Space monitoring of the nesting areas of locust species in Kazakhstan since 2000

Abstract

Two approaches to the locust nesting conditions modelling using remotely sensed data in Kazakhstan are described and discussed in the paper. The first approach is focused on the estimation of land cover classes dynamics, that is determined with use of unsupervised classification of low to moderate spatial resolution satellite data. The second method is a logic development of the first one and based on the concept of the species' ecological niche. The base model of the locust nesting conditions was developed with abiotic variables taken from BioClim and WorldClim datasets and a series of field registration of egg-clutches and young nymphs of the locust. Further elaboration of the existing ecological niche model implies the inclusion of additional dynamic variables in the analysis, such as: the reeds area for the current year; the water bodies area; and the floods presence or absence in the spring season. Dynamic variables are calculated using actual remotely sensed moderate spatial resolution information. The combination of two methods allows obtaining the reliable information on the abundance of locust. Further development of the described techniques is aimed at creating predictive model for potential locust outbreaks and its application for the rational pest control measures management. The use of two methods in different years allowed obtaining the results of Asian locust outbreaks monitoring according to remote sensed data for the period 2000-2017.