Climate influence on compound solar and wind droughts in Australia

Australia aims to significantly increase its renewable energy share by 2030, relying heavily on solar and wind power. This transition heightens the energy sector's vulnerability to weather and climate fluctuations. Wind and solar power's intermittency, governed by complex weather patterns, necessitates mitigation strategies such as battery storage and dispatchable renewables. The Australian Energy Market Operator (AEMO) manages the National Electricity Market (NEM), connecting five regions with a vast transmission network. Thirty-nine Renewable Energy Zones (REZs) have been designated for large-scale renewable energy investment. The spatial distribution of REZs offers an opportunity to enhance grid resilience by offsetting low production in one area with high production in another. However, the risk of weather-related grid-wide impacts remains. Studies in other regions have successfully linked renewable energy supply to specific weather patterns, informing energy system design and forecasting. In contrast, the relationship between weather patterns and grid-wide energy production in Australia, and the influence of large-scale climate modes, lacks systematic investigation. This study aims to address this gap by analyzing the spatial and temporal variability of solar and wind droughts, both individually and in combination (compound droughts), across multiple REZs within the NEM, identifying associated weather systems and the role of major climate modes in modulating drought frequencies.

Existing literature highlights the increasing impact of weather on electricity supply and demand globally. Studies have explored the spatio-temporal characterization of low solar irradiance events and the resilience of wind energy to climate change. Research has also focused on evaluating the complementarity of different renewable energy sources to mitigate the effects of climate variability, such as the El Niño-Southern Oscillation (ENSO). While studies in other regions have effectively linked renewable energy supply to weather patterns, the Australian context remains less explored. Previous work has shown the impact of ENSO on solar and wind power in Australia, but a comprehensive analysis of climate mode influences on grid-wide energy production is lacking. The need for a systematic analysis is crucial for informing strategic planning and seasonal forecasting of renewable resources.

The study utilizes climate data from the ECMWF ERA5 reanalysis product to analyze solar radiation and 100-meter wind speed as proxies for renewable energy resources. A solar or wind drought is defined when daily mean values fall below the 25th percentile of the 1959-2021 climatology across all REZs. Compound droughts are defined as the simultaneous occurrence of solar and wind droughts. The analysis focuses on 36 REZs with existing or planned solar or wind capacity. Regional differences in drought frequency are examined, and the synoptic weather conditions associated with widespread droughts (top 5% of days with the most REZs in drought) are analyzed using 10-meter wind speed and direction, cloud cover anomalies, and 2-meter temperature anomalies. The role of major climate modes (ENSO, IOD, SAM) is investigated by compositing sea-surface temperature and mean sea-level pressure anomalies during the seven seasons with the highest mean number of simultaneous droughts. The influence of climate modes on drought frequencies is assessed by analyzing individual grid cells and comparing drought frequencies between positive and negative phases of each mode. The analysis uses a variety of variables and indices to diagnose the synoptic and large-scale climate associated with renewable energy droughts. Data from ECMWF ERA5 reanalysis product, Hadley Centre Global Sea Ice and Sea Surface Temperature data set (HadISST), and climate mode indices (Niño3.4, DMI, SAM) are used.

The study reveals that compound droughts frequently affect multiple REZs simultaneously, particularly during winter. Solar droughts exhibit strong seasonality, with higher frequencies in southern regions, whereas wind droughts show more complex geographical patterns. Widespread solar droughts are characterized by anomalously high cloud cover across eastern Australia, often linked to moist onshore easterly flow and anticyclones. Widespread wind droughts are associated with anticyclonic conditions, particularly over Victoria and New South Wales. Compound droughts in winter and autumn show weather patterns similar to wind droughts but with significantly higher cloud cover. Widespread solar and compound droughts are sometimes associated with surface temperature anomalies, which might imply reduced energy demand. Analysis of climate mode indices indicates that the occurrence of widespread droughts is not strongly linked to specific phases of ENSO, IOD, or SAM, highlighting the spatial variability of climate mode teleconnections. However, regional differences exist, with some areas experiencing significantly higher drought frequencies during positive or negative phases of specific climate modes. For example, La Niña-like conditions and positive SAM phases are associated with increased solar drought frequencies in parts of eastern Australia, while El Niño-like conditions are associated with increased wind drought frequencies in various regions. The difference in drought frequencies between opposing phases of the climate modes can exceed ten days per season in some regions.

The findings highlight the vulnerability of Australia's energy grid to weather patterns causing widespread solar and wind droughts, especially during winter. The contrasting weather patterns associated with solar and wind droughts demonstrate the potential for compound events, which significantly impact renewable energy generation. The study's focus on daily data provides insights into the physical climate variables driving droughts but doesn't capture shorter-timescale events affecting energy sector risks. The limited predictability of widespread droughts based on major climate modes emphasizes the importance of regional analysis and strategic resource placement to mitigate the impact of droughts. The spatial variability of climate mode teleconnections suggests the potential for grid resilience by offsetting low generation in one region with high generation in another. Future work should incorporate more detailed energy metrics, demand models, and higher resolution data to better represent the complexity of the system.

This research demonstrates the significant risk of simultaneous solar and wind droughts across Australia's electricity grid, particularly in winter. The study highlights the complex interplay between synoptic weather patterns and large-scale climate modes in shaping drought frequencies. While major climate modes are not strong predictors of grid-wide droughts, regional variations offer opportunities for mitigation through strategic renewable energy deployment. Further research should explore higher-resolution data, incorporate energy-specific metrics, and refine models to improve subseasonal and seasonal forecasts of renewable energy resources.

The study's use of solar radiation and wind speed as proxies for energy production may introduce uncertainties, given the nonlinear relationships between climate variables and energy generation. The analysis focuses on daily data, potentially overlooking shorter-timescale events impacting the energy sector. The weighting of installed or planned generation capacities within REZs was not explicitly considered, which could influence the interpretation of results, particularly concerning solar droughts. Additionally, the analysis does not encompass all potential factors influencing energy demand.