Investigating the relationships between wheat-specific rainfall characteristics, large-scale modes of climate variability and wheat yields in the Swartland region, South Africa
Master Thesis
2014
Permanent link to this Item
Authors
Journal Title
Link to Journal
Journal ISSN
Volume Title
Publisher
Publisher
University of Cape Town
Faculty
License
Series
Abstract
Wheat producers in the South Western Cape (SWC) of South Africa need to cope with biophysical and socio-economic systems exposing farmers to a multidimensional decision- making environment. The rain fed wheat production in the Swartland region is highly susceptible to the interannual variability of winter rainfall. Producers, therefore, need relevant climatic information to identify ways to improve profitability and to make sound economic decisions. Seasonal forecasting has the potential to provide wheat producers with invaluable information regarding the climatic conditions. However, due to the complex nature of the atmospheric dynamics associated with winter rainfall in South Africa, seasonal forecasting models have been found to have very little skill in predicting the variability of winter rainfall. Such a shortfall has created a gap for which this study has attempted to bridge. This study aimed to investigate the relationship between wheat-specific rainfall characteristics, large-scale modes of climate variability and wheat yields in the Swartland region to assess whether these relationships could provide useful climatic information to the wheat farmers. Six wheat-specific rainfall characteristics (total rainfall ; number of wet days ; number of ‘good’ rainfall events; number of heavy rainfall events; percentage ‘good’ rainfall ; and the number of dry dekads ) on various time scales (winter; seasonal; monthly and dekadal) were correlated against wheat yield records over a 17 year period from 1994 to 2010. From this analysis, the distribution and timing of the rainfall throughout the wheat growing season (April to September) emerged as an important determinant of wheat yield. An accurate statistical wheat prediction model was created using farmer stipulated rainfall- wheat yield thresholds. Three teleconnections (El Niño-Southern Oscillation [ENSO], Antarctic Oscillation [AAO] and South Atlantic sea surface temperatures [SSTs]) represented by eight climate indices (Nino 3.4 Index, Ocean Nino Index [ONI], Southern Oscillation Index [SOI], AAO index, Southern Annular Mode Index [SAM], South Atlantic Dipole Index [SADI], South Western Atlantic SST Index [SWAI] and South Central Atlantic SST Index [SCAI]), were correlated against wheat yield data over a 17 year period from 1994 to 2010. The relationships between the three teleconnections and wheat yield in the Swartland were established. Teleconnection-wheat yield correlations were found to be limited, with regards to the application of this information to farmers, due to the lack of a comprehensive understanding of the dynamics of how the three teleconnections influence the local climate and, therefore, the wheat yield in the Swartland. The eight climate indices, representing the three teleconnections, were correlated against the six wheat-specific rainfall characteristic indices from each of the three study areas over the period from 1980 to 2012. The state of ENSO during the first half of the year was shown to be correlated with rainfall characteristics during both the first (April to July) and second (July to September) halves of the wheat growing season; however, these correlations differ ed in their sign. Correlations suggested a negative phase of AAO was associated with above normal rainfall throughout the year across the Swartland region. Sea surface temperatures in the central South Atlantic during March to October showed significant negative correlations with rainfall during the latter half of the wheat growing season (July to October) across the Swartland region. This study presented evidence supporting the plausibility and validity for the use of the state of large-scale modes of variability in the prediction of wheat-specific rainfall characteristics and aggregated yields in the Swartland region. This has the potential to provide useful information to wheat farmers in the Swartland to aid in their decision making process
Description
Includes bibliographical references.
Reference:
- Investigating the relationships between wheat-specific rainfall characteristics, large-scale modes of climate variability and wheat yields in the Swartland region, South Africa. Thesis. University of Cape Town ,Faculty of Science ,Department of Environmental and Geographical Science. http://hdl.handle.net/11427/13214