Predicción de la erosión del suelo mediante random forest: caso de estudio cuenca río grande, Antioquia

Contextualization: Currently, the knowledge of natural phenomena associated with the preservation of the systems is of interest both for researchers in the natural sciences, and for the environmental authorities in charge of decision-making on resource management. In this sense, work has been carried out on the interpretation and prediction of different physical phenomena such as erosion, to create scenarios that allow strengthening the response criteria for the conservation of the natural capital of the soil.

Knowledge gap: The ability to predict the phenomenon of erosion is limited on many occasions due to the quantity and variability of the parameters and variables that are related to erosion; besides that, in many cases, a high computational processing is required to achieve that they are associated with each other.

Purpose: The aim is to implement a machine learning model as an alternative tool for complex modeling and erosion prediction.

Methodology: In this study, a model is developed from the training of the non-parametric Random Forest method through supervised learning, to predict erosion occurrences in the Rio Grande basin, considering the variables that have previously been used in other methods to model this phenomenon.

Results and conclusions: The results showed a capacity to predict erosion in the basin with an approximate precision of 77%, so this method can be applied to obtain fast and reliable predictions. In addition, it was found that the variables used in the RUSLE model mainly explain the occurrence or not of erosion. The great importance of the temperature variable introduced in the model is also surprising.

keywords: Erosion, Random Forest, non-parametric classification models, supervised learning

Barral, M. P. (2016). Tutorial para el mapeo de funciones ecosistémicas y servicios ecosistémicos con protocolo ECOSER (Vol. 1). Unidad Integrada Balcarce (EEA INTA Balcarce – Facultad de Ciencias Agrarias, UniversidadNacional de Mar del Plata

Barrera, J. E., Rivera, J. H., y Cadena, M. E. (2013). Influencia del sistema radical de cuatro especies vegetales en la estabilidad de laderas a movimientos masales. Cenicafé, 64(2), 59-77. http://biblioteca.cenicafe.org/bitstream/10778/531/1/arc064%2802%2959-77.pdf

Burkhard, B., Kandziora, M., Hou, Y., & Müller, F. (2014). Ecosystem service potentials, flows and demands-concepts for spatial localisation, indication and quantification. Landscape Online, 34(1), 1-32. https://doi.org/10.3097/LO.201434

Cárdenas, J. A. (2019). Clasificación de aceptación de campañas para una entidad financiera, usando random forest con datos balanceados y datos no balanceados. [Tesis de Maestría]. Universidad Ricardo Palma Escuela de Posgrado.

Corantioquia y UNAL. (2015). Actualización y ajuste Plan de Ordenación y Manejo de la Cuenca de los ríos Grande y Chico. Contrato 967 de 2013.

Criminisi, A., & Shotton, J. (2013). Decision forests for computer vision and medical image analysis. Springer Science & Business Media. https://doi.org/10.1007/978-1-4471-4929-3

Duan, X., Chen, Y., Wang, L., Zheng, G., & Liang, T. (2023). The impact of land use and land cover changes on the landscape pattern and ecosystem service value in Sanjiangyuan region of the Qinghai-Tibet Plateau. Journal of Environmental Management, 325(PB), 116539. https://doi.org/10.1016/j.jenvman.2022.116539

García-Leoz, V., Villegas, J. C., Suescún, D., Flórez, C. P., Merino-Martín, L., Betancur, T., & León, J. D. (2018). Land cover effects on water balance partitioning in the Colombian Andes: improved water availability in early stages of natural vegetation recovery. Regional Environmental Change, 18(4), 1117-1129. https://doi.org/10.1007/s10113-017-1249-7

Genuer, & Poggi, J. (2020). Random Forests with R. In Use R. https://doi.org/10.1007/978-3-030-56485-8

Gómes, N., & Carmona, M. (2022). An application of Machine learning Techniques to the Prediction of Purchase in the Tourism Sector. EasyChair.

Hao, J., & Ho, T. K. (2019). Machine Learning Made Easy: A Review of Scikit-learn Package in Python Programming Language. Journal of Educational and Behavioral Statistics, 44(3), 348-361. https://doi.org/10.3102/1076998619832248

Le Clec’h, S., Oszwald, J., Decaens, T., Desjardins, T., Dufour, S., Grimaldi, M., Jegou, N., & Lavelle, P. (2016). Mapping multiple ecosystem services indicators : Toward an objective-oriented approach. Ecological Indicators, 69, 508-521. https://doi.org/10.1016/j.ecolind.2016.05.021

Louppe, G. (2014). Understanding Random Forests: From Theory to Practice. July. http://arxiv.org/abs/1407.7502

Machado, J. (2018). Impacto Potencial de pérdida del servicio ecosistémico intermedio de control de erosión por cambios en el capital natural del suelo. Caso de estudio: Cuenca de Riogrande, Departamento de Antioquia [Universidad Nacional de Colombia]. https://doi.org/10.1016/j.ecolind.2017.07.051

Machado, J., Villegas-Palacio, C., Loaiza, J. C., & Castañeda, D. A. (2019). Soil natural capital vulnerability to environmental change. A regional scale approach for tropical soils in the Colombian Andes. Ecological Indicators, 96(May 2018), 116-126. https://doi.org/10.1016/j.ecolind.2018.08.060

Martín-López, B., González, J., & Vilardy, S. (2012). Guía Docente Ciencias de la Sostenibilidad. CO-BAC. https://doi.org/10.1016/j.ajhg.2011.11.018

Martínez-Harms, M. J., & Balvanera, P. (2012). Methods for mapping ecosystem service supply : a review. International Journal of Biodiversity Science, Ecosystem Services & Management, 8(1-2), 17-25. https://doi.org/10.1080/21513732.2012.663792

Medina-Merino, R. F., y Ñique-Chacón, C. I. (2017). Bosques aleatorios como extensión de los árboles de clasificación con los programas R y Python. Interfases, 0(010), 165. https://doi.org/10.26439/interfases2017.n10.1775

Mohammed, S., Alsafadi, K., Talukdar, S., Kiwan, S., Hennawi, S., Alshihabi, O., Sharaf, M., & Harsanyie, E. (2020). Estimation of soil erosion risk in southern part of Syria by using RUSLE integrating geo informatics approach. Remote Sensing Applications: Society and Environment, 20(2019), 100375. https://doi.org/10.1016/j.rsase.2020.100375

Probst, P., Wright, M. N., & Boulesteix, A. L. (2019). Hyperparameters and tuning strategies for random forest. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 9(3). https://doi.org/10.1002/widm.1301

Ramírez, C. D., Orrego, S. A., & Schneider, L. C. (2018). Identifying Drivers and Spatial Patterns of Deforestation in the Río Grande Basin, Colombia. Journal of Latin American Geography, 17(1), 108-138. https://doi.org/10.1353/lag.2018.0005

Sepúlveda, L. (2013). Modelo para la definición de áreas estratégicas para la conservación de suelos a partir de la determinación de la susceptibilidad a la erosión hídrica. Universidad de Antioquia.

Suescún, D., León, J. D., Villegas, J. C., García-Leoz, V., Correa-Londoño, G. A., & Flórez, C. P. (2018). ENSO and rainfall seasonality affect nutrient exchange in tropical mountain forests. Ecohydrology, e2056, 1-10. https://doi.org/10.1002/eco.2056

Suescún, D., Villegas, J. C., León, J. D., Flórez, C. P., García-Leoz, V., & Correa-Londoño, G. A. (2017). Vegetation cover and rainfall seasonality impact nutrient loss via runoff and erosion in the Colombian Andes. Regional Environmental Change, 17(3), 827-839. https://doi.org/10.1007/s10113-016-1071-7

UPRA, U. de P. R. A.-. (2021). Evaluaciones Agropecuarias Municipales – EVA. Módulo de Consulta de Información. https://www.upra.gov.co/web/guest/consulta-de-informacion

Wang, P., Li, R., Liu, D., & Wu, Y. (2022). Dynamic characteristics and responses of ecosystem services under land use/land cover change scenarios in the Huangshui River Basin, China. Ecological Indicators, 144(May), 109539. https://doi.org/10.1016/j.ecolind.2022.109539

Wischmeier, W. H., & Smith, D. D. (1978). Predicting rainfall erosion losses: a guide to conservation planning. Department of Agriculture, Science and Education Administration.

license

How to Cite

Arango Carvajal, L. I. (2023). Prediction of soil erosion by random forest: case study of the rio grande basin, Antioquia. Revista De Investigación Agraria Y Ambiental, 15(1), 317-339. https://doi.org/10.22490/21456453.6755

Download Citation

Almétricas

Metrics

File downloads