Use of Machine Learning in Evaluation of DroughtPerception in Irrigated Agriculture: The Case of anIrrigated Perimeter in Brazil

• Autor(a) principal: Xavier, Louise Caroline Peixoto
• Data de Publicação: 2020
• Outros Autores: Silva, Samiria Maria Oliveira da, Carvalho, Taís Maria Nunes, Pontes Filho, João Dehon de Araújo, Souza Filho, Francisco de Assis de
• Tipo de documento: Artigo
• Idioma: por
• Título da fonte: Repositório Institucional da Universidade Federal do Ceará (UFC)
• Texto Completo: http://www.repositorio.ufc.br/handle/riufc/58238
• topic: Irrigated agriculture; Standardized Precipitation Index; Machine learning; Random Forest;Decision Tree; Drought perception; Water resource management

This study aimed to understand the perception of drought among farmers, in order to supportdecision-making in the water allocation process. This study was carried out in theTabuleiro de Russasirrigated perimeter, in northeast Brazil, over the drought period of 2012–2018. Two analyses wereconducted: (i) drought characterization, using the Standardized Precipitation Index (SPI) based ondrought duration and frequency criteria; and (ii) analysis of farmers’ perceptions of drought viaselection of explanatory variables using the Random Forest (RF) and the Decision Tree (DT) methods.The 2012–2018 drought period was defined as a meteorological phenomenon by local farmers;however, an SPI evaluation indicated that the drought was of a hydrological nature. According tothe RF analysis, four of the nine study variables were more statistically important than the others ininfluencing farmers’ perception of drought: number of cultivated land plots, farmer’s age, years ofexperience in the agriculture sector, and education level. These results were confirmed using DTanalysis. Understanding the relationship between these variables and farmers’ perception of droughtcould aid in the development of an adaptation strategy to water deficit scenarios. Farmers’ perceptioncan be beneficial in reducing conflicts, adopting proactive management practices, and developing aholistic and efficient early warning drought system.