Identification of erosion-prone areas in connection with ranges of peak ground acceleration using fuzzy logic and entropy Shannon models in the Talar Drainage Basin, Mazandaran province

Soil erosion is a natural geomorphic process. An increasing number of studies show that changes in the rate of erosion and sediment yield in watershed are often strongly correlated with seismicity. Peak ground acceleration is equal to the maximum acceleration of the ground that occurred during the shaking of the earthquake in a place. Therefore, the purpose of this study is to investigate the relationship between soil erosion and seismic activities and erosion control factors, how high erosion zones are related to the Peak ground acceleration and other factors, and to compare the efficiency of two fuzzy logic and entropy models in erosion zoning.

Talar drainage basin is located on both sides of Qaimshahr-Tehran axis. In terms of geographic coordinates, it is located between ˚52 '35 "22 to˚53 '23 "34 east longitude and ˚35 '44 "23 to ˚36 '19 "1 north latitude. For this purpose, important indices such as erosivity, erodibility, slope-length, cover management have been calculated in the RUSLE equation. Then, the zoning map of the peak ground acceleration has been used as the erosion control factor using the seismic hazard analysis method obtained in the study (Ashtari et al., 2023).

Fuzzy logic and entropy:

For the purpose of zoning, all the layers produced in the previous step were placed as base layers in fuzzy logic and entropy models. Linear function was used to quantify all layers in fuzzy logic. In order to combine the layers, the fuzzy gamma operator was used. In the entropy model, based on the map of erosion levels from specialized studies of the Talar drainage basin, the number of sample sites in each of the erosion zones is as follows: The areas of surface erosion are (16), lateral stream erosion (10), groove erosion (9), gully erosion (8), rock and rock mass erosion (6) and badland erosion (1). The steps of the entropy method are as follows: forming the decision matrix, Ej function and determining the value of entropy, calculating the degree of deviation of criteria dj, determining the weight of each criterion Wij, regional erosion risk model Hi.

Rainfall erosivity factor: This factor expresses the kinetic energy of rain when the drops hit the soil particles, and in other words, it is the intensity of precipitation and erosion resulting from the impact of rain drops on the ground during rainfall. The highest and lowest amount of this factor is respectively in the northern side of the sub-basin 3 with (369-457-8) and in the southern side of the sub-basin 1 with (13.9-102).Erodibility factor: It is in accordance with the average sensitivity factor of formations to erosion of sub-basin 1 (6.09), sub-basin 2 (6.39), sub-basin 3 (7.6). Due to the extent and variety of geological formations, there is a high potential for erosion in all 3sub-basins.length - slope: which is also known as the topography factor, is a function of the height in the basin. The lowest amount of this factor is in the areas of valleys and river streams (>6) and the highest amount corresponds to slopes and heights (>23).cover management: Vegetation factor represents the amount of vegetation waste in different uses. The lower the amount of c, the more vegetation in the area. The highest coefficient c (0.5-0.6) is in sub-basin 1, which indicates the decrease of vegetationand the potential for erosion, and the lowest c (0.2-0.3) in sub-basin 3, which indicates the richness of land uses from forest vegetation.Peak ground acceleration factor: The highest amount of ground acceleration is near the active faults in the region. Due to the structure of the faults, which have an east-west trend, in all 3 sub-basins there are areas of maximum ground acceleration during each event. The highest areas of ground acceleration levels (0.5-0.6) are sub-basin 1 and the lowest are sub-basin 3, which is due to the structure of active faults such as Firuzukuh fault, IRQ112 and IRQ357.

The highest area is in the very low class (839 km2 and 39.9 %) and the lowest area is in the low class (162.9 km2 and 7.7 %). The largest expansion of the mentioned areas is located in sub-basin 1. The distribution of high and very high-risk areas especially in sub-basin 2 and the low coverage of these areas in sub-basin 1 do not match with the relative contribution of high sediment production in sub-basin 1 based on studies.Erosion zonation map by entropy methodThe results of the entropy model showed that the erodibility factor was 20.62%, the peak ground acceleration factor was 20.20%, the cover management was 20%, the erosivity factor was 19.60%, and the slope-length factor was 19.58% in the occurrence of erosion in the region. The area of very high risk with 13.5% of the area of the basin includes erosional facies containing surface, groove, gully and lateral stream. In general, it can be said that the highest erosion risk zones are not related to the highest weighting of the layers, but the placement of a set of factors influencing the occurrence of erosion, which shows the highest amount of erosion.

The peak ground acceleration has a direct impact on the control of sediment yield and erosion processes. The placement of a small part of the IRQ112 fault has caused the extent of ground acceleration levels (0.5-0.6) to be less in sub-basin 3 and more in sub-basin 1 due to the placement of 3 faults, Firuzukuh, IRQ112 and IRQ357. The placement of erodible formations in the ranges of peak ground acceleration has caused the process of sediment production and erosion to accelerate. The coordination of the erosion occurrence zones with the ground acceleration levels is due to the regionality of the entropy model in contrast to the fuzzy model.