University of Twente Student Theses
Estimation of crop water use with Sentinel 1 and 2 data integration.
Kyeku, Paul (2023) Estimation of crop water use with Sentinel 1 and 2 data integration.
|
PDF
1MB |
| Abstract: | Evapotranspiration (ET) is an important parameter that influences the availability of water for crop production ET comprises of two processes: evaporation, which is water loss from soil and vegetation surface s , and transpiration, which is water release through plant roots and leaf openings. Accurate and timely information about ET is essential for understanding the global water ba lance , which is vital for irrigation scheduling, assessing plant water requirements , and overall water resource management. However, accurate estimati on of ET, especially on a larger scale, is clogged with a lot of uncertainties T he choice of data used as input for ET estimates is a key factor contributing to these uncertainties. Similarly, the partitioning of evapotranspiration into its components of evaporation and transpiration also contributes to the uncertainties of ET estimate To address these uncertainties and enhance accurate ET estimation, this study leveraged the capabilities of both Sentinel 1 (sensitive to canopy structure and soil moisture content) and Sentinel 2 ( sen sitive to photosynthesis, canopy structure, and moisture contents )), b y integrating their indices. The analysis was performed on two different water regimes, allowing for exclusive evaluati on of the performance of Sentinel and Sentinel 2 indices This was essential to see if there would be difference s in the predictive power of the RF model. Furthermore, the most important vegetation and polarimetric indices were evaluated. The result revealed that Chlorophyll red edge (Chlre) was the most important vegetation index given its sensitivity to chlorophyll contents in plants . The results proved that the integrated use of Sentinel 1 and Sentinel 2 improved ET estimates on both rainfed and irrigated agriculture achieving an overall R 2 =0.6 7 . Sentinel 2 accounted fo r 66% of the variability in ET . While Sentinel 1 indices explained 26% variability in ET suggesting that Sentinel 1 cannot explain variability in ET in isolation. The result of the ET estimate for rainfed agriculture using all sensors S1+S2 (rainfed) was reduced by 2%. The outcomes of irrigated agriculture, though subject to scrutiny due to the limited sample size, exhibited a 2% improvement . This improvement was expected given the substantial occurrence of evapotranspiration in irrigated agriculture found in the literature Overall, Sentinel 2 has contributed to the total ET in this study d ue to its sensitivity to chlorophyll and moisture content. These parameters are related to photosynthesis, which is directly connected to ET process |
| Item Type: | Essay (Master) |
| Faculty: | ITC: Faculty of Geo-information Science and Earth Observation |
| Subject: | 38 earth sciences |
| Programme: | Geoinformation Science and Earth Observation MSc (75014) |
| Link to this item: | https://purl.utwente.nl/essays/97223 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page