MULTIVARIATE ANALYSIS OF WEATHER FACTORS ASSO CIATED WITH BLACK THRIPS, THRIPS PARVISPINUS IN FESTATION IN CHILLI IN TELANGANA

Abstract

Thrips, in particular the invasive black thrips, Thrips parvisspinus (Karny) have become a serious pest of chilli (Capsicum annuum L.) in Telangana, India. Understanding the relationship between weather variables and thrips population is vital in developing effective integrated pest management (IPM) strategies. A four-year field experi ment (2022-2025) was conducted at the Vegetable Research Station, Sri Konda Laxman Telangana Horticultural University, Mulugu, Siddipet District, Telangana, in Kharif seasons (SMW 32-52 to SMW 1-5) to investigate the influence of nine weather variables (Temp_Max, Temp_Min, RH_I, RH_II, Rainfall, Rainy Days, Sunshine, Wind speed, Evaporation) on thrips populations on leaves and flowers. Stepwise multiple regression, Pearson correla tion, Principal Component Analysis (PCA) and K-Means clustering analyses were used to derive this infor mation. The pooled regression model for flower thrips had an excellent predictive capability (R² = 0.951, adjusted R² = 0.945), where Temp_Min (β = -1.29, p<0.001), Evaporation (β = -6.59, p<0.001) and Rainfall (β = -0.064, p<0.01) were significant negative contributors. Correlation analysis revealed Temp_Min as the strongest nega tive correlation with flower thrips (r = -0.90, p<0.001). PCA revealed that the first two principal components (PCs) accounted for 82.23% of the variation in the pooled data, with PC1 (Wet/Humidity axis: 59.12%) and PC2 (Thermal/Evaporation axis: 23.11%) being the most significant. K-Means analysis of the pooled data revealed the most distinct separation (k=2, silhouette=0.43) into "Wet-humid" and "Dry-sunny" weather groups. The pre dictive models developed here can be used as effective tools for weather-based early warning systems to pre dict thrips outbreaks in chilli.