The Impact of New Zealand's Topography on Quantitative Precipitation Forecasting Based on the Probability Matching Method

The forecast skill of the Probability Matching Method (PMM) was evaluated based on ensemble precipitation forecasts over New Zealand, with a detailed analysis of the influence of New Zealand's topography. This study used New Zealand ensemble forecast data in 2023 and employed multiple objective verification methods to statistically analyze the difference in performance of PMM over mountainous and plain areas of New Zealand. Results indicate that topographic factors cause significant differences in the skill of the ensemble mean (EM) and PMM between mountainous and plain regions of New Zealand. In the mountainous areas of New Zealand, the performance of the EM is largely comparable to, or slightly better than, PMM. In contrast, PMM outperforms the EM over plain areas. The primary reason for this difference is that the most precipitation affecting New Zealand, moving from west to east, first encounters the western mountainous regions. The unified topography induces uplift motions in all ensemble members, resulting in high spatial consistency in precipitation patterns over the mountains. The smoothing effect caused by inter-member differences is thus weaker. Over plain areas, however, the lack of uniform topographic forcing makes precipitation more sensitive to differences in flow-over-mountain conditions and local thermal-dynamic conditions, leading to greater relative differences among members. Consequently, PMM exhibits higher forecast skill relative to the EM in plains. This also gives PMM a greater advantage over the North Island of New Zealand, as the topographic influence is more dominant in the South Island.