Evaluation of crop model performance in simulating biophysical indexes based on proximal sensing and statistical data

Crop science

Enhancing the robustness of crop modeling simulation in prediction of biophysical variables (phenology, leaf area growth, biomass, yield etc…) is important for investigating impacts of climate change as well as for informing social-economic models in cost-benefit analysis of ecosystem services at regional scales. Scaling up the modeling application will be challenged due to the high spatio-temporal heterogeneity of soil characteristics and climatic conditions due to lack of experimental ground-truth data. Proximal sensing and statistical data could provide good alternatives to support modeling validation.

The aim of the study is to investigate the crop model capability in simulating biophysical variables through combining Sentinel-2 products and statistical records of growth data for the important crops in Brandenburg. Specifical aims are to (i) extract and investigate the existing Sentinel-2 final products (i.e. LAI) and statistical crop growth variables (i.e. phenology and yield) for targeted crops and locations (ii) to compare the simulated outputs of crop model with the extracted data products (iii) to evaluate the modeling performance across locations and growing seasons in Brandenburg.

Type of study: simulation, data analysis, modeling uncertainty, R, QGIS. Students will have an opportunity to work with crop model outputs from SIMPLACE, remote sensing (Sentinel-2 products) and statistical data (yield and phenology). Students are expected to perform statistical analysis and to familiarize with geographical presentations and analysis.

03/2026 – 09/2026

English, German 

Basis, field, crop modeling application, statistical analysis, Excel, R, QGIS

Dr. Thuy Nguyen

tngu@uni-bonn.de