Photosynthetic traits are slow to measure by hand. In a new European Journal of Agronomy paper, we link UAV multispectral imagery to leaf physiology with a physics-informed machine learning framework — predicting CO₂ assimilation, stomatal conductance, photosystem II efficiency, biomass, and grain yield across winter wheat varieties.

On 10 June 2026, researchers from the University of Georgia (Athens, USA) and TUM met at the Weihenstephan campus for a full-day joint seminar — from variable-rate management and precision irrigation in Georgia to UAV and Satellite crop sensing and digital nutrient management at TUM.

Hyperspectral sensors see crop nitrogen in detail but are expensive; multispectral drones are cheap but spectrally coarse. Our new M2H-SWIR framework bridges the gap, reconstructing continuous 400–2500 nm reflectance from a 10-band UAV camera and improving nitrogen estimation in winter wheat.

TUM alumnus Prof. Christian Bredemeier (UFRGS, Brazil) returned to Weihenstephan on 13 May 2026 to give a guest lecture — “Research in precision and digital agriculture in the context of the Brazilian agribusiness” — followed by a visit to the Precision Agriculture Lab at Dürnast.

Our TUM Precision Agriculture Lab contributed three talks to the “Hyperspectral remote sensing and AI in Agriculture” workshop at the University of Luxembourg workshop (4–5 Feb 2026), covering trait-based crop monitoring, UAV SIF for NUE, and uncertainty from atmospheric correction.