Integrated Modeling of Soil Erosion by Water and Wind Using Machine Learning Methods

Document Type : Research

Authors

1 Ph.D. Student, Department of Natural Resources Engineering, Faculty of Agriculture and Natural Resource Engineering, University of Hormozgan, Bandar-Abbas, Iran

2 Associate Professor, Department of Natural Resources Engineering, Faculty of Agriculture and Natural Resource Engineering, University of Hormozgan, Bandar-Abbas, Iran

3 Assistant Professor, Department of Natural Resources Engineering, Faculty of Agriculture and Natural Resource Engineering, University of Hormozgan, Bandar-Abbas, Iran

4 Associate Professor, Department of Natural Resources Engineering, University of Kashan, Kashan, Iran

Abstract

Introduction and objective
Soil erosion by water and wind is one of the biggest environmental threats worldwide which it has various negative consequences such as soil degradation, mitigation of soil fertility, depletion of nutrients and micro-elements, degradation of soil structure, dust storms, reservoir siltation, and etc. on the natural and anthropogenic ecosystems. Therefore, accurate information of land susceptibility to soil erosion hazard by water and wind, and production of spatial maps of these hazards are necessary for mitigation of their consequences. Therefore, the goal of this research is application of two machine learning models for the spatial modelling of water and wind erosion in the Bakhtegan basin, Fars province.
Materials and methods
In order to generate the spatial maps of soil erosion by water and wind, we have used 20 and 16 effective factors on water and wind erosion, respectively. Before the modelling stage, the multivariate adaptive regression spline (MARS) feature selection algorithm and multicollinearity test were used to identify the most important factors controlling water and wind erosion. In next stage, two random forest and support vector learning machine techniques were applied to model soil erosion. Area under curve (AUC) was applied to assess the model performance.
Result and discussion
According to results, land use, slope, lithology, roughness, coarse fragment, clay, vegetation cover, aspect, soil bulk density and topographic wetness index were identified as the most important factors controlling water erosion. The most important factors controlling wind erosion are including elevation, land use, wind speed, slope, vegetation cover, soil nitrogen, cation exchange capacity, lithology, aspect, clay and bulk density. The AUC value for wind erosion map generated by random forest model was 99, and for water erosion map produced by support vector machine is 96. Finally, we integrated two maps and results indicate an area of study area about 18% has very high susceptibility to wind erosion and low susceptibility to water erosion.
Conclusion and suggestions
Due to study area has several lakes consisting of Bakhtegan, Maharlo and Tashk, and also several dams, it is so important to supply water for different purposes such as drinking, agricultural, industry and environment. In case of not paying attention to this challenge especially wind erosion in the study area, this case study especially dried beds for lakes can play as sources for generating dust particles. Therefore, the findings of this research can use by managers to mitigate negative effects of this phenomena.

Keywords

References

Anderson TW, Anderson TW, Anderson TW, Anderson TW. 1958. An introduction to multivariate statistical analysis. New York: Wiley. pp. 1468-1482.
Bordbar M, Aghamohammadi H, Pourghasemi HR, Azizi Z. 2022. Multi-hazard spatial modeling via ensembles of machine learning and meta-heuristic techniques. Scientific Reports. 12(1): 1451.
Breiman L. 2001. Random forests. Machine learning. 45: 5-32.
Boroughani M, Pourhashemi S, Gholami H, Kaskaoutis DG. 2021. Predicting of dust storm source by combining remote sensing, statistic-based predictive models and game theory in the Sistan Watershed, southwestern Asia. Journal of Arid Land. 13(11): 1103-1121.
Carrara A, Cardinali M, Guzzetti F, Reichenbach P. 1995. GIS technology in mapping landslide hazard. In geographical information systems in assessing natural hazards, Dordrecht: Springer Netherlands. pp. 135-175.
Mirhashemi SH, Haghighat Jou P, Mirzaei F, Panahi M. 2020. The study of environmental and human factors affecting aquifer depth changes using tree algorithm. International Journal of Environmental Science and Technology. 17: 1825-1834.
Chen W, Pourghasemi HR, Naghibi SA. 2018. Prioritization of landslide conditioning factors and its spatial modeling in Shangnan County, China using GIS-based data mining algorithms. Bulletin of Engineering Geology and the Environment. 77: 611-629.
Chou SM, Lee TS, Shao YE, Chen IF. 2004. Mining the breast cancer pattern using artificial neural networks and multivariate adaptive regression splines. Expert Systems With Applications. 27(1): 133-142.
Constantin M, Bednarik M, Jurchescu MC, Vlaicu M. 2011. Landslide susceptibility assessment using the bivariate statistical analysis and the index of entropy in the Sibiciu Basin (Romania). Environmental Earth Sciences. 63: 397-406.
Dehbozorgi M, Jafari M, Malekian A, Zehtabian G, Fallah Shamsi SR. 2020. Assessment the instability and land degradation using fuzzy methods, hierarchical analysis and weighted linear combination (Case study: Bakhtegan Watershed, Fars). Journal of Range and Watershed Management. 73(2): 321-335.
Ferraresi M. 1989. The regionalization of fluvial sediment yield in Emilia-Romagna (northern Italy). IAHS-AISH publication. 191: 253-260.
Gholami H, Mohamadifar A, Collins AL. (2020). Spatial mapping of the provenance of storm dust: Application of data mining and ensemble modelling. Atmospheric Research. 233: 104716.
Gholami H, Mohammadifar A, Golzari S, Kaskaoutis DG, Collins AL. 2021. Using the Boruta algorithm and deep learning models for mapping land susceptibility to atmospheric dust emissions in Iran. Aeolian Research. 50: 100682.
Gholami H, Mohammadifar A, Fitzsimmons KE, Li Y, Kaskaoutis DG. 2023. Modeling land susceptibility to wind erosion hazards using LASSO regression and graph convolutional networks. Frontiers in Environmental Science. 11: 1187658.
Gholami H, Mohammadifar A. 2022. Novel deep learning hybrid models (CNN-GRU and DLDL-RF) for the susceptibility classification of dust sources in the Middle East: A global source. Scientific Reports. 12(1): 19342.
Guzzetti F, Carrara A, Cardinali M, Reichenbach P. 1999. Landslide hazard evaluation: A review of current techniques and their application in a multi-scale study, Central Italy. Geomorphology. 31(1-4): 181-216.
Kanungo DP, Arora MK, Sarkar S, Gupta RP. 2012. Landslide susceptibility zonation (LSZ) mapping–a review.
Kursa MB, Rudnicki WR. 2010. Feature selection with the Boruta package. Journal of Statistical Software. 36: 1-13.
Lucà F, Conforti M, Robustelli G. 2011. Comparison of GIS-based gullying susceptibility mapping using bivariate and multivariate statistics: Northern Calabria, South Italy. Geomorphology. 134(3-4): 297-308.
Mina M, Rezaei M, Sameni A, Ostovari Y, Ritsema C. 2022. Predicting wind erosion rate using portable wind tunnel combined with machine learning algorithms in calcareous soils, southern Iran. Journal of Environmental Management. 304: 114171.
Mohammadifar A, Gholami H, Comino JR, Collins AL. 2021. Assessment of the interpretability of data mining for the spatial modelling of water erosion using game theory. Catena. 200: 105178.
Nasrnia F, Zibaei M. 2017. Determination farmers' vulnerability patterns to drought in Iran: case study of Bakhtegan Basin. Agricultural Economics Research. 9(34): 1-37.
Nicodemus KK. 2011. On the stability and ranking of predictors from random forest variable importance measures. Briefings in Bioinformatics. 12(4): 369-373.
Pournader M, Ahmadi H, Feiznia S, Karimi H, Peirovan HR. 2018. Spatial prediction of soil erosion susceptibility: an evaluation of the maximum entropy model. Earth Science Informatics. 11: 389-401.
Rahmati O, Pourghasemi HR, Melesse AM. 2016. Application of GIS-based data driven random forest and maximum entropy models for groundwater potential mapping: a case study at Mehran Region, Iran. Catena. 137: 360-372.
Rezaei M, Mohammadifar A, Gholami H, Mina M, Riksen MJ, Ritsema C. 2023. Mapping of the wind erodible fraction of soil by bidirectional gated recurrent unit (BiGRU) and bidirectional recurrent neural network (BiRNN) deep learning models. Catena. 223: 106953.
Safaei M, Omar H, Yousof ZB, Ghiasi V. 2010. Applying geospatial technology to landslide susceptibility assessment. Electronic Journal of Geotechnical Engineering. 15(G): 677-696.
Senanayake S, Pradhan B. 2022. Predicting soil erosion susceptibility associated with climate change scenarios in the Central Highlands of Sri Lanka. Journal of Environmental Management. 308: 114589.
Talukdar S, Naikoo MW, Mallick J, Praveen B, Sharma P, Islam AR. MT., ... Rahman A. 2022. Coupling geographic information system integrated fuzzy logic-analytical hierarchy process with global and machine learning based sensitivity analysis for agricultural suitability mapping. Agricultural Systems. 196: 103343.
Vafakhah M, Rajabi M. 2005. Efficiency of meteorological drought Indices for monitoring and assessment of drought in Bakhtegan, Tashk, and Maharlo Lakes Watershed.
Van Westen CJ, Van Asch TW, Soeters R. 2006. Landslide hazard and risk zonation—why is it still so difficult?. Bulletin of Engineering Geology and the Environment. 65: 167-184.
Yao X, Tham LG, Dai FC. 2008. Landslide susceptibility mapping based on support vector machine: A case study on natural slopes of Hong Kong, China. Geomorphology. 101(4): 572-582.
Yesilnacar EK. 2005) The application of computational intelligence to landslide susceptibility mapping in Turkey. University of Melbourne, Department. 200 p.
Watershed Management Research
Volume 36, Issue 3 - Serial Number 140
September 2023
Pages 128-145

Files

History

  • Receive Date: 11 April 2022
  • Revise Date: 16 August 2022
  • Accept Date: 21 September 2022

Share

How to cite

Statistics