Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
The use of satellite-derived data and neural-network analysis to examine variation in maize yield under changing climate
Thesis (PhD)--University of Pretoria, 2019.
Saved in:
| Other Authors: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
University of Pretoria
2019
|
| Subjects: | |
| Tags: |
No Tags, Be the first to tag this record!
|
| _version_ | 1867613536353517569 |
|---|---|
| access_status_str | Open Access |
| author2 | Darkey, Daniel |
| author_browse | Darkey, Daniel |
| author_facet | Darkey, Daniel |
| collection | Thesis |
| dc_rights_str_mv | © 2019 University of Pretoria. All rights reserved. The copyright in this work vests in the University of Pretoria. No part of this work may be reproduced or transmitted in any form or by any means, without the prior written permission of the University of Pretoria. |
| description | Thesis (PhD)--University of Pretoria, 2019. |
| format | Thesis |
| id | oai:repository.up.ac.za:2263/70338 |
| institution | University of Pretoria (South Africa) |
| language | English |
| last_indexed | 2026-06-10T12:37:42.457Z |
| license_str | Other — see source repository |
| provenance_str_mv | Harvested via OAI-PMH from UPSpace — University of Pretoria Institutional Repository |
| publishDate | 2019 |
| publishDateRange | 2019 |
| publishDateSort | 2019 |
| publisher | University of Pretoria |
| publisherStr | University of Pretoria |
| record_format | dspace |
| source_str | UPSpace — University of Pretoria Institutional Repository |
| spelling | oai:repository.up.ac.za:2263/70338 The use of satellite-derived data and neural-network analysis to examine variation in maize yield under changing climate Darkey, Daniel lolaadisa@yahoo.com Botai, Joel Hassen, Abubeker Tesfamariam, Eyob Adisa, Omolola Mayowa Geoinformatics UCTD Thesis (PhD)--University of Pretoria, 2019. Climate change and variability is foreseen to have direct and indirect effects on the existing agricultural production systems potentially threatening local, regional and/or global food security depending on the spatial scale of the change. The trend and level of impact caused by climate change and/or variability is region dependent and adaptive capacity. Climate change is projected to have more adverse impact in high vulnerability areas of sub-Saharan Africa. This study aimed to examine the variation in maize yield and develop a framework for predicting maize yield in response to climate change. To achieve this aim, this study has analyzed the impact of agro-climatic parameters on maize production across the major four maize producing provinces of South Africa. This study went further to investigate changes in the satellite derived phenological parameters and its relationship with maize production. In addition, the influence of drought (a derivative of climate change) on maize production was investigated. The study concluded by integrating all datasets used in each objective to develop an empirical predicting model using artificial neural network. Previous studies have quantified the impact of climatic variables on maize and at a small geographic area. Attempts to predict maize yield have been minimal and the use of artificial intelligence such as the artificial neural network has not been conducted. In this study, alternative sources of climatic and environmental data have been employed using remotely sensed data which offers possibilities of collecting continuous data over a large area (including remote areas) through the use of satellite. The analysis of agro-climatic variables (precipitation, potential evapotranspiration, minimum and maximum temperatures) spanning a period of 1986–2015, over the North West, Free State, Mpumalanga and KwaZulu-Natal (KZN) provinces, indicated that there is a negative trend in precipitation for North West and Free State provinces and positive trend in maximum temperature for all the provinces over the study period. Further more, the result showed that one or more different agro-climatic variables has more influence on maize across the provinces. Analysis of the phenological parameters of maize indicated that climate change and climate variability affect plant phenology largely during the vegetative and reproductive stages. NDVI values exhibited a decreasing trend across the maize producing provinces of South Africa. The results further demonstrate that the influences of climate variables on phenological parameters exhibit a strong space-time and common covariate dependence. Agro-climatic variables can predict about 46% of the variability of phenology indicators and about 63% of the variability of yield indicators for the entire study area. The study also illustrated the spatial patterns of drought depicting drought severity, frequency, and intensity which has the potential to influence crop yield. The study found that maize yield is most sensitive to 3-month timescale coinciding with maize growing season (r = 0.59; p <0.05) affecting maize yield by up to 35% across the study area. In ensuring and fulfilling one of the seventeen sustainable development goals; to eradicate extreme poverty and hunger, the development of a system capable of monitoring and predicting crop yield becomes imperative. Machine learning tools such as the artificial neural network becomes handy and useful to provide a platform that is data intensive and robust to meet the requirements for an effective monitoring and predictive system for crop; particularly maize. The accuracy of the comparison between the actual and predicted maize yield is averaged at about 92% across the study area. The empirical model developed in this study can also be adopted to other grain crops such as Sorghum, wheat, soya beans etc. EU FP7 AnimalChange project under the grant agreements no. 266018 Geography, Geoinformatics and Meteorology PhD Unrestricted 2019-07-01T12:46:51Z 2019-07-01T12:46:51Z 2019-09-05 2019 Thesis Adisa, OM 2019, The use of satellite-derived data and neural-network analysis to examine variation in maize yield under changing climate, PhD Thesis, University of Pretoria, Pretoria, viewed yymmdd <http://hdl.handle.net/2263/70338> S2019 http://hdl.handle.net/2263/70338 en © 2019 University of Pretoria. All rights reserved. The copyright in this work vests in the University of Pretoria. No part of this work may be reproduced or transmitted in any form or by any means, without the prior written permission of the University of Pretoria. application/pdf University of Pretoria |
| spellingShingle | Geoinformatics UCTD The use of satellite-derived data and neural-network analysis to examine variation in maize yield under changing climate |
| title | The use of satellite-derived data and neural-network analysis to examine variation in maize yield under changing climate |
| title_full | The use of satellite-derived data and neural-network analysis to examine variation in maize yield under changing climate |
| title_fullStr | The use of satellite-derived data and neural-network analysis to examine variation in maize yield under changing climate |
| title_full_unstemmed | The use of satellite-derived data and neural-network analysis to examine variation in maize yield under changing climate |
| title_short | The use of satellite-derived data and neural-network analysis to examine variation in maize yield under changing climate |
| title_sort | use of satellite derived data and neural network analysis to examine variation in maize yield under changing climate |
| topic | Geoinformatics UCTD |
| url | http://hdl.handle.net/2263/70338 |