03800nam a22002897a 4500008004100000040004200041084005100083100004800134245014200182260006300324300004700387500009500434500231800529502007702847504002502924610004002949610005102989650002703040650001903067650004603086650002203132710003203154942001303186998008803199999001703287952020603304260608t2025    bx ab||g |||| 00| 0 eng d  aUniversiti Teknologi BruneibengcUTB  aUTB 120 REPORT THESIS & DISSERTATIONaRTDS 4081 aNor Hamizah binti Haji Mohd Rhymeeeauthor.10aClimate change impact on irrigation water requirement and crop yield for paddy in Brunei Darussalam /cNor Hamizah binti Haji Mohd Rhymee  aBandar Seri Begawan :bUniversiti Teknologi Brunei,c2025.  a154 pages :billustrations, maps ;c30 cm.  aThesis submitted in fulfilment of the requirements for the degree of Doctor of Philosophy.  aABSTRACT Climate change poses unprecedented challenges to agricultural systems, particularly for small nations reliant on rice imports like Brunei Darussalam. With the vast majority of rice consumed being imported and climate change expected to significantly alter rainfall patterns and increase temperatures across Brunei, developing climate-resilient rice production strategies has become critically urgent. This study provides the first integrated assessment of climate change impacts on rice yields and irrigation water requirement (IR) in Brunei by combining statistical downscaling of Coupled Model Intercomparison Project 6 (CMIP6) scenarios, normalised vegetation difference index (NDVI)-based machine learning yield forecasting, and Water Evaluation And Planning (WEAP) software; a novel methodological framework for tropical rice systems. Future climate scenarios were assessed using statistical downscaling (SD) of multi-model ensemble (MME) techniques to simulate precipitation, minimum and maximum temperatures under different Shared Socioeconomic Pathways (SSPs) outlined by the CMIP6.
The downscaled climate models were then used to forecast peak NDVI using linear regression and random forest (RF) regression, with RF showing better performance. The developed peak-NDVI models predicted paddy yield for both main and off seasons for three future periods within 2020-2100. Key findings reveal distinct seasonal and temporal patterns: the main season's NDVI showed consistent trends across SSPs, with paddy yields declining slightly in the near future, increasing mid-future, and dropping sharply in the far future (30-40% reduction under SSP585), while off-season yields exhibited significant near-term trends alongside higher IWR due to hotter, drier conditions. Seasonal rainfall shifts disrupted current planting calendars, necessitating adaptation strategies. Shifting the main season to October improved irrigation water use efficiency (IUE) by 31-33%, while delaying the off-season to June enhanced IWE by 40-46. By integrating modern technologies (NDVI, machine learning) to overcome traditional challenges, this study bridges climate projections with on-farm decisions, offering actionable strategies and laying the groundwork for future agricultural and climatology advancements in Brunei.  aDissertation (Doctor of Philosophy) - Universiti Teknologi Brunei (2025)  aIncludes references. 4aUniversiti Teknologi BruneivThesis 4aUniversiti Teknologi BruneivFinal Year Report 4aDissertation, Academic 4aThesis writing 4adissertation Universiti Teknologi Brunei. 4aCivil Engineering  aUniversiti Teknologi Brunei  2lccRTDS  eReport, Thesis & Dissertations850580 : 002452 c.1 UTBxUniversiti Teknologi Brunei  c24216d24216  00102lc4070aUTBbUTBcLevel 2d2025-07-19eUniversiti Teknologi Bruneil0oUTB 120 REPORT THESIS & DISSERTATION, RTDS 408p850580r2026-06-08tc. 1w2026-06-08yRTDSzReg. No. 002452_UTB [RTDS 408]