Assessing the physical background of spatial variation in biomass production of corn: A case study for the Hupselse Beek catchment

Organised by Laboratory of Geo-information Science and Remote Sensing

Thu 25 October 2018 10:00

Venue Gaia, gebouwnummer 101
Room 1

By Quanxing Wan (China)

Satellite remote sensing of normalized difference vegetation index (NDVI) provides a method for characterizing spatial patterns of crop biomass production. One of the objectives of this research was to estimate corn biomass production in the Hupsel Brook Catchment from remotely sensed NDVI maps. By means of establishing a linear relationship between cumulative NDVI in the whole growing season of corn and corn biomass production, we roughly estimated the biomass production of all corn parcels in the catchment in 2017. The variation of the predicted production (dry weight) ranges from 14.18 ton/ha to 24.85 ton/ha. Inhomogeneous distribution of estimated values was observed in most of corn parcels. Two parcels close to the center of the catchment were witnessed relatively low production than the rest. The estimation accuracy would be significantly improved with higher temporal resolution of NDVI time series and available measured biomass data in fields. However, it is extremely hard to build up NDVI time series out of only Sentinel-2 database, judged from the nonnegligible cloud effects in a large number of ready-to-download images. Data fusion technique can be applied in future studies to increase the estimated accuracy. As for the correlation between corn biomass production and physical influencing factors, no significant correlation was derived between surface elevation and production. As a whole, weak relationship also existed in the relationships between soil physical units, groundwater depth and biomass production. Moderate variation could still be found in between specific classes/units of physical backgrounds. Lastly, the maximum NDVI value appeared approximately 60 days before the cumulative ETref (reference crop evapotranspiration) was super closed to its yearly maximum in late autumn. Evapotranspiration was negligible in winter. Variation of NDVI is positively correlated to crop biomass. Thus, we predicted that the maximum of biomass was also reached around two months earlier. The gap in between must be explained by other influencing factors which were not taken into consideration in this study, such as agrotechnical measures, irrigation and drainage efficiency, solar radiation and so forth.

Key words: NDVI; corn; biomass production; Hupsel Brook; precision agriculture.