An integrated framework for hourly rainfall generation at multiple sites in the tropical region
This study aims to introduce an integrated framework to disaggregate daily rainfall data into hourly intervals at multiple sites. Traditional methods that evenly distribute daily rainfall over 24 hours often fail to either accurately reflect observed hourly rainfall due to their inability to capture...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis-Master by Research |
| Language: | English |
| Published: |
Nanyang Technological University
2024
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/181026 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Nanyang Technological University |
| Language: | English |
| Summary: | This study aims to introduce an integrated framework to disaggregate daily rainfall data into hourly intervals at multiple sites. Traditional methods that evenly distribute daily rainfall over 24 hours often fail to either accurately reflect observed hourly rainfall due to their inability to capture finer scale variability and intensity or capture the spatial correlation among various sites. Over the past decades, the Bartlett-Lewis Rectangular Pulse Model (BLRPM) can effectively capture temporal patterns at a single site, offering better representation of hourly rainfall data than simple disaggregation methods. Meanwhile, the MuDRain model, by accounting for spatial correlations among multiple sites, more accurately reproduces rainfall events across different locations, which is crucial for regions with complex rainfall patterns influenced by geographical and meteorological factors. This spatial consideration at a fine resolution (e.g. hourly) in synthesizing rainfall series is particularly valuable for urban rainfall-runoff modeling and flash flood simulations. However, such studies are very limited in the literature, especially for tropical cities like Singapore. This study specifically addressed this issue, discussing the pros and cons of simple disaggregation methods and investigating the advantages of using an integrated modeling scheme by sequentially coupling two stochastic models into a general framework. The results demonstrated that the integrated method could more comprehensively depict rainfall patterns in multi-site scenarios, enhancing the analysis of urban hydrological processes. |
|---|