Project Topics
Get Complete Project Material File(s) Now! »
Materials and Methods
Study region and design
This research was carried out in the Meta Department of eastern Colombia. At roughly 200 m in elevation, the region experiences a humid tropical climate, with an average annual temperature of 26 °C, rainfall averaging 2500 mm yr-1, and a marked dry season between December and March (Decaëns et al., 2001). Sampling was conducted along a 200 km transect extending from Puerto Lopez to Carimagua (to the Northeast) and bounded to the North by the Meta River (Figure 1). Soil-based ecosystem services and biodiversity were assessed within five representative production systems in the region: 1) annual crops (rice, maize and soybeans), 2) rubber plantations, 3) oil palm, 4) improved pastures, and 5) semi-natural savannas. Fifteen fields for each production system were distributed regularly along the transect, such that all systems sampled were equally present in each of three municipalities studied (Puerto Lopez, Puerto Gaitan and Carimagua), yielding a total of 75 sampled fields (Figure 1).
Figure 1. Location of the 75 plots and associated land use sampled between June and July 2011, in the Meta Department, Colombia.
Ecosystem service evaluation and development of synthetic indicators
Within each field, three equally-spaced sampling points (200 m apart) were located along a 400 m transect, starting at least 50 m from the field edge. Ant communities and soil-based ecosystem services were sampled at each of these points between June and August 2011. Methodologies and results for the assessment of soil-based ecosystem services have already been detailed by Lavelle et al. (2014), and will thus not be described in full. Briefly, a suite of soil physical, chemical and biological parameters were measured to evaluate key ecosystems services including: nutrient provision, water regulation, maintenance of soil structure, biodiversity, and climate regulation. Nutrient provision services were assessed by measuring soil chemical fertility parameters (pH, SOM, total N, available P, Al saturation, cation exchange capacity, and macronutrient concentrations) at the 0-10 cm and 10-20 cm depths. Soil hydrological services (water regulation) were assessed by soil physical characteristics including volumetric and gravimetric moisture content, porosity, plant available water (based on water retention curves), aggregate stability, bulk density, penetration resistance and shear strength resistance. The maintenance of soil structure (particularly relevant for these easily compacted soils) was assessed by visual inspection of soil macroaggregates (> 4 mm) to determine the origins of soil structure (i.e., fauna vs. root vs. microbial or physical aggregation processes vs. non-macroaggregated soils). This method, outlined by Velasquez et al. (2007), provides an integrative measure of recent soil biological activity and offers an important proxy measurement for C stabilization, crop rooting potential, water infiltration and aeration. Soil biodiversity and biological activity was assessed via the collection and sorting of soil macrofauna (including ants) into 16 taxonomic groups (e.g., Oligochaeta, Isoptera, Coleoptera) largely separated by order (see details below for sampling of ant communities). Finally, climate regulation was evaluated by measuring aboveground C in living biomass, soil C in the surface layer (0-20 cm), and greenhouse gas (GHG) emissions of CO2, CH4, and N2O at monthly intervals between June and November of 2011.
The data sets generated for nutrient provision, soil hydrological services, maintenance of soil structure, and biodiversity were summarized into a set of four synthetic indicators of ecosystem services (as well as a more direct indicator for climate regulation mentioned below) according to methods adapted from Velasquez et al. (2007). Principal components analysis (PCA) was used to determine the parameters that best capture the variance within each dataset across the five land uses sampled. This was done by selecting those with a significant contribution (>50% of the variance explained by the most influential variable) to either of the first two principal axes. The selected variables were then combined into a single value and scaled to a number ranging from 0.1 to 1.0 using a homothetic transformation. Since widely accepted standard conventions already exist for climate regulation services, GHG fluxes were converted into global warming potential based on CO2 equivalents, while ecosystem C storage was calculated as the sum of aboveground biomass and soil C in the surface (0-20 cm) layer. The resulting numbers for all services were then scaled to between 0.1 and 1.0 (using the inverse value for GHG emissions, since high emissions imply a negative service and low emissions a positive service) and then averaged to generate a single indicator of climate regulation.
Ant diversity sampling and calculations
At the sampling points described above, ants were collected along with other groups of soil macrofauna by employing a modified TSBF collection method (Anderson and Ingram, 1993). Sampling involved the excavation and hand-sorting of litter and soil from a central monolith (25 x 25cm x 20 cm deep) and two adjacent monoliths (25cm x 25cm x 10 cm deep) located 10 m to the North and South of each central monolith. Standing plant biomass was cut 2-3 cm above the soil surface and removed prior to sampling. With three sampling points per transect, a total of nine soil monoliths were collected for each field. In the laboratory, ants were separated from other macrofauna and the samples were cleaned and preserved in 96% alcohol. Identification of ants to the genus level was performed following keys of Palacio and Fernandez (2003) and Bolton (1994); updated keys that are specific for each gender were used for finer level identifications according to AntWeb (http://www.antweb.org) and Longino (2003).
We performed a descriptive analysis of the percentage of collection by the different subfamilies, genera and species, based on morphological assessment of the number of ant species for each sampling point and land use. Effectiveness of sampling was estimated with species accumulation curves using the program EstimatesS v.8.2.0 (Colwell 2013).
Comparison of land uses
One-way ANOVA was used to compare abundance and species richness between the different land uses using JMP 10.0 statistical software (SAS Institute 2012). Analyses were conducted at the field level, using the average value for the three sampling points (nine soil monoliths) in each field. Tukey’s honestly significant difference was used to examine specific differences between the five land use systems. Natural log transformations were applied to meet the assumptions of ANOVA when necessary.
Identification of ant indicators using the IndVal method
Indicator ant species for ecosystem services were identified using the IndVal method developed by Dufrêne and Legendre (1997), based on the degree of specificity (occurrence of a species in a particular category, but not in others) and fidelity (frequency of a species in samples from a category) of a species to a particular habitat or soil condition (i.e., level of ecosystem service provision). The association of species with ecosystem service provision was accomplished by defining three categories of performance (irrespective of land use) based on the synthetic indicators described above. Each of the sampled fields was considered to have low (0.1-0.4), medium (0.4-0.7), or high performance (0.7-1.0) for each of the five consolidated ecosystem services, based on a similar methodology applied by Rousseau et al.
where: Aij represents specificity, N.ind (ij) is the average number of individuals of the species (i) within samples of a particular category (j) and N.ind (i) is the sum of the average number of individuals of species (i) across all categories examined. The fidelity is represented by Bij, where N.samples (ij) is the number of samples in category (j) where species (i) is present and N.samples (j) is the total number of samples in this category. Aij yields a maximum value when a species is present in only one category and Bij is maximum when the species is present in all samples of this category. The IndVal index was obtained by multiplying these two terms together, thus yielding a value which ranges from 0 to 100, such that 100 represents a perfect indicator species (Dufrêne and Legendre, 1997). Finally, Monte Carlo simulations were performed using 1000 simulation runs with PC-ORD v.4.34 (McCune and Grace, 2002). This test results in a p-value based on the proportion of randomized runs with an indicator value greater than or equal to the observed indicator value.
Results
Composition of ant communities
In total, 5154 ant individuals were collected, comprised of 91 ant species, 33 genera, and 9 subfamilies (Table S1). This corresponds to roughly 10 % of the species, 37 % of genera and 64% of the known subfamilies in Colombia (Fernández and Sendoya, 2004). Of all the taxonomic groups collected, 64 were identified to the species level and the remaining 28 are mainly morphospecies belonging to diverse genera with poorly known taxonomy. Among these genera, two of them, Pheidole and Solenopsis, contained 10 and 7 morphospecies, respectively, representing 18.6 % of total species richness. The most diverse subfamily was Myrmicinae, with 43 % of the species collected, followed by Ponerinae, with 14 % of the species observed. The genus with the highest number of species was Pheidole (11), and the most abundant species were: Brachymyrmex sp. 1 (1511 individuals), Acromyrmex sp. 1 (853) and Pheidole subarmata (551; Table 1). Assessment of the species accumulation curve suggests that the average sampling efficiency was 75.5%, with 29 species singletons and 18 species doubletons (data not shown). Although the sampling method used in this work is not specifically designed for the capture of ants, the effectiveness appears acceptable.
Table of contents :
CHAPTER 1: GENERAL INTRODUCTION
1.1 GENERAL CONTEXT
1.1.1 Importance of Soils and the ecosystem services they provide
1.1.2. Importance of soil macrofauna for soil functioning
1.1.3. The role of termite communities
1.1.4. The role of ant communities
1.1.4. Impacts of agriculture practices on soil fauna
1.2. THE CASE OF COLOMBIAN LLANOS
1.3. STRUCTURE OF THE THESIS AND GOALS
1.4 REFERENCES
CHAPTER 2: ANTS AS INDICATORS OF SOIL-BASED ECOSYSTEM SERVICES48
2.1 INTRODUCTION
2.2 MATERIALS AND METHODS
2.2.1 Study region and design
2.2.2 Ecosystem service evaluation and development of synthetic indicators
2.2.3 Ant diversity sampling and calculations
2.2.4 Comparison of land uses
2.3 RESULTS
2.3.1 Composition of ant communities
2.3.2 Indicators for ecosystem services
2.3.3 Comparison of ant communities between land use systems
2.4. DISCUSSION
2.4.1 Developing indicators of soil-based ecosystem services
2.4.2 Response of ant communities to manangement
2.7 REFERENCES
2.8 ANEXES
CHAPTER 3 : INFLUENCE OF LANDSCAPE AND SOIL PROPERTIES ON SOIL ENGINEERS COMMUNITIES
3.1 INTRODUCTION
3.2 MATERIALS AND METHODS
3.2.1 Study region and sampling design
3.2.2 Physical and Chemical Soil analyses
3.2.3 Ants and termites biodiversity
3.2.4 Statistical analysis
3.3 RESULTS
3.3.1 Soil engineers diversity
3.3.2 Effect of region or land use on soil properties
3.3.3 Effect of region or land use on community structure
3.3.4 Effect of environmental factors on soil engineer species occurrence
3.4 DISCUSSION
3.4.1 Effect of the regions or land uses on soil properties
3.4.2 Effect of the regions or land uses and ant or termite communities.
3.4.3 Effect of soil properties, regions and land uses on ant or termite species
3.5 CONCLUSION
3.6 REFERENCES
3.7 ANNEXES
CHAPTER 4: RESPONSE OF ANT TRAITS TO AGRICULTURAL CHANGES
4.1 INTRODUCTION
4.2 MATERIALS AND METHODS
4.2.1 Study area
4.2.2 Sampling scheme
4.2.3 Ant identification.
4.2.4 Soil characteristics
4.2.5 Trait description
4.2.6 Data analysis
4.3 RESULTS
4.3.1 Joint analisys of ecological traits, environmental variables and ant community
4.3.2 Classification of species based on environmental variables and species ecological traits
4.3.3 Co-inertia analysis between morphologic and ecological traits
4.4 DISCUSSION
4.4.1 Joint structure between ant species distribution, environment and species traits
4.4.2 What explains the ant species distribution?
4.4.3 Relation between morphological and ecological traits.
4.5 CONCLUSION
4.6 REFERENCES
4.7 ANNEXES
CHAPTER 5: DISCUSSION, PERSPECTIVES AND CONCLUSIONS
5.1 DISCUSSION
5.1.1 Overview and main results of the thesis
5.1.2 Species richness and composition
5.1.3 Effect of the agricultural landscape on soil communities
5.1.4 Ecosystem services and soil engineers
5.1.5 An approach to landscape ecology
5.2. CONCLUSION AND PERSPECTIVES
5.6 REFERENCES
