Phenotyping of productivity and resilience in sweetpotato under water stress through UAV-based multispectral and thermal imagery in Mozambique

Sweetpotato is a crucial crop to guarantee food security in sub-Saharan Africa, and drought events are considered one of the most critical factors affecting sweetpotato productivity in this region. In this study, airborne imagery based on reflectance (NDVI, CI_red-edge) and canopy temperature minus air temperature (dT) indices was used to characterize sweetpotato genotypes under drought treatments in Mozambique. Two field experiments established in rainy/hot (Trial A) and dry/cool (Trial B) seasons were assessed. In Trial A, 24 genotypes were subjected to early- (ESD), mid- (MSD) and late-season (LSD) drought stress treatments and compared against a control. In Trial B, 120 genotypes were subjected to LSD only. The percentage of reduction in vine weight (PR_VW) under drought was related primarily to temporal variation of NDVI and CI, regardless of drought treatment and seasons. dT in relation to control (dT_Amp) was associated with PR_VW in ESD-Trial A and LSD-Trial B, whereas under LSD-Trial A, dT_Amp was related to total fresh storage root weight (TRW). During the rainy/hot season, higher TRW reduction was promoted under ESD; however, under LSD, it was possible to identify productive genotypes able to withstand drought stress, highlighting their relevance for drought-tolerance selection purposes.