Water-use efficiency Decision Support System for irrigation in smallholder farms using machine learning

Mndela, Yonela

Title: Water-use efficiency Decision Support System for irrigation in smallholder farms using machine learning
Creator: Mndela, Yonela
Subject: Irrigation farming
Subject: Agriculture--Economic aspects
Subject: Agricultural development projects
Date Issued: 2024-09
Date: 2024-09
Type: Master's theses
Type: text
Identifier: http://hdl.handle.net/10353/29933
Identifier: vital:79207
Description: Small-scale farmers in arid and semi-arid regions rely mostly on irrigation to boost agricultural production and reduce dependency on unpredictable rainfall. However, due to poor farming knowledge, the efficiency of irrigation in these farms is low as crops are irrigated equally across the entire field without considering spatial variations in crop water requirements. This study was aimed at devising irrigation scheduling for various small-scale crops cabbage, sweet potato, maize, garlic, Florida broadleaf mustard, Solanum Retroflexum, green beans, sugar beans and spinach in the Mutale River catchment, based on changes in crop water status obtained from the regression models generated from the association between remotely sensed data and field data. Remotely sensed data in the form of unmanned aerial vehicle UAV imagery was used as the base from which crop spectral signatures and spatial patterns in water content were obtained. Endmember spectral analysis was performed to analyse spectral patterns of the crops. The Green Normalized Difference Vegetation Index GNDVI, Normalized Difference Vegetation Index NDVI, Normalized Difference Red edge NDRE, and Optimized Soil Adjusted Vegetation Index OSAVI were generated from the UAV imagery to serve as the base from which crops were characterized, and spatial patterns in water content were observed. Field survey method was carried out on the various crop plots to acquire leaf water content data fresh weight, turgid weight, and dry weight. The stratified random sampling method was used to select crops from which leaf samples would be extracted for measurements, based on crop type and growth stage. The measured leaf water content data was used to compute the Relative Water Content RWC which was used as an indication of water content in the crops. Field based RWC was used to observe spatial patterns in crop water content, as well as calibration and validation of crop water content empirical models retrieved from the Unmanned Aerial Vehicle UAV imagery. The Levene’s k-comparison test was performed to determine the spatial patterns in crop water content. A simple linear regression technique was employed to determine the nature and significance of the association of field based RWC and spectral vegetation indices. The linear regressions with the highest coefficient of determination r2 were used for modelling crop water content using spectral vegetation indices maps as the explanatory variable in the models. The time series regression technique was employed to simulate water changes across the crop types in the study area. The endmember spectral analysis results revealed variations in the spectral reflectance patterns of crops across the UAV spectral channels. A slight difference in spectral reflectance was noted across Solanum Retroflexum, green beans, sweet potato, pepper, maize, and sugar beans, as they all exhibited low and high reflectance in the red and near infrared channels, respectively. The spectral patterns of peas and cabbage were relatively easy to distinguish from the other crops because they exhibited high reflectance in the red-edge and green channels, respectively. The analysis of the spectral vegetation maps revealed that GNDVI, NDVI, NDRE, and OSAVI can be used to observe spatial variability of water content across various crops. In this study, the healthier plants were identified by a higher spectral index value (closer to +1) while unhealthy plants showed lower values closer to -1. The linear regression analysis revealed a significant association between GNDVI and water content of sweet potato, maize, sugar beans, and Florida broadleaf mustard, with r2 values of 0.948, 0.995, 0.978, and 0.953, respectively. NDVI revealed a strong association with water content of Solanum Retroflexum, pepper, and cabbage, with the r2 values of 0.949, 0.956, and 0.995, respectively. NDRE, on the other hand, revealed a strong relationship with water content in peas and green beans, with r2 values of 0.961, and 0.974, respectively. OSAVI was observed to be the least sensitive spectral index to water content across the surveyed crop types. The remotely sensed models produced in this study revealed that, RWC can be successfully predicted from UAV imagery. Time series regression revealed a gradual decrease in water content with an increase in the number of days for all the surveyed crops. The simulations revealed that Solanum Retroflexum, sweet potato, maize, sugar beans, and Florida Broadleaf Mustard reached their respective wilting points at day four after irrigation, implying that irrigation of these crops should be scheduled after every four 4 days basis. Peas, green beans, pepper, and cabbage reached their respective wilting points at day five after irrigation, implying that irrigation of these crops should be scheduled after every five days. The results of this study revealed that, with the current irrigation scheduling interval of seven days in the study area, the crops are subjected to water stress which has a huge impact on the quality and yield of the crops.
Description: Thesis (MSci) -- Faculty of Science and Agriculture, 2024
Format: computer
Format: online resource
Format: application/pdf
Format: 1 online resource (156 leaves)
Format: pdf
Publisher: University of Fort Hare
Publisher: Faculty of Science and Agriculture
Language: English
Rights: rights holder
Rights: All Rights Reserved
Rights: Open Access

Hits: 9
Visitors: 7
Downloads: 0

Collections

UFH Department of GIS & Remote Sensing

		Thumbnail	File	Description	Size	Format
View Details Download			SOURCE1	DISSERTATION_MNDELA_YONELA.pdf	53 MB	Adobe Acrobat PDF	View Details Download