Assessing spatial variability of soil macro fauna and tree canopy using fractal theory (Case study: Riparian Forest of Maroon River)

Spatial distribution of soil macro-fauna and vegetation are controlled by several parameters. Distribution pattern of these variables have many variations. In current study, auto-correlation function and fractal theory were used to evaluate spatial variability of tree canopy and soil macro fauna diversity in Riparian Forest of Maroon River in Khuzestan. In this research, soil macro fauna were sampled using 175 sampling point along parallel transects (perpendicular to the river). The distance between transects was 100 m. We considered the distance between samples as 50 m. soil macro fauna was extracted from 50 cm×50 cm×10 cm soil monolith by hand-sorting procedure. Evenness (Sheldon index), richness (Menhinick index) and diversity (Shannon H’ index) were determined in each sample. Tree canopy was measured in 5* 5 plots.  The results showed that none of the variables had autocorrelation confirming lack of their significant spatial structure. The study of distribution and fractal behavior of variables showed intensive variability in the study area and spatial variability did not have remarkable structure. Fractal dimension of both variables also were high.
Extended Abstract
1-Introduction
 Sustainability and maintenance of riparian vegetation or restoring of degraded sites is critical to sustain inherent ecosystem function and values. There is an increasing recognition of fundamental role of space in population. For example, it is well recognized that habitat spatial heterogeneity can be an important contributor to the coexistence of species in communities. In spite of this, within spatial ecology, empirical and modeling studies have concentrated on aboveground biota, limiting belowground detail, if any, to abiotic properties of soil. However, spatial patterns of soil biota, and the factors that determine them, will influence spatial patterns of decomposition, nutrient supply and, ultimately, the spatial structure of plant communities. As such, spatial variability is perceived as a problem for understanding how highly species-rich soil communities are maintained, and what their ecosystem functions are. However, taking lessons from spatial ecology, it is likely that spatial variability is the key, rather than the obstacle, to understanding the structure and function of soil biodiversity. Soil macro fauna biodiversity plays a recognized role in the productivity and soil functioning of these systems. But the factors that drive its distribution are still poorly documented. Spatial distribution of soil macro fauna and vegetation are controlled by several parameters. Interactions between aboveground and belowground biota are among the main drivers of ecosystem processes in soils. As autogenic ecosystem engineers, plants modify food quality, quantity, and the microclimate of soil macro fauna. They modify the microclimate in their vicinity by cooling down the soil and air in the shade of their leaves. In the area of the research, the covering vegetation is highly variable, from dense to loose, which leads to heterogeneous habitats for soil organisms. So, distribution pattern of these variables might have many variations. Consequently, autocorrelation function and fractal theory were used to assess spatial variability of tree canopy and soil macro fauna diversity in Riparian Forest of Maroon River in Khuzestan.
2- Materials and Methods
The present study was carried out in Maroon riparian forest of the southeastern Iran (30^o 38^/- 30 ^o 39^/N and 50 ^o 9^/- 50 ^o 10^/ E). The climate of the study area is semi-arid. Average yearly rainfall is about 350.04 mm with a mean temperature of 24.5^oc. Plant cover, mainly comprises populous euphratica Olivie and Tamarix arceuthoides Bge and Lycium shawii Roemer & Schultes. Soil macro fauna were sampled using 175 sampling point along parallel transects (perpendicular to the river). The distance between transects was 100m. We considered the distance among the samples as 50 m. Tree canopy was measured in 5* 5 plots. Soil macro fauna was extracted from 50 cm×50 cm×10 cm soil monolith by hand-sorting procedure. All soil macro fauna were identified to family level. Evenness (Sheldon index), richness (Mechanic index) and diversity (Shannon H’ index), were determined in each sample. Classical statistical parameters, i.e. mean, standard deviation, coefficient of variation, minimum and maximum, were calculated. We calculated the autocorrelation of soil properties and macro fauna diversity to determine the spatial structure. As a next step, we investigated the frequency of variables to access the variability rate of soil macro fauna diversity and vegetation cover. The autocorrelation function provides information about the separation distance with which a measured value is related to its neighbors and, it is a manifestation of the fact that at or beyond the lag distance, observations will vary at random. Finally, we calculated the fractal dimension to describe deeply the spatial structure of variables.
3- Results and Discussion the results showed that none of the variables had autocorrelation and indeed confirmed the lack of their significant spatial structure. The study of distribution and fractal behavior of variables showed intensive variability in the study area. Fractal dimension of both variables also were high. Totally it seems that as a results of disturbance and pressure on this ecosystem spatial dependency of these variables had decreased to the level that it could be accepted the value of these variables are independent of each other.
4- Conclusions Spatial studies considerably improve the analysis of macro fauna distribution and vegetation cover. Fractal analysis yields more information about the macro fauna diversity and trees cover beside assessment of spatial pattern using autocorrelation. As revealed by autocorrelation analysis, macro fauna diversity and trees cover didn’t show any spatial structure but fractal analysis helps us to explore spatial structure that may not detected using autocorrelation to describe the spatial structure.