|
|
Exploring correlation between vegetation indices and plant nitrogen uptake
Pavlačková, Alena ; Doležalová Weissmannová, Helena (oponent) ; Kučerík, Jiří (vedoucí práce)
Excessive fertilization can cause environmental pollution, such as water contamination and greenhouse gas emissions, along with economic losses. To mitigate these issues, it is important to adjust fertilization rates to the specific needs of crops. This thesis explores the use of remotely sensed vegetation indices to monitor crop nitrogen uptake and guide fertilization application. The study was conducted in Oensingen, Switzerland, during an internship at ETH Zürich. The main objective was to develop a prediction model based on vegetation indices to estimate the nitrogen uptake of grass-clover mixtures and winter wheat. Additionally, the correlation between various vegetation indices and crop characteristics, especially nitrogen uptake, was analyzed. Vegetation indices (NDVI, NDRE, GNDVI, MCARI, EVI) were derived from Sentinel-2 images using Google Earth Engine. Various crop characteristics, including the Leaf Area Index (LAI) and crop height, were measured, and winter wheat samples were analyzed for nitrogen uptake using an elemental analyzer. Additional nitrogen uptake data for grass from previous years was also included. In total, data from the years 2021-2023, that included both grass-clover mixture and winter wheat values, were used in the analysis. Correlation and regression analysis were performed to examine the relationships between vegetation indices and the measured crop characteristics. The index showing the strongest relationship with crop nitrogen uptake was then used to create a prediction model. The analysis revealed that the Enhanced Vegetation Index (EVI) was the most effective predictor of nitrogen uptake. The constructed prediction model based on EVI values achieved a high coefficient of determination (R$^2$) of 0.89, a low root mean square error (RMSE) of 1.05, and a mean absolute error (MAE) of 0.89. The results indicate that EVI is a reliable index for predicting nitrogen uptake in crops. The developed EVI-based model could be potentially used for optimizing nitrogen application in crops, which can reduce the negative environmental and economic impacts of over-fertilization.
|
|
|
Dynamic metabolomic prediction from genetic variation
Nemčeková, Petra ; Weckwerth, Wolfram (oponent) ; Schwarzerová, Jana (vedoucí práce)
Hordeum vulgare, like many other crops, suffers from the reduction of genetic diversity caused by climate changes. Therefore, it is necessary to improve the performance of its breeding. Nowadays, the area of interest in current research focuses on indirect selection methods based on computational prediction modeling. This thesis deals with dynamic metabolomic prediction based on genomic data consisting of 33,005 single nucleotide polymorphisms. Metabolomic data include 128 metabolites belonging to 25 Halle exotic barley families. The main goal of this thesis is to create dynamic metabolomic predictions using different approaches chosen from relevant publications. The created models can be helpful for the prediction of phenotype or for revealing important traits of Hordeum vulgare.
|
|
|
Dynamic metabolomic prediction from genetic variation
Nemčeková, Petra ; Weckwerth, Wolfram (oponent) ; Schwarzerová, Jana (vedoucí práce)
Hordeum vulgare, like many other crops, suffers from the reduction of genetic diversity caused by climate changes. Therefore, it is necessary to improve the performance of its breeding. Nowadays, the area of interest in current research focuses on indirect selection methods based on computational prediction modeling. This thesis deals with dynamic metabolomic prediction based on genomic data consisting of 33,005 single nucleotide polymorphisms. Metabolomic data include 128 metabolites belonging to 25 Halle exotic barley families. The main goal of this thesis is to create dynamic metabolomic predictions using different approaches chosen from relevant publications. The created models can be helpful for the prediction of phenotype or for revealing important traits of Hordeum vulgare.
|
host ::
RSS.