EGU2020-8078, updated on 14 Jun 2023
https://doi.org/10.5194/egusphere-egu2020-8078
EGU General Assembly 2020
© Author(s) 2023. This work is distributed under
the Creative Commons Attribution 4.0 License.

Studying soil erosion rates through landscape fragmentation. A case study in Crete, Greece.

Dimitrios D. Alexakis2, Christos Polykretis1,2, and Manolis G. Grillakis1,2
Dimitrios D. Alexakis et al.
  • 1School of Environmental Engineering, Technical University of Crete, Chania 73100, Crete, Greece (grillakis@hydrogaia.gr)
  • 2Lab of Geophysical-Satellite Remote Sensing & Archaeo-environment, Institute for Mediterranean Studies, Foundation for Research and Technology-Hellas, Rethymno 74100, Crete, Greece (dalexakis@ims.forth.gr; polykretis@ims.forth.gr)

The regular patterns of soil erosion tend to change at different scales of observation, affecting the mechanism of soil erosion and its evolution characteristics. Fragmentation and land loss are two critical, interrelated processes that influence the entire landscape. In this study, we examine how the relationship between landscape fragmentation and soil loss is diversified in different scales and contexts. Thus, different Earth Observation (EO) products, in terms of spatial analysis, such as Landsat 8, Sentinel 2 and Planetscope imageries are utilized to search the influence of scale effect in fragmentation rate. Land use / Land Cover (LULC) maps were developed in different scales through the use of sophisticated classification algorithms. Following, FRAGSTATS software was employed to calculate spatial metrics in order to capture important aspects of landscape patterns such as edge density, largest patch index, number of patches, contagion etc. In this context we calculated fractal dimensions and Moran’s I spatial autocorellation statistics and used them to represent the degree of landscape fragmentation. Soil loss data, estimated in different scales were incorporated in the overall study as derived from RUSLE (Revised Universal Soil Loss Estimation) soil loss estimation model. Ordinary least square (OLS) and Geographical Weighted Regression (GWR) methodologies were applied in order to correlate both spatially and quantitavely soil loss rates with landscape fragmentation The results denoted the fact that areas of larger patches with least fragmentation suffer less from soil loss phenomena. The overall approach can be used as a road map in order to extract crucial conclusions about landscape’s diachronic evolution and how this is affected both from natural and anthropogenic interventions.

How to cite: Alexakis, D. D., Polykretis, C., and Grillakis, M. G.: Studying soil erosion rates through landscape fragmentation. A case study in Crete, Greece., EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-8078, https://doi.org/10.5194/egusphere-egu2020-8078, 2020.

Displays

Display file