Is Overgrazing Really Influencing Soil Erosion?

Abstract

Soil erosion is a serious problem spread over a variety of climatic areas around the world. The main purpose of this paper is to produce gully erosion susceptibility maps using different statistical models, such as frequency ratio (FR) and information value (IV), in a catchment from the northeastern part of Romania, covering a surface of 550 km 2 . In order to do so, a total number of 677 gullies were identified and randomly divided into training (80%) and validation (20%) datasets. In total, 10 conditioning factors were used to assess the gully susceptibility index (GSI); namely, elevation, precipitations, slope angle, curvature, lithology, drainage density, topographic wetness index, landforms, aspect, and distance from rivers. As a novelty, overgrazing was added as a conditioning factor. The final GSI maps were classified into four susceptibility classes: low, medium, high, and very high. In order to evaluate the two models prediction rate, the AUC (area under the curve) method was used. It has been observed that adding overgrazing as a contributing factor in calculating GSI does not considerably change the final output. Better predictability (0.87) and success rate (0.89) curves were obtained with the IV method, which proved to be more robust, unlike FR method, with 0.79 value for both predictability and success rate curves. When using sheepfolds, the value decreases by 0.01 in the case of the FR method, and by 0.02 in the case of the success rate curve for the IV method. However, this does not prove the fact that overgrazing is not influencing or accelerating soil erosion. A multi-temporal analysis of soil erosion is needed; this represents a future working hypothesis.

GIS-based Landform Classification of Eneoli thic Archaeological Sites in the Plateau-plain Transition Zone (NE Romania): Habitation Practices vs. Flood Hazard Perception

“…The summaries of archaeological site placement classified into 10 landform classes for the four combined versions of small-TPI and large-TPI neighbourhood sizes according to Criterion (2) are shown in Table 2 and Figure 6. were extracted using a 1000 m buffer zone around an Eneolithic site selected randomly from the study area: (a) calculation of TPI rasters for 100 m, 300 m, 600m, 1000 m, and 2000 m thresholds using the algorithm developed by [3] and [55]; (b) calculation of standardized TPI for each threshold rasters based on SD and Mean after the ArcGIS algorithm described by [75]; (c) generate the DEV models for each threshold rasters based on standardized TPI after [29]; (d) classification of landscape features into six slope position classes using the DEV and slope for each threshold rasters (Method 1) after [3]; The accuracy of the results has been verified using the visual interpretation of aerial imagery and by comparing the landform classification generated by TPI with the specific morphological features of over 100 settlement locations provided by previous geomorphological and archaeological surveys in the study area [39,[47][48][49][50][51][52][53][54]. Also, according to [28,29], which applied the same methodology of landscape classification within a heterogenous landscape in Belgium, in this case, the same DEV thresholds and various neighbourhood sizes were used; the results obtained by [28,29] are applicable for our study area, being a plateau-plain transition zone.…”

“…In this way, the front of cuestas provides a good defence of the settlement, offers a wider perspective on the landscape, and a very good (inter)visibility [47][48][49][50][51][52][53][54]. The sites located in the valleys have a lower relative frequency, and it is possible to correlate their placement with seasonal mobility for the practice of agriculture [35][36][37]75].…”

“…3850-3500 BCE) [35,[39][40][41]. Besides the literature on the morphological features in the study area [42][43][44][45][46], only a few works deal with the local or the relative topographic position of archaeological settlements in the landscape [47,48]; most of them refer to the evaluation of cultural heritage sites to natural hazards (landslides, gully erosion) and anthropogenic impact [49][50][51][52][53][54].Understanding the connections between the small-scale features, large-scale landforms, flood hazard perception, and the types of archaeological settlement is an important method applied in the study of the prehistoric peoples because the landscape can reveal insights into settlement distribution and dynamics over time [4,27]. This paper provides the first landform classification of 730 Eneolithic sites, using the TPI (Topographic Position Index) [3,5,28,55], and the SD (standard deviation) of the mean elevation, abbreviated as DEV by [29], around archaeological sites [3,4,29], which can classify the landscape in terms of slope position and landform categories and morphological classes based on the geomorphology [1,4,56,57].…”

“…3850-3500 BCE) [35,[39][40][41]. Besides the literature on the morphological features in the study area [42][43][44][45][46], only a few works deal with the local or the relative topographic position of archaeological settlements in the landscape [47,48]; most of them refer to the evaluation of cultural heritage sites to natural hazards (landslides, gully erosion) and anthropogenic impact [49][50][51][52][53][54].…”

GIS-Based and Statistical Approaches in Archaeological Predictive Modelling (NE Romania)

“…As highlighted by [39], the road network in this area has increased significantly from 0.67 km/km 2 in 1894 to 2.64 km/km 2 in 2012. The future A8 motorway [64], which is projected to intersect the southern part of the study area, will be one of the biggest infrastructure projects from the north-eastern part of the country (Figure 5). Currently, the projected path of the motorway does not intersect any known Eneolithic sites; however, it overlaps with areas identified as having a high and very probability of "hosting" archaeological sites.…”