more cows and pastureland . However, the annual yield
Wojack and Lueke 7 for Monteverde in 2002 was approximately 8500 kg/year/cow, a comparable yield rate Porr as et al. 2005. In fact, CO2 levels were lower in Latin America than in the USA FAO 2010. The USA emitted 0.23 kg CO2/kg FPCM, while Latin America emitted only 0.11 kg CO2/kg FPCM FAO 2010 , Figure 3 . The efficiency statistic that lags behind North Am erica in Monteverde is the stocking rate, which, for organic farms in the US in 2005 was an average of 1.5 cows/hectare Benson 2008. In Monteverde, the stocking rate in 2002 was 0.6 cows/hectare, an increase in productivity largely attributed to technolo gical advances such as windbreaks Porras et al. 2005. Aside from soil damage, transportation, and biodiversity losses, livestock production uses a great deal of water. On average, cows need to drink 80 liters 21 gallons of water a day to maintain stas is, with an addition 2 .5 liters of water for every 1 liter of milk produced Singh et al. 2004. On average, Jersey cows, the pre dominant breed in the Monteverde region, produce approximately 22.7 liters 6 gallons of milk a day, drinking on average 137 liters of water a day Schivera 2005 , Figure 4 . With 230 producers in the Monteverde region contracted with Productores de Monteverde S.A., producing an average of 33,906 liters 8957 gallons a day, we can conjecture that there are approximately 1,493 c ows in the region as of 2005 , drinking approximately 204,168 liters 53,935 gallons of water a day Porras et al. 2005. In the US, there are about 60,000 dairy operations, with an average herd size of 135 cows Purdue. Holsteins are the predominant bre ed in the US, producing an average of 5 gallons of milk a day Benson 2008 . Using these figures, we can estimate that there are 8.1 million cows in the US producing milk, each drinking approximately 128 liters of water a day, leading to an overall estimat e of 1,032.8 m illion liters of water consumed by the dairy industry in the US every day Figure 4 . Of course, these estimates only cover water used solely for drinking. Singh et al. 2004 estimates that less than 1% of water use in dairy production is drinking water; the other 99% is embedded, or virtual water , from irrigation of fodder crops . In these calculations, Singh et al. 2004
Wojack and Lueke 8 only account for water from irrigation, and not from r ainfall or soil moisture, meaning that pasture raised dairy is far less water intensive with far less virtual water usage and a far smaller water footprint due to the lesser amount of crop feedstock they require compared to confinement operations. Using th ese rough estimates, the US dairy industry, which is predominately composed of confinement operations, uses 103,280 million liters of water, 103.280 billion liters of water, every day , with each cow consuming 12,570.62 liters of water a day, both drinking water and virtual water Figure 5 . CONCLUSIONS When trying to figure out whether to buy cheese from home or from abroad, there are some preconceived notions that spring to the mind of an environmentally conscious young woman of the world. Among these no tions are that it is always better to buy local, to buy small, and to buy humane. When examining the effects of dairy production in Costa Rica and the United States, Costa Rican modes of production small scale pasturing definitely win as far as environ mental costs and greenhouse gas emissions are concerned. However, when one accounts for loss of biodiversity, ecosystem resilience and associated local climate change, the answer seems less clear. In addition, pastureland in Costa Rica can be managed so it impacts surrounding forests as little as possible. Windbreaks, living fences, and forest patches all integrate pasture into the landscape and protect surrounding ecosystems Harvey et al. 2001 . Windbreaks in particular lead to an increase in the quality of pasture, with reports of a 20% increase in milk production by hectare Porras et al. 2005. All these structures keep biological corridors open through pasture land. Biological corridors facilitate the movement of forest dwelling organisms and keep the g enetic diversity of these animals healthy, as they are not cut off from each other in forest fragments Harvey et al. 2001 . Choosing native species of trees instead of potentially invasive species makes these efforts even more successful Harvey et al. 20 01
Wojack and Lueke 9 In the authors opinion, buying locally produced, particularly home made, dairy products is still better for the environment than buying imported cheese, even at the cost of biodiversity and climate change. The effects to the environment are smaller, and at a far smaller scale, and more humane , in the dairy production techniques of local ranchers. In 2005, there were 230 producers for the Productores de Monteverde S.A., with an average of 6.45 cows per herd. In the United States, the average herd size is 135 cows, a number skewed by the rise of 500+ herds Porras 2005, Purdue. In addition, while the majority of dairy operations in the U . S . are family owned, there was a 39% decrease in the number of dairy farms from 1998 to 2007 Benson 2008. There app ears to be a shift from smaller, family run operations to larger corporations, even if they are still family owned". It is also important to consider the implications of buying imported foods from foreign companies, particularly capital flight, as the pr ofits then immediately leave the country. If one buys cheese from one s neighbor, that money stays in the local economy for at least a little bit longer, stimulating economic growth and community resilience. While Productores de Monteverde S.A. is no longe r locally owned, buying milk from a neighbor with a cow and then making one s own queso blanco in the traditional, tico way, promotes not only the local economy but also traditional food ways and culture. Productores de Monteverde S.A. is still one of the biggest employers in the region, and without it, the only jobs left are the in the eco tourism and hospitality industries Porras 2005. The answer is clear buying locally and humanely in Monteverde is not only better for the environment, but for the com munity as well.
Wojack and Lueke 10 Works Cited Barnes, M.K. 2011. Low input grassfed livestock production in the American West: case studies of ecological, economic, and social resilience. Society for Rangeland Management. Rangelands , 332:31 40. Benson, G.A. 2008. Pastu re Based and Confinement Dairy Farming in the United States: An Assessment. North Carolina State University, Department of Agricultural & Resource Economics. Raleigh, NC, USA. Journal of International Farm Management 42. Biodiversity and Conservation. B iodiversity. Instituto Nacional De Biodiversidad INBio, n.d. Web. Apr. 2014. . Bruckner, M. 2012. The Gulf of Mexico dead zone. Montana State University. Web. Apr. 2014. < http://serc.carleton.edu/microbelife/t opics/deadzone/index.html >. Buschbacher, R.J. 1986. Tropical deforestation and pasture development. Oxford University. Oxford, UK. BioScience , 361:22 28. Facts about Holstein Cattle. Holstein Association USA, Inc. Brattleboro, VT, USA. Web. Apr. 2014. < http://www.holsteinusa.com/holstein_breed/holstein101.html >. Koneswaran, G., D. Nierenberg. 2008. Global farm animal production and global warming: impacting and mitigating climate change. Human Society of the United States. Washington, DC. Environmental Health Perspectives , 1165:578 582. Lawton, R.O., U.S. Nair, R.A. Pielke Sr., R.M. Welch. 2001. Climatic impact of tropical lowland deforestation on nearby montane cloud forests. Science . 294:584 587. Myers, N. 1996. Environmental services of biodiversity. Green College, Oxford University. Oxford, UK. Proceedings of the National Academy of Sciences of the United States of America, 937:2764 2769. Porras, I.T., R. Hope, M. Miranda. 2005. Introducing th e key issues: an exploratory paper looking at the socio economic impacts of biophysical changes in land use in Monteverde, Costa Rica. Department for International Development, UK. Centre f or Land Use and Water Resources Research. Newcastle, UK. Series: M arket Opportunities Associated with Hydrological Changes in a Tropical Montane Cloud Forest. Accessed April 2014 from . Potthast, K., U. Hamer, F. Makeschin. 2011. Land use change in a tropical mountain rainforest region of southern E cuador affects soil microorganisms and nutrient cycling. Institute of Soil Science and Site Ecology. Dresden, Germany. Biogeochemistry , 111:151 167. Powers , J.S. 2004. Changes in Soil Carbon and Nitrogen after Contrasting Land Use Transitions in Northeast ern Costa Rica . Ecosystems 7 2 : 134 146 . Purdue University School of Agriculture. 2008. Dairy facts. Food Animal Education Network. Purdue University. Web. Apr. 2014. < http://www.ansc.purdue.edu/faen/dairy%20facts.html >.
Wojack and Lueke 11 Schivera, D. 2005. Getting a famil y cow &lots to consider. Maine Organic Farmers and Gardeners Association. Maine Organic Farmer and Gardener, Winter 2005/2006. Web. Apr. 2014. < http://www.mofga.org/Publications/MaineOrganicFarmerGardener/Winter20052006/FamilyC ow/ >. Singh, O.P., A. Sharma, T. Shah, R. Singh. 2004. Virtual water trade in dairy economy: irrigation water productivity in Gujarat. Economic and Political Weekly, 2931:3492 3497. Stokstad, E. 2008. A second chance for rainforest biodiversity. Science , 320 5882: 1436 1438. Tanner, L.H., J. Curry, D.L. Smith, J. Twist. 2014. Effect of land use change on carbon content and CO 2 flux of cloud forest soils, Santa Elena, Costa Rica. Le Moyne College. Syracuse, NY, USA. Open Journal of Soil Science 4:64 71. Unite d Nations Food and Agriculture Organization. 2010. Greenhouse gas emissions for the dairy sector: a life cycle assessment. United Nations, New York.
Wojack and Lueke 12 Appendix Average % Change for Land Use Transitions Forest to Banana Pasture to Crop % C 0 10 cm 37.1 4.6 34.6 4.6 Soil C inventory Mg C ha 1 0 10 cm 16.5 4.8 33.0 4.3 % N 0 10 29.3 4.4 35.8 4.6 Table 1: Average percentage change for land use transitions for forest to banana and from pasture to annual crops Powers 2004 Figure 1: Relative contribution of world regions to milk production and GHG emissions from cradle to grave FAO 2010
Wojack and Lueke 13 Figure 2: GHG emissions/kg of FPCM FAO 2010 Figure 3: GHG emissions at farm gate from the processing of raw milk FAO 2010
Wojack and Lueke 14 Figure 4: Comparison of average amount of drinking water per cow in Monteverde versus in the United States. The average in Monteverde is 137 liters/cow/day, and in the USA it is 128 liters/cow/day. Figure 5: Comparison of virtual drinking water averages in Monteverde versus the United States. Virtual water accounts solely for movement of goods such as feed and does not measure rain or soil moisture. With virtual water accounted for, the average in Mo nteverde remains 137 liters/cow/day, while in the USA the number jumps to 12,570.62 liters/cow/per day. 137 128 0 20 40 60 80 100 120 140 160 Monteverde USA Average Liters/Day Drinking Water 137 12570.62 0 2000 4000 6000 8000 10000 12000 14000 Monteverde USA Average Liters/Day Drinking + Virtual Water