Poster

         Ecophysiology

Common bean yield prediction using remote sensing

Presenting Author
Manuel Guzman
Description
Common bean (Phaseolus vulgaris L.) yield modeling using remote sensing is an emerging method in Colombia´s production regions where landraces are used. In general, these landraces are climbing plants with dissimilarities in phenological phases such as flowering and grain maturity. This condition is challenging to predict due to vertical biomass distribution, heterogeneous production and soil chemical conditions. The Normalized Difference Vegetation Index (NDVI) is useful to evaluate plant biomass and nutrient content in different growth stages. A field experiment was conducted with eight common bean landraces and climbing type and three Phosphorus restriction levels (25, 50, and 75% of P-relative to the control), according to the recommended dose. NDVI was measured using a GreenSeeker® canopy sensor, with a two-week frequency, from the V4 up to the R8 growth stages. P-restriction were laid out in a randomized complete block design with a split-plot arrangement and three replications, where P-doses were the main plots and landraces subplots. Biomass production was higher at the 50% P-restriction level. For the NDVI, an interaction between the P-restriction and days to flowering was observed. NDVI values increased from 0.92 at 25% restriction to 0.79 at 75% P-restriction. The common bean yield was linear and positively correlated with NDVI values (P < 0.05). Our results indicated that the NDVI can be used for estimating biomass production and P uptake in the early growth stages of common bean crops.