Statistical analysis and prediction of climate impacts on sugarcane yield in southeastern Africa

Simple item page
dc.contributor.authorMbhamali, Thulebona Wellington
dc.date.accessioned2023-03-14T10:32:03Z
dc.date.available2023-03-14T10:32:03Z
dc.date.issued2020
dc.descriptionA dissertation submitted in fulfillment of the academic requirements for the degree of Master of Science in the Department of Geography and Environmental Studies in the Faculty of Science, Agriculture and Engineering, University of Zululand, 2020.en_US
dc.description.abstractSugarcane production is chiefly influenced by climate, but there are also farming operations that diminish the sugarcane yield. The study concentrated on discovering the statistical relationship between the climate and subsequent sugarcane yield by statistical analysis of monthly and annual data averaged over the area 31-23°S and 28-33°E in the period from 1970-2016 with respect to FAO data and 1988-2010 for SASA data. R procedures were employed to derive statistical associations between sugarcane yield and climate variables and indices. Time series analysis was performed, including Mann-Kendall tests for trend detection, Pearson correlation and MLR for exploring the statistical links between yield and climate parameters. Moreover, the spatio-temporal correlations and regressions were performed between the sugarcane yield index and relevant local crop drivers such as rainfall and temperature, and the global sea surface temperatures and winds through online tools. The results demonstrated a downward trend in sugarcane yield over SE Africa as signalled by 𝑧 ≈ −4 for FAO time series and 𝑧 ≈ −2 for SASA yield data at 95% level. Time series analysis showed high influence of local climate indices such as PDSI (e.g 𝑟 ≈ +0.7) with global sugarcane yield, and the value increased to 𝑟 ≈ +0.80 locally (e.g in South Coast region). A positive correlation for rainfall (𝑟 > +0.5) and negative for temperature (𝑟 < −0.5) conditions in the preceding December to May with the harvested sugarcane yield in the spatial analysis was observed. Hence, it is understood that high temperatures induced by offshore air flow and low pressure tend to suppress sugarcane yield, and correspond with rainfall below 1000 mm/yr. The surface air temperature is by far the most important indicator of sugarcane yield in ESwatini while PDSI explains most of the yield variation in South Africa. The findings show that climate influences on sugarcane yield cover 40-80% of variance, and that the decline and variability of sugarcane production has distinct climatic factors. Meridional wind flow over the western Indian Ocean and anticyclonic circulation in the Mozambique Channel manifest themselves as key climatic features for sugarcane production over SE Africa. Recommendations for ways to mitigate and adapt to climate variability are vi given to benefit the sugar industry in southeastern Africa. In addition, a statistical model to predict sugarcane yield over the study area was developed through MLR algorithms extracted from large-scale climate drivers.en_US
dc.identifier.urihttps://hdl.handle.net/10530/2284
dc.language.isoenen_US
dc.publisherUniversity of Zululanden_US
dc.subjectClimate impactsen_US
dc.subjectTemperatureen_US
dc.subjectPrecipitationen_US
dc.subjectSugarcane yielden_US
dc.titleStatistical analysis and prediction of climate impacts on sugarcane yield in southeastern Africaen_US
dc.typeThesisen_US
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Statistical analysis and prediction of climate impacts on sugarcane yield in Southeastern Africa.pdf
Size:
5.56 MB
Format:
Adobe Portable Document Format
Description:
Statistical analysis and prediction of climate impacts on sugarcane yield in Southeastern Africa
Download
License bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
license.txt
Size:
1.83 KB
Format:
Plain Text
Description:
Download
Collections
Geography and Environmental Studies