An analysis of precipitation variability across New Zealand.
| dc.contributor.author | Marsh, Tom | |
| dc.date.accessioned | 2025-01-07T21:02:16Z | |
| dc.date.available | 2025-01-07T21:02:16Z | |
| dc.date.issued | 2024 | |
| dc.description.abstract | The distributed volunteer computing project, Weather@Home (W@H), has provided thousands of years of atmospheric simulation data at a 50 km grid resolution to explore precipitation distributions of New Zealand under various conditions. Two separate datasets were chosen to be analysed in detail. The first dataset included two climate regimes, a current era (2007- 2016) collection featuring the climate forcings present today, and a future era (2090s) collection featuring a projected warmer climate under a mixture of RCP4.5 and RCP8.5 CMIP5 scenarios. The second dataset included a 30-year (1986-2014) climatology under current anthropogenic and natural climate forcings to study the effects of El Nino Southern Oscillation (ENSO) on New Zealand precipitation patterns. Ordering the first dataset ensemble members by daily precipitation allowed us to determine the number of days taken to accumulate the top (wet) and bottom (dry) 25% of annual rainfall for each grid-cell over the country under both climate regimes. These data were then used to construct a metric for how the cumulative rainfall is expected to change at both a wet and dry end of the distribution in a warmer climate. It was found that the wettest days are expected to increase nearly uniformly across the country by around 14%, keeping in line with the idealised Clausius-Clapeyron relation of 7% K−1 for moist air. The driest days are expected to get 10-20% drier for much of the North Island and interior South Island, however, both coasts of the South Island are expected to get significantly wetter (in some areas by up to 20%) under the warmer climate. The key finding here was that both extreme ends of the precipitation distribution can work together to mask the unevenness of annual rainfall patterns that traditional mean shift studies might miss. For the second dataset, in-model Southern Oscillation Index (SOI) values were generated and compared against real-world values. Monthly cumulative precipitation data was determined to further aggregate the data into seasonal bins for each grid-cell. A simple “signal metric” for both El Nino and La Nina events (determined by in-model SOI) was designed to capture the population imbalance of cumulative monthly precipitation relative to the median of the ensemble. If the ratio of number of members was top-heavy this signified a wetter response for the grid-cell under that ENSO condition, while a bottom-heavy implied a drier response. It was found that under El Nino conditions, a general drying in the North Island and east coast of the South Island was present, with the strongest effects occurring during autumn and winter, while a wetter signal was present for regions in the lower left flank of the South Island peaking in autumn and spring. Conversely, under La Nina conditions this pattern was reversed with a wet signal for much of the country peaking in strength during winter and spring, with a mild drying of Fiordland. | |
| dc.identifier.uri | https://hdl.handle.net/10092/107877 | |
| dc.identifier.uri | https://doi.org/10.26021/15587 | |
| dc.language | English | |
| dc.language.iso | en | |
| dc.rights | All Rights Reserved | |
| dc.rights.uri | https://canterbury.libguides.com/rights/theses | |
| dc.title | An analysis of precipitation variability across New Zealand. | |
| dc.type | Theses / Dissertations | |
| thesis.degree.discipline | Physics | |
| thesis.degree.grantor | University of Canterbury | |
| thesis.degree.level | Masters | |
| thesis.degree.name | Master of Science | |
| uc.bibnumber | in1403185 | |
| uc.college | Faculty of Science | en |