Projected changes in New Zealand drought risk

An updated assessment using multiple drought indicators

Previous work has shown that the intensity, duration and frequency of future droughts for New Zealand will largely depend on the future greenhouse gas concentrations increase, with drought characteristics becoming more severe for the higher emission scenarios in most regions.

This report applies a series of indices used to monitor drought in New Zealand to develop an improved method of analysing future drought risk under scenarios of climate change, and thereby supports improved understanding of risks to potable water supplies resulting from climate change.

Some key findings of this report are:

• New Zealand drought is characterised by the “New Zealand Drought Index” (NZDI) and four constituent indices. The leading cause of increase in drought risk is the temperature-driven increase in atmospheric water demand which leads to drier soil conditions, water stress for
  vegetation and low hydrological flows. Lower emission scenarios will result in lower drought risk increase since temperature increase is considerably lower nationwide.
• Precipitation deficit due to changing precipitation patterns is also a major driver of drought in some regions, but increased drying due to increased temperature alone may be sufficient to trigger droughts even when there is no change in or even an increase in precipitation.
• Whilst in some regions the frequency of “back-to-back” droughts may appear to decrease due to higher winter and spring precipitation due to climate change, the national trend is that they will increase. Even in some regions, such as around Nelson and Tasman, with lower
  “back-to-back” drought risk conditions, higher risk of deeper drought in the main growing season is likely to have a strong impact.
• The lower drought risk conditions that are mostly prevalent in the winter and early spring season are in contrast to the higher risk of drought later in the main growing season.
• Stream and river flow and drinking water reservoir levels are likely be adversely affected especially in the late spring and summer seasons at all sites used in this study impacted by “very dry” to “drought” conditions.
• Drought risk in most regions, and for almost all sites used in this study, is mostly impacted by increases in summer temperature drying the soils, as evident in significant increase in Soil Moisture Deficit (SMD) and the climate drought index, Clim-NZDI. In the winter season, some eastern and southern North Island and the two South Island sites are wetter than in the past (1990s) and can avoid “back-to-back” drought risk due to improved winter recharge. The Northland and Gisborne sites, on the other hand, are at higher risk of “back-to-back” drought due to drier winters.
• This report represents a major advance in the diagnosis of future New Zealand droughts. For the first time, the annual cycle of the drought indices is represented at a daily time-step, and the four constituent indices making up the Clim-NZDI allow us to assign causes to the
  drought changes, in particular whether they are driven by temperature or rainfall changes.

The drought calculations used in this study are derived from simulations by six global climate models, driven by four IPCC Fifth Assessment emission scenarios known as RCPs 2.6, 4.5, 6.0 and 8.5. The global models are dynamically downscaled using a 27-km regional climate model, and subsequently bias-corrected and further downscaled to a 5-km grid covering the whole country.

The drought index used is known as the New Zealand Drought Index (NZDI), which is similar to NIWA’s operational calculations from observed data but applied to the climate model output. However, some technical differences in the calculation that we use here require us to distinguish this
new climate drought index which we refer to as “Clim-NZDI”. Clim-NZDI is comprised of four component indices, representing the Standardised Precipitation Index (SPI), the Soil Moisture Deficit (SMD), the Soil Moisture Deficit Anomaly (SMDA), and the Potential Evapotranspiration Deficit (PED).