University of Twente Student Theses


Crop stress during the 2018 drought study area: North Brabant, The Netherlands

Uwumukiza, Ake Josiane (2021) Crop stress during the 2018 drought study area: North Brabant, The Netherlands.

Link to full-text:
(only accessible for UT students and staff)
Abstract:Water shortage is a severe environmental constraint to plant productivity and food security worldwide. Due to the severity and duration of drought, it can exceed all other causes of crop yield reduction. Crop growth and yield production are negatively affected by deficient water supply and abnormal temperature due to physical damages and biochemical changes. For better management, it is vital to understand the physiological, ecological, and biological processes related to drought stress. Drought stresses reduce leaf area, leaf water content, root proliferation, and CO2 assimilation by leaves due to the stomatal closure. This study aims to develop a method to monitor drought stress on potatoes and differentiate this from other effects such as diseases, wildfires, lack of fertilizers, and floods by using multiple indices and considering physical processes related to drought. Satellite data combined with in-situ data have been used to achieve the objectives. The first step was to identify the drought period, based on the time series of 20 years Normalized Difference Vegetation Index (NDVI), four current years’ time series of precipitation deficit, and comparative analysis of two different growing seasons (2017 and 2018) of the temperature difference between the surface and air(dTsa). The second step was to retrieve the vegetation properties ((leaf area index (LAI), leaf water content(Cw), chlorophyll content(Cab), and dry matter content(Cdm)) from a radiative transfer model for solar radiation in vegetation (RTMo), which is part of the 'Soil Canopy Observation of Photosynthesis and Energy fluxes' model (SCOPE),' then evaluated whether their seasonal course can be used as stress indicators. This retrieval is based on Sentinel-2 reflectance data. The simulated vegetation properties coupled with weather data from climate reanalysis data produced by the computer simulation model (ERA5) have been used in the SCOPE to simulate photosynthesis and evapotranspiration, which are the variables used to assess the severity and duration of the 2018 agricultural drought in the Raam region. The results obtained have been compared with groundwater levels, rainfall, leaf area index, and the estimated root zone soil moisture based on the weighted average method. The approach has been made by relating two different growing seasons: non-stressed toward water-stressed crops conditions 2017 versus 2018. This analysis showed that the critical period that indicates drought was the 2018 summer in the Raam region. The time series of NDVI was related to in-situ data of annual agricultural crop yield from 2000 to 2020 as documented by CBS. The NDVI time series demonstrates a significant relationship between the 2018 NDVI maturity stage and the seasonal 2018 annual yield reduction production. Most vegetation properties show the stress indicators in their season course, except chlorophyll content due to the insufficient cloud-free data in 2017. The simulated photosynthesis indicated drought in the 2018 summer as expected. The in situ data used for comparison were first analyzed to check if they show drought at the same period. The data used for root zone soil moisture of potato are for 40cm and 80 cm depth as potato roots can go shallow or deeper depending on where they are cultivated. In conclusion, the Sentinel-2 reflectance data can be relied on in agricultural drought monitoring of a specific crop. It has a high spatial resolution to provide accurate and detailed data that can be used to retrieve relevant vegetation properties. In this study, the combination of various indicators which react on all aspects of drought (meteorological, remote sensing indices, vegetation properties, and in-situ data) was used to develop a reliable method for agricultural drought monitoring on potato. This approach was based on multispectral Sentinel-2 data of 20m spatial resolution, which provided reliable photosynthesis results that matched with the hydrological drought as indicated by the in-situ data of groundwater and soil moisture.
Item Type:Essay (Master)
Faculty:ITC: Faculty of Geo-information Science and Earth Observation
Programme:Geoinformation Science and Earth Observation MSc (75014)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page