Assessing different soil properties in primary forest, pasture, and regenerative forest in Los Llanos, Monteverde Arianna Wolff Department of Environmental Science UCLA EAP Tropical Biology and Conservation Program, Spring 2017 9 June 2017 ABSTRACT Anim al agriculture is one of the largest causes of deforestation, converting primary and secondary forest into pasture for cattle. By severely altering the landscape, major ecosystem changes occur, including changes in soil properties. I compared the soil of f our types of land plots in Las Llanos, Monteverde: primary forest, pasture, reforested pasture, and fallow pasture. I tested the compaction, decomposition rate, presence of water stable aggregates, and arthropod diversity within each plot. Primary forest h ad significantly lower compaction, and seemed to have higher decomposition rates and higher proportions of large aggregate sizes. The reforestation site resembled both pasture and fallow pasture, especially in the case of compaction and decomposition. The four sites had individuals from many of the same arthropod orders, but primary forest had the highest number of individuals. Of the four sites, the primary forest seemed to have better soil quality than the others. The lack of regeneration in the reforeste d soil may suggest that forest does not return quickly to its natural state. Thus, when considering ecosystem protection, conservational efforts should be prioritized. Estudio de dif erentes propiedades del suelo en bosque primario, pasto, y bosque en reg eneracin en Los Llanos, Monteverde RESUMEN La agricultura animal es una de las mayores causas de deforestacin, ya que convierte el bosque primario y secundario en pastos para el ganado. Al alterar el paisaje, ocurren cambios importantes en el ecosistema, incluyendo cambios en las propiedades del suelo. Compar el suelo de cuatro tipos de uso de tierra en Las Llanos, Monteverde: bosque primario, pastos, pastizales refores tados y pastos en barbecho. Estudi la compactacin, la tasa de descomposic in, la presencia de agregados estables en agua y la diversidad de artrpodos en cada sitio El bosque primario present una compactacin si gnificativamente menor, y parecier a tener mayores ndices de descomposicin y mayor proporcin de agregados grandes. El sitio de reforestacin present resultados similares al de pastizales y pastos en barbecho, especialmente en el caso de compactacin y descomposicin. Los cuatro sitios presentaron atrpodos de muchas de las mismas rdenes de artrpodos, sin embargo el bosque primario obtuvo el mayor nmero de individuos. De los cuatro sitios, el suelo del bosque primario parece tener mejor calidad que los otros. La falta de regeneracin en el suelo reforestado sugiere que el bosque no regresa rpidamente a su estado na tural. Por lo tanto, al considerar la proteccin de los ecosistemas, los esfuerzos de conservacin deben ser priorizados.
Soil properties of four differently managed land plots Wolff 2 Cattle ranching which often entails deforestation, introduction of novel organisms, and in many cases the introduction of agro chemicals, has intense effects on cultivated land and surrounding ecosystems. It is a wid espread practice in Costa Rica and has the potential to severely reduce soil quality. Soil quality assesses the success and presence of biologic al and biochemical processes fundamen tal to ecosystem function (Dick 1994). Soi ls vary, especially in response to use and management, and these changes can be measured and tested (Larson and Pierce 1994) In response to the deforestation of primary forest for cattle pasture, reforestation projects attempt to regenerate natural habitats. At La Calandria Biological Station in Los Llanos, Monteverde, reforestation projects started in 2001 have aimed to reforest some land that was once pasture. In order to und erstand if reforested areas are truly regenerating the natural habitat, it is crucial to understand the soil conditions The quality of s oil properties can provide information on overall ecosystem health. To determine the soil quality of various types of managed land, I assessed four different properties : (1) soil compaction (2) presence of water stable aggregates (3) decomposition rate (4) arthropod diversity. Soil compaction is essentially a measure of soil density; s oil that is more compact will have hi gher density and less porosity. Soil compaction can influence root growth, inhibiting smaller roots from growing in more compacted soil (Materechera et a l 1992). Aggregates are particles of soil that bind together. Testing their water stability measures how well they stay together under forces of water. Larger aggregates are generally considered favorable, as they increase soil oxygenation, decrease ero sion, and enhance plant growth (Kemper and Rosenau 1986). Decomposition rate m easure s how quickly organic matter decays. It is an essential process for nutrient regeneration within the soil In this study, arthropod diversity measures the number of orders found at each site. A study by W. A. Reiners and collaborators (1994) noted th at conversion of forest to pasture showed obvious and significant effects on species composition and richness. Arthropods are ess ential in trophic interactions. C hanging vegetation can decrease their d iversity thus altering the trophic structure, impactin g ecosystem function (Haddad et al 2009) Since cattle farming is such a large industry in Costa Rica (and the world), it is crucial to understand how this industry affects long term soil health. Additionally, it is valuable to understand how well refore station programs can recover the natural habitat, years after regenerative projects have been established. In order to assess this, I ask: How does soil compaction water stable aggregates decomposition rate, and arthropod diversity differ among pasture, reforested pasture, fallow pasture, and primary forest? METHODS I conducted my study in May 2017 at La Calandria Biological Station and an adjacent farm, Finca las Americas. La Calandria has various reforestation plots from multiple types of cultivated land. I studied two sections of La Calandria and two sections of Finca Las Americas, each with a d ifferent land history. Site 1 was current pasture used for cattle grazing ( Finca ) Site 2 was pasture that was left fallow since 2001 (La Calandria). Site 3 was reforested pasture (La Calandria); the site was reforested beginning in 2001. Site 4
Soil properties of four differently managed land plots Wolff 3 was primary forest that is virtually unto uched (Finca). The study sites were adjacent so in theory, the soil quality had the same st arting health before human activity. Additionally, each soil site was at the same elevation, so there is no differentiation due to altitude. On each site, I tested soil compaction, aggregate stability, decomposition, and arthropod diversity Soil Compact ion I used a soil core to find soil density and collected ten samples from each site ( n=40 ). Compacted soil is associated with higher bulk density (Brady 1984). For each trial, I collected a core of soil and measured its length The length was then multip lied by the soil. I then dried each soil sample overnight to evaporate excess water within the soil and eliminate mass discrepancies due to additional water weight. After weighing each sample separately, I calculated the density using the formula density = mass/volume I used ANOVA to determine differences Water Stable Aggregates I used a wet sieving methodology developed by Cambardella and Elliott (1993). This method divides aggregates by size, and stacked in sieves of increasing hole size the four sieve hole sizes were 125m, 250m, 500m, and 2000m I first collected and dried soil from each of the four sites. For each trial, I measured 300 grams of dried soil, and poured this into t he stacked sieve apparatus The sieves were submerged in a bucket of water for 2 3 minutes. The apparatus was then removed and re immersed for three minutes I drained excess water and the sorted aggregates were dried overnight separately After the drying period, I weigh ed the aggregates to determine the ir proportions in each soil sample. I conducted three trials for each plot a total of twelve tria ls. I used ANOVA to determine differences Decomposition Rate I used an adapted methodology from Cornelissen (1996). I collected leaves from the same tree, to keep the species uniform. I dried the leaves overnight to remove excess water. I then measured t hree grams of dried leaves and put the weighed amount into a mesh bag with small holes At each of the four site s I buried two bags 10 cm underground. After 12 days, I collected the bags, removed excess dirt, and dried the leaves over night. I weighed the leaves and assessed the decomposition rate by calculating the difference in leaf weight. Arthropod Diversity I built pitfall traps. I set up three pitfall traps at each site (12 traps in total) I ethodology (1964) I dug a five cm wide hole, the size of a plastic container, and placed the container in the hole, making sure the entrance of the cup was level with the ground. Each container was the same size. I mixed a solution of alcohol and water in each pitfall trap, so that the caught arthropods could not climb out. I placed elevated p etri dishes over the traps to protect them from rainfall I left them for 24 hours before collection. I identified c ollected arthropods to o rder level. I checked the traps on two different occasions.
Soil properties of four differently managed land plots Wolff 4 RESULTS Soil Compaction The primary forest plot was significantly less compact than the reforested, pasture, and fallow plots. (Figure 1, F (3,36) =31.01, p<0.0001). Compared to the primary forest plot, the reforested, pasture and fallow plots showed higher average density ( Figure 1 ). S ince s oil compaction is relate d with mass bulk density (Brady 19 84), these three plots are more compact than the primary forest. Among the pasture, fallow, and reforested plots, the soil density did not differ significantly T h e reforested plot had approximately the same average density a s the fallow and pasture plots and thus resembled pasture more than primary forest 0.82 0.83 0.56 0.85 0.00 0.20 0.40 0.60 0.80 1.00 1.20 -0.5 0.5 1.5 2.5 3.5 4.5 Density (g/cm 3 ) Trials Average Pasture Fallow Reforested Primary Figure 1. D ensity of soil cores. Each individual trial is denote d as an X. There were ten trials per plot, and the average of each set of trials is denoted with a circle.
Soil properties of four differently managed land plots Wolff 5 Water Stable Aggregates There was no statistically significant difference between all four plots. Note that the sum s of the average percentages are not exactly 100%, but range between 99.96 100.04%. This slight discrepancy occurs because these percentages are averages taken from the individual trials Pri mary and reforested plots show a general trend of having more 200 0 m aggregates 27.8% and 37.6% respectively ( Figure 2). Additionally, fallow and pasture plots had higher proportions of 500 m 250 m and 125 m aggregate class sizes. In all four plots, the 125 m aggregates had significantly smaller proportions than the other aggregates sizes. Decomposition Rate The decomposition rate of primary forest is higher than the rates of the reforested, pasture, and fallow plots; t he decompositi on rates among the reforested, pasture, and fallow plots showed no empirical difference. Pasture had a change of 0.1 and 0.2 grams over the twelve days. Fallow pasture also had a change of 0.1 and 0.2 grams. Reforested Pasture ha d a change of 0.3 and 0.1 g rams ( Figure 3). The data, therefore, suggests no major difference between the decomposition rates of these three sites The primary forest however, showed higher decomposition rate s than the three other sites with a change in mass of 0.8 and 0.9 grams. I did not conduct a statistical analysis, as there were only two trials per plot. 0.582 0.972 0.464 0.436 28.71 35.1 23.8 34.5 46 46.2 38.1 37.3 24.7 17.7 37.6 27.8 0% 20% 40% 60% 80% 100% Pasture Fallow Reforested Primary Percentage 2000m 500m 250m 125m Figure 2. Soil aggregate percentages by sieve size Each bar sums to 100%, and the proportions of each aggregate class are distributed within the bar.
Soil properties of four differently managed land plots Wolff 6 Arthropod Diversity I found a total of seven orders and one taxa of arthropods among the four sites. I collected the most individuals at the primary forest plot (432), the second most at the reforested plot (239), the third most at the pasture plot (229), and the least a t the fallow plot (126) (Table 1 ). I found high numbers of purple springtails (ranging bet ween 110 to 400 individuals) in each pitfall trap, which accounts for the large number of Collembola at each sit e. With the exception of the Coleoptera Acari, and Diptera, primary forest had the largest number of individuals within each order. I found similar orders of insects at each of the four sites with the exception of Diptera, which was only present at the pr imary forest site While I found a greater number of individuals at the reforested plot compared to pasture, the difference is only 10 individuals. 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 -0.5 0.5 1.5 2.5 3.5 4.5 Plot Type Pasture Fallow Reforested Primary Figure 3. Change in mass over a twelve day period Each data point represents the change in mass in one decomposition bag. There were two decomposition bags per site.
Soil properties of four differently managed land plots Wolff 7 DISCUSSION The decomposition and compaction tests showcased a trend of superior soil quality within the primary forest plot While I predicted that primary soil would show the best results of the four plots, I did not expect the reforested plot to closely resemble the pasture and fallow plots C attle farming has been shown to increase soil compaction, potentially resulting in severe str uctural damage (Mulholland and Fullen 1991). My results coincide with this study, as the pasture had relatively high compaction. However, the fallow and reforested plots also had high relative soil compaction. R eforestation efforts at La Calandria commenced in 2001, giving the reforested soil approximately 15 years to regenerate. ompaction was statistically the same as pasture. This may be due to a long temporal requirement for detectable changes in soil (Dick 1994) Additionally compaction may be related to the pasture, fallow, and reforested decomposition rates. A mong these three plots, higher compaction was associated with slower decomposition, and vice versa for primary forest. Compaction of soil may be associated with a decrease in porosity and oxygen pockets within the soil, causing a decrease in the microbial decompositio n process (Dittmer and Schrader 2000). Physical and biochemical changes in soil can deter rapidly from human land use (Islam and Weil 2000), but based on my results, regeneration of natural function is a much longer process. The water stable aggregates showed no significant difference among the four plots. Soil with more organic matter generally has higher stability (Haynes and Swift 1990 ) so I originally predicted that primary forest would have the largest proportion of the 2000 m aggregate class. However, Nyamangara et al (2001) found that cow manure treatments increased aggregate stability The influx of cow manure may increase the amount of organic matter within the soil, increasing its stability. Another study conducted by Islam and Weil (2000) found that grassland and naturally forested soils had similar aggregate stability. Based on my data, the reforested and primary forest sites may hav e higher 2000 m aggregate proportions but further testing with a larger sample size is necessary. The arthropod diversity analysis show s slightly contradicto ry results. Primary forest had the highest number of arthropod individuals in both total number and in five out of eight orders However, I also found most of the same orders at the pasture plot I was Plot Type Coleoptera Hymenoptera Acari Orthoptera Araneae Dermaptera Diptera Collembola Total Pasture 7 7 2 4 3 0 1 205 229 Fallow 2 5 2 5 1 0 1 110 126 Reforested 2 4 12 4 1 0 1 215 239 Primary 4 9 10 6 3 1 0 400 433 Table 1. Number of Arthropod Individuals Classified by Order and taxa (Acari) at Each Site The collected arthropod individuals are organized by the site they were found at and the order they belong to.
Soil properties of four differently managed land plots Wolff 8 initially surprised by these results, as I expected the simplified vegetation in pasture to minimize the amount of habitable arthropods. Many arthropods, however, are predators. If prey items can survive in pasture grass, than these predators will be attracted to the area. Additionally, I was only able to check the pitfall traps twice. Conducting more checks and laying out more traps may have provided for a better survey. During the two weeks I conducted my study, the rainfall was unusually heavy, which also may have accounted for the sma ll number of individuals found Considering my results, where primary forest soil seem to have better quality in at least two of the four categories, conservational efforts are of high importance My study alludes to the notion that soil does not regenerate quickly. The reforestation plot of 15 years tested more similar ly to pasture tha n primary forest. While there may have been trees, vegetation, and forest inhabitants, life underground may not be functioning as well as it once did Soil is the basis of healthy ecosystems and functioning biochemical processes. Understanding the slow cha nge and regeneration of soil and the rapid changes caused by humans may indicate the need for serious consideration when changing ecosystem landscapes for human related purposes. ACKNOWLEDGEMENTS I would like to thank my primary advisor, Sof a Arce Flores, for all her help and guidance. She stuck with me through my struggles and uncertainties during the course of this project I would also like to thank Federico Chinchilla, my secondary advisor, for his helpful suggestions. Thank you to Randy Chinchi lla at the Monteverde Institute who helped c a Las Americas for allowing me to conduct my study on their land. L Monteverde Institute for letting me use their lab space and supporting my project. LITERATURE Brady, N C. 1984. The Nature and Properties of Soil MacMillan Publishing Company. Cambardella C A and Elliott E T. 1993. Carbon and Nitrogen distribution in aggregates from cultivated and native grassland soils. Soil Science of Society of America Journal V 57 No. 4 pp. 1071 1076 Corenelissen J H C. 1996. An Experimental Comparison of Leaf Decomposition Rates in a Wide Range of Temperate Plant Species and Types. Journal of Ecology. V84 No. 4 p p. 573 582 Dick P, Richard. 1994. Soil Enzyme Activities as Indicators of Soil Quality. Soil Science Society of America. Dittmer S, Schrader S. 2002. Longterm effects of soil compaction and tillage on Collembola and straw decomposition in arable soil. Pedobiologia. V44 Issues 3 4 pp. 527 538. Greenslad P J M. 1964. Pitfall Trapping as a Method for Studying Populations of Carabidae (Coleoptera). Journal of Animal Ecology V33 No.2 pp. 301 310 Haddad N M, Crutsinger G M, Gross K, Haarstad J, Knops J M H, Tilman D. 2009. Plant species loss decreases arthropod diversity and shifts trophic structure Ecology Letters V12 Issue 10 pp 1029 1039.
Soil properties of four differently managed land plots Wolff 9 Haynes R J, Swift R S. 1990. Stability of soil aggregates in relation to organic constituents and soil water content European Journal of Soil Science. V41 Issue 1 pp. 73 83. Kemper W D, Rosenau R C. 1986. Aggregate Stability and Size Distribution. American Society of Agronomy Soil Science Society of America. Larson W E, Pierce F J. 1994. The Dynamics of Soil Quality as a Measure of Sustainable Management. Soil Science Society of America. Materech era S A, Alstron A M, Kirby J M, Dexter A R. 1992. Influence of root diameter on the penetration of seminal roots into a compacted subsoil. Plant and Soil. V144 Issue 2 pp. 297 303 Mulholland B, Fullen M A. 1991. Cattle tramping and soil compaction on loamy sands. Soil Use and Management. V7 Issue 4 pp. 189 193. Nyamangara J, Gotosa J, Mpofu S E. 2001. Cattle manure effects on structural stability and water retention capacity of a granitic sandy soil in Zimbabwe. Soil and Tillage Research. V62 Isssues 3 4 pp. 157 162 Reiners, W. A., Bouwman, A. F., Parsons, W. F. J. and Keller, M. 1994. Tropical Rain Forest Conversion to Pasture: Changes in Vegetation and Soil Properties. Ecological Applications 4: 363 377