Monitoring and investigation of wind erosion warning areas in the northwest of Golestan Province (Case study: Aq Qala and Gomishan Cities)

Introduction In recent years, one of the most significant environmental crises has been the phenomenon of wind erosion and dust emission. Wind erosion is considered one of the effective factors in desertification and land degradation in dry regions. The process of wind erosion, due to the transport of soil nutrients along with fine particles, is recognized as one of the limiting factors for soil fertility in many parts of the world. Evaluating this type of erosion and estimating the soil loss requires the installation of various measuring stations. Establishing and equipping these stations and providing the necessary equipment require large costs and a long time. In addition to direct measurement methods in wind erosion, the use of modeling results, especially in combination with remote sensing techniques, to study and predict environmental effects, trends, and risk assessment has greatly contributed to research in the last two decades. Therefore, this study aims to monitor and investigate the warning zones of wind erosion in northwest Golestan Province using the wind erosion hazard index (WEHI) model.                                        Materials and Methods To assess wind erosion in the western part of Golestan Province, the WEHI model was implemented. This model predicts the sensitivity of the landscape to wind erosion by considering a set of surface and climatic thresholds and using a geographic information system (GIS). In this model, wind erosion severity is determined in three classes: low, moderate, and severe, by multiplying wind speed by the percentage of bare soil and dividing it by the percentage of soil moisture percentage. To monitor wind erosion, the normalized difference moisture index (NDMI) was used to evaluate soil moisture, and the modified bare soil index (MBI) was used to assess bare soil. For validation, these indices were compared to field data and plots in each working unit. Additionally, three statistical parameters, Pearson correlation coefficient, coefficient of determination (R2), and root mean square error (RMSE), were employed to calculate the correlation between these indices and ground data. Furthermore, a Markov chain model was used to examine changes in wind erosion classes. Finally, after monitoring wind erosion and considering a threshold value for this model, the area of warning zones during the statistical period was investigated.   Results and Discussion The results obtained from the WEHI model indicate severe wind erosion intensity in the working units of Atark floodplain deposits, saline lands, longitudinal dunes, Barchan dunes, bare lands, and margins area of wetlands. According to the WEHI model, the region was divided into three classes: low (54% frequency), moderate (21% frequency), and severe (25% frequency). The model classified the northern regions into severe and moderate classes, while the southern areas of the region fell into the low wind erosion class. High correlation coefficients between the WEHI model indices based on remote sensing and field data demonstrate the model's ability to monitor wind erosion over time and at different scales. Wind erosion monitoring results showed that the high wind erosion class increased from 59,940.88 ha to 71,698.3 ha, indicating an increase of 11,757.43 ha. Spatial analysis of wind erosion classes indicated that most changes occurred in central areas, with most areas around the Sangartappeh playa and central regions changed to severe wind erosion class, while western, eastern, and central areas changed to the low wind erosion class. Finally, monitoring the warning zones revealed an increase of 41,000 hectares in the areas under warning, in the western, northwestern, and central regions.   Conclusion In this study, the performance of the WEHI model in assessing wind erosion risk in the western part of Golestan Province was confirmed. Although factors such as roughness, soil structure, and organic matter content are not directly considered in the model, they are indirectly incorporated in scoring the percentage of bare soil. To control wind erosion in these areas, planting salt-tolerant species and implementing soil fertility enhancement strategies, such as soil mulching in heavy-textured areas, are recommended. Finally, mechanical operations and the establishment of windbreak networks are suggested for controlling wind erosion in abandoned land units. This research can serve as a useful approach for planning and managing vulnerable areas to wind erosion in northwest Golestan Province.