Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
The applicability of large-scale rain-on-grid hydrological modelling in South Africa
Thesis (MEng)--Stellenbosch University, 2025.
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Language: | English |
| Published: |
Stellenbosch : Stellenbosch University
2025
|
| Subjects: | |
| Tags: |
No Tags, Be the first to tag this record!
|
| _version_ | 1867613957811863552 |
|---|---|
| access_status_str | Open Access |
| author | Bolt, Niels |
| author2 | Du Plessis, Kobus |
| author_browse | Bolt, Niels Du Plessis, Kobus |
| author_facet | Du Plessis, Kobus Bolt, Niels |
| author_sort | Bolt, Niels |
| collection | Thesis |
| dc_rights_str_mv | Stellenbosch University |
| description | Thesis (MEng)--Stellenbosch University, 2025. |
| format | Thesis |
| id | oai:scholar.sun.ac.za:10019.1/134523 |
| institution | Stellenbosch University (South Africa) |
| language | English |
| last_indexed | 2026-06-10T12:44:24.378Z |
| license_str | Other — see source repository |
| provenance_str_mv | Harvested via OAI-PMH from SUNScholar — Stellenbosch University Repository |
| publishDate | 2025 |
| publishDateRange | 2025 |
| publishDateSort | 2025 |
| publisher | Stellenbosch : Stellenbosch University |
| publisherStr | Stellenbosch : Stellenbosch University |
| record_format | dspace |
| source_str | SUNScholar — Stellenbosch University Repository |
| spelling | oai:scholar.sun.ac.za:10019.1/134523 The applicability of large-scale rain-on-grid hydrological modelling in South Africa Bolt, Niels Du Plessis, Kobus Stellenbosch University. Faculty of Engineering. Dept. of Civil Engineering. Rain and rainfall -- South Africa Hydrologic models -- South Africa Runoff -- Mathematical models Thesis (MEng)--Stellenbosch University, 2025. Bolt, N. 2025. The applicability of large-scale Rain-on-Grid hydrological modelling in South Africa. Unpublished masters thesis. Stellenbosch: Stellenbosch University [online]. Available: https://scholar.sun.ac.za/items/911455f3-0ead-4d88-a532-767a3e6d9dbf ENGLISH ABSTRACT: Rain-on-grid (RoG) modelling offers a direct way to simulate runoff generation from rainfall and two-dimensional routing without pre-defined hydrographs, but its skill at larger spatial scales under data constraints remains uncertain. This research evaluates the applicability and limits of RoG using HEC-RAS 2D (Version 6.4.1) in the Berg River Basin, South Africa (Misverstand outlet), with a 30 m Shuttle Radar Topography Mission-based (SRTM) Digital Elevation Model (DEM) and derived stream network. The research spans smaller-scale and large-scale applications, moving from controlled event-scale tests to catchment-wide calibration and diagnostics. The methodological programme first established a smaller-scale baseline through calibration and validation, then conducted large-scale calibration using three global adjustments (burned digital elevation model to enforce connectivity, reduced synthetic wetting depth, and selective South African National Land Cover (SANLC) river-channel corrections) and systematic parameter trials of Curve Number (CN) and Manning's n (n). A second stage varied channel-burn width/depth to test routing sensitivity. Robustness was then probed across four independent events (2001, 2008, 2016, 2018). Finally, upstream diagnostic tests and close-gauge rainfall comparisons were used to separate model-structure effects from rainfall input challenges. Results show a consistent pattern. At smaller-scale, event timing and hydrograph shape were reproduced with high fidelity, while magnitude errors persisted. At large-scale, routing refinements and plausible CN/n adjustments improved timing and shape locally, but did not remove the magnitude bias. Across independent events, the 2001 peak was overestimated by 12% with good shape; the first 2008 peak was underestimated by 4% but arrived 10.6 h early; the 2016 first peak was overestimated by 26% with 1.3 h delay; and the 2018 event showed pervasive overestimation, with the historical peak exceeded by 75%. These cross-event outcomes indicate that discrepancies are not purely structural and motivated targeted diagnostics. Upstream sub-catchment tests during the high-bias 2018 period retained substantial overestimation despite markedly smaller contributing areas (G1H008 and G1H078), indicating that reducing modelled area does not eliminate the discrepancy and shifting the weight of evidence toward rainfall input uncertainty (depth, timing, and spatial allocation). Complementary comparisons between pairs of rainfall gauges only 1.6 km and 2.4 km apart revealed multiple days with >10 mm differences in daily totals, demonstrating that sizeable inconsistencies can occur even over short distances with minimal elevation contrast. Together, these diagnostics support the interpretation that rainfall input uncertainty is the dominant driver of large-scale magnitude error in this setting. Terrain-routing experiments clarified the role of channel burning: enforcing a continuous thalweg reduced spurious storage and improved alignment, but stronger burns traded hydrograph realism for sharper timing as floodplain exchange was curtailed. No burn configuration alone resolved the remaining magnitude bias, underscoring that routing refinements cannot compensate for uncertain forcing. At 30 m resolution, smoothing of hillslope micro-topography and omission of minor headwater lines reduce hillslope-channel coupling, further complicating magnitude recovery. The research contributes: (i) a transparent calibration-to-diagnostic pathway for RoG under real data constraints; (ii) evidence that RoG can be relied upon for event timing and hydrograph evolution at small and medium scales; and (iii) a practical interpretation that large-scale magnitude fidelity is contingent on corroborated rainfall inputs and terrain representation that preserves connectivity. Limitations include reliance on available gauge networks, a 30 m DEM, and the lack of detailed information on hydraulic structure locations and dimensions. Recommendations prioritise a rainfall gauge reliability assessment, cautious use of burn strength, selective parameter refinement where justified, consideration of warm-start initialisation, and exploration of hydrologically conditioned DEMs for preserving hillslope pathways at coarse grids. AFRIKAANSE OPSOMMING: Geen opsomming beskikbaar. Masters 2025-12-12T07:36:59Z 2025-12-12T07:36:59Z 2025-12 Thesis https://scholar.sun.ac.za/handle/10019.1/134523 en Stellenbosch University xiv, 105 pages : illustrations application/pdf Stellenbosch : Stellenbosch University |
| spellingShingle | Rain and rainfall -- South Africa Hydrologic models -- South Africa Runoff -- Mathematical models Bolt, Niels The applicability of large-scale rain-on-grid hydrological modelling in South Africa |
| title | The applicability of large-scale rain-on-grid hydrological modelling in South Africa |
| title_full | The applicability of large-scale rain-on-grid hydrological modelling in South Africa |
| title_fullStr | The applicability of large-scale rain-on-grid hydrological modelling in South Africa |
| title_full_unstemmed | The applicability of large-scale rain-on-grid hydrological modelling in South Africa |
| title_short | The applicability of large-scale rain-on-grid hydrological modelling in South Africa |
| title_sort | applicability of large scale rain on grid hydrological modelling in south africa |
| topic | Rain and rainfall -- South Africa Hydrologic models -- South Africa Runoff -- Mathematical models |
| url | https://scholar.sun.ac.za/handle/10019.1/134523 |
| work_keys_str_mv | AT boltniels theapplicabilityoflargescalerainongridhydrologicalmodellinginsouthafrica AT boltniels applicabilityoflargescalerainongridhydrologicalmodellinginsouthafrica |