Abstract
dc:description.abstractIn this study, maximum potential yield and maturity periods of three soybean varieties: Dina, SC-Safari, and SC-Spike were studied using remote sensing. To optimise this process, Landsat-8, PlanetScope, and Sentinel-2 satellites were combined into a virtual constellation that tracked the chlorophyll levels of the soybean fields with high spatial, spectral and temporal resolutions. A Random Forest algorithm was used to classify pixels thereby masking clouds and cloud’s shadows from the images. NDVI (Normalised Difference Vegetation Index), and EVI (Enhanced Vegetation Index) formed the basis for determining chlorophyll levels of the soybean’s canopies. Average values of NDVI and EVI per field were plotted against time using Gaussian process modelling and cubic spline interpolation to obtain their time-series profiles for 2016/2017, 2017/2018 and 2018/2019 farming seasons. Analysis done using linear, logarithm and power functions modelled the relationship of the maximum average EVI and NDVI to yield for each of the varieties. The analysis found that linear and logarithm functions explain this trend more accurately than power functions. Maximum average NDVI showed a higher correlation to yield in all three regressions than maximum average EVI. Maximum average NDVI versus yield R 2 values ran between 0.65 and 0.85, while maximum average EVI versus yield R 2 values ran between 0.28 and 0.83. Different regression equations were observed per variety. Extrapolation of their equation trend lines was used to rank the varieties according to maximum potential yield. SC-Spike showed the highest maximum yield potential in terms of metric tons per hectare after the crops were threshed. Followed by SC-Safari, Dina showed the least maximum potential yield. Maturity period was determined by measuring the time taken for the EVI and NDVI values at germination to recur during senescence. SC-Safari showed the earliest maturity period, followed by SC-Spike. Dina showed the most extended maturity period.
Degree
thesis:*- Name dc:type.qualificationname
- M.Sc by research
- Level dc:type.qualificationlevel
- masters
- Grantor dc:publisher.institution
- University of York
- Year dc:date.issued
- 2020
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Ngonga, Emmanuel Muteba
- Advisors dc:contributor.advisor
-
- Denby, Katherine
- Chigeza, Godfree
- Joseph, Fennel
- ORCID person:identifier.orcid
Rights
- Language dc:language.dcterms_iso639_2
- eng