Search

• Estimates of the time of emergence of climate change (ToE) for Australian rivers. • Stochastic data generation helps to better define baseline variability. • Uncertainty in ToE represented by 37 CMIP6 projections and 3 emissions scenarios. • ToE in river flows can precede ToE in precipitation by up to 50 years. • ToE substantially delayed or avoided in a low emissions pathway (SSP1-2.6). Climate change impacts on river flows are leading to substantial changes in water availability for human and natural systems. The significance of these impacts largely depends on the degree of change relative to natural variability. The 'Time of Emergence' (ToE) is the point in time (past or future) when the signal of climate change emerges from the noise of background variability. The ToE has been assessed for some climate variables but rarely streamflows. Here we investigate ToE of changes in streamflow across Australia for various characteristic periods relevant to different water management objectives. We assess the uncertainty in ToE estimates from different emissions pathways and a large ensemble of CMIP6 climate model projections. ToE is likely to occur earlier in hydrologic systems subject to low natural variability than those subject to high variability, and to systems that are vulnerable to failure over longer rather than shorter periods. There are clear regional differences in the patterns of ToE, with changes first emerging in south-west regions of Australia, followed by the south-east, then east and north. We estimate that ToE has already occurred in over 10% of sampled rivers for longer characteristic periods, primarily in south-west Australia. Importantly for planning, the results show that ToE in flows precedes ToE in precipitation by up to 50 years due to compounding effects of changes in annual precipitation, precipitation seasonality, and increasing temperatures. The number of catchments projected to have reached ToE by 2080 is heavily influenced by the trajectory of future emissions. A low emissions pathway generally consistent with aspirational goals of the Paris Climate Agreement (SSP1-2.6) substantially delays the emergence of significant hydrological change by up to 50 years, or negates emergence entirely by 2080, depending on location. Our methods can be applied to catchments globally, and our findings have implications for the prioritisation of climate adaptation efforts across different river systems and the urgency of continued action on climate change mitigation. [ABSTRACT FROM AUTHOR]

• The influence of management decisions on water allocations are often overlooked. • Effects of different water management rules were simulated in two Australian rivers. • Changing management rules alone varied long-term water allocations by −6 to +22%. • Influence of drivers varied with river system and reliability of water license. • More systematic and transparent water allocation decisions will be beneficial. A global challenge for water resource management in rivers worldwide is ensuring water supply reliability satisfies consumptive and environmental demands. High variability in water supply, water policy and management decisions, and uncertainty about the effects of climate change compound this challenge. Understanding factors driving water allocations and consequences for water users and the environment is essential. Rainfall and streamflow traditionally drive allocations, yet the influence of water management decisions are often overlooked because they are complex, rarely codified and dependent on the regional context. We compared the relative influences of water management, climate and the river system characteristics including demand and water storage infrastructure on regulated and unregulated water allocations in two regulated rivers with large dams for water supply to towns, irrigated agriculture and wetlands of international significance (Gwydir and Macquarie Rivers of the Murray-Darling Basin, southeastern Australia), using hydrological simulation models and regression-based sensitivity analyses under an historical and drier climate. Water management decisions influenced regulated water allocations (explaining 23% to 52% of the variance) considerably more than unregulated water allocations (explaining 2%). For regulated water allocations, water management decisions and the river system were more important than a plausible change in climate, while for unregulated water allocations, the river system and climate change were dominant drivers. Changing management rules alone could vary long term water allocations by between −6% and +0.1% in the Gwydir River and between −4% and +22% in the Macquarie River. Management rules which allocated future inflows (i.e. credit model) were more likely to fail in a drying environment with projected reductions in runoff, compared to rules where only water in storage is allocated (i.e. debit model). Management rules differed despite both river systems being governed by a single management agency under the same legislative and policy framework. More systematic approaches to justify water management rules and greater transparency in their influence on water allocations are critical for maximizing the benefits to water users and river health, and for managing risks to water supply in a variable and changing climate. [ABSTRACT FROM AUTHOR]

Climate change affects natural systems, leading to increased acceleration of global water cycle and substantial impacts on the productivity of tropical rivers and the several ecosystem functions they provide. However, the anticipated impacts of climate change in terms of frequency and intensity of extreme events (e.g., droughts and floods) on hydrological systems across regions could be substantially different. This study therefore aims to assess the impacts of climate change on the streamflow of a large river basin located in central Australia (Cooper Creek-Bulloo River Basin). Modified version of the hydrological model Hydrologiska Byråns Vattenbalansavdelning (HBV) was used in this study to generate daily streamflow. This model was first calibrated (2001–2010) and then validated for two independent periods (1993–1997 and 2011–2015). The model depicted a good performance in simulating observed streamflow. Climate projection data from multiple general circulation models (ACCESS1.0, CanESM2, CESM1-CAM5, CNRM-CM5, GFDL-ESM2M, HadGEM2-CC, MIROC5, NorESM1-M, ACCESS1-0, ACCESS1-3, CCSM4, CNRM-CM5, CSIRO-Mk3.6, GFDL-CM3, GFDL-ESM2M, HadGEM2, MIROC5, MPI-ESM-LR, and NorESM1-M) in various forms (raw, statistically downscaled, dynamically downscaled, and bias adjusted) were considered in this study. Results showed that three high resolution dynamically downscaled and bias adjusted models (ACCESS1-3, CNRM-CM5, and MPI-ESM-LR) from Terrestrial Ecosystem Research Network (TERN) dataset (v1.0.2) have better performance than other models considered, that is, in terms of capturing observed precipitation over the basin. Future climate projections of ensemble of these three models forced with RCP 4.5 and RCP 8.5 emission scenarios were then used to generate streamflow for 2050s (2040–2069) and 2080s (2070–2099). Results of the study indicated that mean annual precipitation was projected to decrease by up to −8% in 2050s and temperature was projected to increase by up to 4.66 °C in 2080s under the average and extreme emission scenarios, respectively. Mean annual, mean seasonal (December–February, March–May, June–August, September–November), and mean monthly streamflow were projected to decrease under different emission scenarios in 2050s and 2080s. These results indicate decreased water availability in the future as well as water cycle intensification. These changes in streamflow might have impacts on agriculture, natural ecosystem, and could lead to water restrictions. The outcome of this study can directly feed into frameworks for sustainable management of water resources and support adaptation strategies that rely on science and policy to improve water resources allocation in the region. [ABSTRACT FROM AUTHOR]