Drought is a critical natural hazard in New Zealand, affecting both agricultural production and hydroelectric generation. However, the detection and characterisation of drought events are complicated by the range of different drought metrics available. Here, time-series variation in drought is investigated for New Zealand using two commonly used indices: the Standardized Precipitation Index (SPI) and the Standardized Precipitation Evapotranspiration Index (SPEI). Both the SPI and SPEI are calculated from WATCH Forcing ERA-Interim data (a bias-corrected version of ERA-Interim). Analyses are based on the proportion of New Zealand experiencing at least "moderate" drought (i.e. index values at the threshold of - 1 or lower). There is no statistically significant difference in area in drought between the two indices averaged over the time series, and no national-scale trends in drought occurrence are identified. However, a statistically significant trend of divergence between the SPI and SPEI is present when examining drought at moderate spatial coverage (10 % and 20 % coverage based on SPEI), with the SPEI showing increasingly greater drought coverage. These typically become more apparent either as a major drought event is building or terminating. As such, the SPEI generally indicates an extended duration (rather than higher peak extent) of drought events. The spatial expression of the largest SPI-SPEI differences reveals "hotspots" in terms of both average and divergent trends in drought extent, results which are reflected in a regional drought analysis using a cluster analysis of the SPI and SPEI area in drought. In particular, SPEI drought coverage is greater in the South Island east coast and central/southwestern North Island regions. Conversely, SPI drought extent is greater in the eastern North Island. Overall, the differences and similarities in drought extent between the SPI and SPEI can be linked to the highly varied nature of New Zealand climate, with the inst
展开▼
