Skills and Expertise
Jul 2012 - Oct 2016
The Commonwealth Scientific and Industrial Research Organisation · Oceans and Atmosphere
Research Items (62)
The Earth has warmed over the past century. The warming rate (amount of warming over a given period) varies in time and space. Observations show a recent increase in global mean warming rate, which is initially maintained in model projections, but which diverges substantially in future depending on the emissions scenario followed. Scenarios that stabilize forcing lead to much lower warming rates, as the rate depends on the change in forcing, not the amount. Warming rates vary spatially across the planet, but most areas show a shift toward higher warming rates in recent decades. The areal distribution of warming rates is also changing shape to include a longer tail in recent decades. Some areas of the planet are already experiencing extreme warming rates of about 1 °C/decade. The fat tail in areal distribution of warming rates is pronounced in model runs when the forcing and global mean warming rate is increasing, and indicates a climate state more prone to regime transitions. The area-proportion of the Earth displaying warming/cooling trends is shown to be directly related to the global mean warming rate, especially for trends of length 15 years and longer. Since the global mean warming rate depends on the forcing rate, the proportion of warming/cooling trend areas in future also depends critically on the choice of future forcing scenario.
Climate variability, climate change and extreme events pose risks that need to be quantified and managed. Dry and hot conditions have notable impacts, and have a strong link to drought risk. Many extreme event analyses focus on one variable at a time. However, compound extremes, involving two or more climate variables, can have a disproportionately large impact. Thus integrated multivariate analyses are necessary to comprehensively assess climate impacts. Here we document 150 years of information about events with low monthly rainfall and high temperature for southeast Australia. The number of hot/dry months per year exhibits decadal variability and increasing trends. Long-term trends are more influenced by temperature than rainfall, consistent with a warming climate. The number of hot and dry consecutive events, defined as three to five consecutive months of compound events, is increasing. Our findings reinforce the need to consider definitions that include multivariate variables such as rainfall and temperature and/or other hydroclimate variables, where possible, when quantifying drought risk. Discussion on how the results could contribute to improvement in climate projection science in Australia or elsewhere is provided.
Anthropogenic climate change and El Niño made small but significant contributions to increasing the likelihood of record low rainfall in October 2015 in Tasmania. Atmospheric variability was the main contributor.
The perception of the accuracy of regional climate projections made in the early 1990s about climate change by 2030 may be influenced by how the temperature trend has changed in the 25 years since their publication. However, temperature trends over this period were influenced not only by external forcings such as greenhouse gases but also natural variations. The temperature of Southern Australia, the Sahel, South Asia and Southern Europe are currently within the warming estimates from statements in the early 1990s from the IPCC and CSIRO, assuming a linear trend between 1990 and 2030. However, northern Australia and central North America are currently at the lower limit or below these projections, having featured areas of multi-year regional cooling during global warming, sometimes called ‘warming holes’. Recent climate model simulations suggest that cooling can be expected in the recent past and near future in some regions, including in Australia and the US, and that cooling is less likely over 1990–2030 than in 1990–2015, bringing observations closer to the IPCC and CSIRO warming estimates by 2030. Cooling at the 25-year scale in some regions can be associated with cyclic variability such as the Inter-decadal Pacific Oscillation, or low trend such as in the Southern Ocean. Explicitly communicating the variability in regional warming rates in climate projections, including the possibility of regional warming ‘holes’ (or the opposite of ‘surges’ or ‘peaks’) would help to set more reliable expectations by users of those projections.
Impact, adaptation and vulnerability (IAV) research underpin strategies for adaptation to climate change and help to conceptualise what life may look like in decades to come. Research draws on information from global climate models (GCMs) though typically post-processed into a secondary product with finer resolution through methods of downscaling. Here we consider the production process as a chain of processes leading to an application-ready data set, where each step may have a significant impact on the climate change signal. Through worked examples set in an Australian context we assess the influence of GCM sub-setting, geographic area sub-setting and downscaling method on the regional change signal. Examples demonstrate that choices impact on the final results differently depending on various factors such as application needs, range of uncertainty of the projected variable, amplitude of natural variability, and size of study region. For heat extremes, the choice of emissions scenario is of prime importance, but for a given scenario the method of preparing data can affect the magnitude of the projection by a factor of two or more, strongly affecting the indicated adaptation decision. For our catchment level runoff projections, the choice of emission scenario is less dominant. Rather the method of selecting and producing application-ready datasets is crucial as demonstrated by results with opposing sign of change, raising the real possibility of mal-adaptive decisions. This work illustrates the potential pitfalls when using unwise GCM sub-sampling or the use of a single downscaled product when conducting IAV research.
Atmospheric circulation change is likely to be the dominant driver of multidecadal rainfall trends in the midlatitudes with climate change this century. This study examines circulation features relevant to southern Australian rainfall in January and July and explores emergent constraints suggested by the intermodel spread and their impact on the resulting rainfall projection in the CMIP5 ensemble. The authors find relationships between models' bias and projected change for four features in July, each with suggestions for constraining forced change. The features are the strength of the subtropical jet over Australia, the frequency of blocked days in eastern Australia, the longitude of the peak blocking frequency east of Australia, and the latitude of the storm track within the polar front branch of the split jet. Rejecting models where the bias suggests either the direction or magnitude of change in the features is implausible produces a constraint on the projected rainfall reduction for southern Australia. For RCP8.5 by the end of the century the constrained projections are for a reduction of at least 5% in July (with models showing increase or little change being rejected). Rejecting these models in the January projections, with the assumption the bias affects the entire simulation, leads to a rejection of wet and dry outliers.
The projected warming of surface air temperature at the global and regional scale by the end of the century is directly related to emissions and Earth’s climate sensitivity. Projections are typically produced using an ensemble of climate models such as CMIP5, however the range of climate sensitivity in models doesn’t cover the entire range considered plausible by expert judgment. Of particular interest from a risk-management perspective is the lower impact outcome associated with low climate sensitivity and the low-probability, high-impact outcomes associated with the top of the range. Here we scale climate model output to the limits of expert judgment of climate sensitivity to explore these limits. This scaling indicates an expanded range of projected change for each emissions pathway, including a much higher upper bound for both the globe and Australia. We find the possibility of exceeding a warming of 2 °C since pre-industrial is projected under high emissions for every model even scaled to the lowest estimate of sensitivity, and is possible under low emissions under most estimates of sensitivity. Although these are not quantitative projections, the results may be useful to inform thinking about the limits to change until the sensitivity can be more reliably constrained, or this expanded range of possibilities can be explored in a more formal way. When viewing climate projections, accounting for these low-probability but high-impact outcomes in a risk management approach can complement the focus on the likely range of projections. They can also highlight the scale of the potential reduction in range of projections, should tight constraints on climate sensitivity be established by future research.
This paper describes the history of national climate change projections for Australia since 1987, with a focus on the series of statements in 1992, 1996, 2001, 2007 and 2015. These were prepared by CSIRO up to 2001, and by CSIRO and the Bureau of Meteorology from 2007 onward. A range of scientific and communication issues were addressed in preparing each statement, including decisions concerning climate model ensembles, emission scenarios, forming ranges of change, use of probability, use of expert judgment, spatial resolution, presentation methods and representing uncertainties. There are a number of _perennial_ issues, _trends_ and _tensions_, which may be of interest to future production of regional projections for Australia and other regions. For example, managing and communicating uncertainty in future climate due to differing emissions and model responses has been a _perennial_ element of the projections. There has been a _trend_ towards wider scope in variables analysed, time periods discussed and use of peer review, as well as greater content in the statements over time, partly reflecting available modelling results and the increasing range, needs and sophistication of users. There are several notable _tensions_ in this work, reflected in some approaches being adopted and then dropped in subsequent statements. Examples include the choice of spatial resolution, the use of probability, model evaluation and expert judgement. These tensions reflect the difficulty in striking the right balance between competing scientific considerations or between scientific credibility and saliency for users.
It is likely that human influences on climate increased the odds of the extreme high pressure anomalies south of Australia in August 2014 that were associated with frosts, lowland snowfalls and reduced rainfall.
There is a high degree of variation in rainfall projections for later this century for Australia's eastern seaboard, partly because of how different climate models represent the relevant physical processes. These processes include local environmental conditions, synoptic phenomena and large-scale atmospheric and oceanic modes of variability. We review these processes using a wide range of analyses from observations and modelling. A synthesis of this review is used to produce likelihood and confidence measures for the rainfall projections, intended to inform planning and adaptation. The most likely projected outcome in southeast Queensland is a small decrease in rainfall for all seasons, with higher confidence in autumn than other seasons due to uncertainties associated with representing the El Niño-Southern Oscillation, thunderstorms, tropical cyclones and the monsoon. In northeast New South Wales (NSW), there is projected to be a small increase for summer, little change for autumn, a decrease for winter and a small decrease for spring, with confidence being highest in winter and lowest in summer. These projections correspond to an intensification of the annual cycle in northeast NSW. Natural variability in mean rainfall is projected to remain significant in comparison to the climate change signal throughout the eastern seaboard region. In contrast to mean rainfall, there is high confidence in a projected increase in the frequency of extreme rainfall throughout this region.
The term ‘downscaling’ refers to the process of translating information from global climate model simulations to a finer spatial resolution. There are numerous methods by which this translation of information can occur. For users of downscaled information, it is important to have some understanding of the properties of different methods (in terms of their capabilities and limitations to convey the change signal, as simulated by the global model), as these dictate the type of applications that the downscaled information can be used for in impact, adaptation, and vulnerability research. This article provides an appraisal of downscaling in terms of its perceived purpose and value for informing on plausible impacts due to climate change and for underpinning regional risk assessments. The concepts climate realism and physical plausibility of change are introduced to qualify the broad scale properties associated with different categories of downscaling approaches; the former concerning the skill of different approaches to represent regional climate characteristics and the latter their skill in simulating regional climate change. Aspects of change not captured by global climate models, due to resolution or regional factors, may be captured by downscaling. If these aspects are of interest, then downscaling may be useful once it has been demonstrated to add value. For cases where the broad scale change to the mean climate is of interest, or where there is no demonstrated added value from downscaling, then there is a wide range of regionalization methods that are suitable for practitioners in the impact, adaptation, and vulnerability field.For further resources related to this article, please visit the WIREs website.Conflict of interest: The authors have declared no conflicts of interest for this article.
The Australian eastern seaboard is a distinct climate entity from the interior of the continent, with different climatic influences on each side of the Great Dividing Range. Therefore, it is plausible that downscaling of global climate models could reveal meaningful regional detail, or 'added value', in the climate change signal of mean rainfall change in eastern Australia under future scenarios. However, because downscaling is typically done using a limited set of global climate models and downscaling methods, the results from a downscaling study may not represent the range of uncertainty in plausible projected change for a region suggested by the ensemble of host global climate models. A complete and unbiased representation of the plausible changes in the climate is essential in producing climate projections useful for future planning. As part of this aim it is important to quantify any differences in the change signal between global climate models and downscaling, and understand the cause of these differences in terms of plausible added regional detail in the climate change signal, the impact of sub-sampling global climate models and the effect of the downscaling models themselves. Here we examine rainfall projections in eastern Australia under a high emissions scenario by late in the century from ensembles of global climate models, two dynamical downscaling models and one statistical downscaling model. We find no cases where all three downscaling methods show the same clear regional spatial detail in the change signal that is distinct from the host models. However, some downscaled projections suggest that the eastern seaboard could see little change in spring rainfall, in contrast to the substantial rainfall decrease inland. The change signal in the downscaled outputs is broadly similar at the large scale in the various model outputs, with a few notable exceptions. For example, the model median from dynamical downscaling projects a rainfall increase over the entirety of eastern Australia in autumn that is greater than the global models. Also, there are some instances where a downscaling method produces changes outside the range of host models over eastern Australia as a whole, thus ex-panding the projected range of uncertainty. Results are particularly uncertain for summer, where no two downscaling studies clearly agree. There are also some confounding factors from the model configuration used in downscaling, where the particular zones used for statis-tical models and the model components used in dynamical models have an influence on results and produce additional uncertainty.
A projected drying of the extra-tropics under enhanced levels of atmospheric greenhouse gases has large implications for natural systems and water security across southern Australia. The drying is driven by well studied changes to the atmospheric circulation and is consistent across climate models, providing a strong basis from which adaptation planners can make decisions. However, the magnitude and seasonal expression of the drying is expected to vary across the region. Here we describe the spatial signature of the projected change from the new CMIP5 climate models and downscaling of those models, and review various lines of evidence about the seasonal expression. Winter rainfall is projected to decline across much of southern Australia with the exception of Tasmania, which is projected to experience little change or a rainfall increase in association with projected increases in the strength of the westerlies. Projected winter decrease is greatest in southwest Western Australia. A 'seasonal paradox' between observations and CMIP5 model projections in the shoulder seasons is evident, with strong and consistent drying pro-jected for spring and less drying projected for autumn, the reverse of the observed trends over the last 50 years. The models have some biases in the simulation of certain synoptic types (e.g. cutoff lows), the rainfall brought by those synoptic types, and the mechanism of rainfall pro-duction. Rainfall projections based on statistical downscaling are examined in relation to some of these biases, projecting stronger future declines in some regions of southeast Australia in autumn than indicated by the host models, as well as little change to the magnitude of the pro-jected declines in spring. Apart from Tasmania in winter, the decline of rainfall in southern Australia during the cool season remains a confident projection but the seasonal expression of change remains an ongoing research topic.
The subtropical ridge (STR) is the mean pressure ridge in the mid-latitudes, and is one of the key features affecting climate variability and change in southeast Australia. Changes to the STR and associated changes to rainfall in a warming climate are of strong interest, and the new Coupled Model Inter-comparison Project phase 5 (CMIP5) model archive provides new opportunities to examine this. Here we show that the STR is projected to strengthen and move pole-ward under global warming, contributing to reduced rainfall in the cool season in south-east Australia. This result is largely consistent among 35 models examined, and CMIP5 shows a greater increase in intensity relative to position than CMIP3 did. We show that the simulation of the STR in the CMIP5 is similar to that of the previous CMIP3 in many respects, including the underestimation of both the historical trends in the STR intensity and the correlation between inter-annual STR intensity and southeast Australian rainfall. These issues mean we still have reduced confidence in regional rainfall projections for southeast Australia and suggest that CMIP5 rainfall projections for this region in April to October may be underesti-mates.
Model evaluation is an important tool to help rate confidence in climate model simulations. This can add to the overall confidence assessment for future projections of the Australian climate. Additionally it can highlight significant model deficiencies that may affect the selection of a subset of models for use in impact assessment. Here we present results from an extensive model evaluation undertaken as part of the Natural Resource Management (NRM) Project in order to inform the newest set of climate change projections for Australia. The assessment covers mean climate skill over Australia as well as variability measures and teleconnections from up to 47 CMIP5 models and 23 CMIP3 models (for comparison where appropriate). Additionally, the skill in representing important climate features such as MJO, SAM, blocking and cut-off lows are also reviewed. Selected extremes are evaluated as well as simulations of two different types of downscaling simulations used within the NRM project. Finally, an attempt is made to synthesise this information in order to highlight a small group of CMIP5 models which show consistent deficiencies in representing the Australian climate and its features.
The projected drying of the extra-tropics under a warmer climate has large implications for natural systems and water security in southern Australia. The downscaling of global climate models can provide insight into regional patterns of rainfall change in the mid-latitudes in the typically wetter cool season. The comparison of statistical and dynamical downscaling model outputs reveals regions of consistent potential added value in the climate-change signal over the 21st century that are largely related to finer resolution. These differences include a stronger and more regionalised rainfall decrease on west coasts in response to a shift in westerly circulation and a different response further from the coast where other influences are important. These patterns have a plausible relationship with topography and regional drivers that are not resolved by coarse global models. However, the comparison of statistical and dynamical downscaling reveals where the method and the configuration of each method makes a difference to the projection. This is an important source of uncertainty for regional rainfall projections. In particular, the simulated change in atmospheric circulation over the century is different in the dynamical downscaling compared to the global climate model inputs, related in part to a different response to patterns of surface warming. The dynamical downscaling places the border between regions with rainfall increase and decrease further north in winter and spring compared to the global climate models and therefore has a different rainfall projection for southeast mainland Australia in winter and for Tasmania in spring.
This Report provides an assessment of observed climate change in Australia and its causes, and details projected future changes over the 21st century. This document, produced by CSIRO and the Australian Bureau of Meteorology, underpins extensive climate change projections for Australia provided as part of a larger package of products developed with funding from the Commonwealth Government’s Regional Natural Resource Management (NRM) Planning for Climate Change Fund. The projections are based on our understanding of the climate system, historical trends and model simulations of the climate response to global scenarios of greenhouse gas and aerosol emissions. Simulations come from the archive of global climate models (GCMs) developed by modelling groups from around the world through the Coupled Model Intercomparison Project phase 5 (CMIP5) which also underpins the science of the Fifth Assessment Report of the Intergovernmental Panel on Climate Change (IPCC).
Fire danger has increased in recent decades, and is projected to increase further with global warming. We assessed the regional changes in fire danger that are projected to occur in Tasmania through to 2100 under a high emissions scenario. In contrast with previous continental–scale studies which show little change in Tasmanian fire danger, our results indicate an overall increase in fire danger, especially in spring, with more days per year likely to require total fire bans. This increase in fire danger will have social and political implications. Projected changes to extreme weather events More extreme events have been recorded over the latter half of the 20th century, coinciding with changes to climate over that time. Higher maximum and minimum temperatures, more hot days and fewer cold days, and more intense rainfall events have all been observed and are expected to increase with future climate change. The Intergovernmental Panel on Climate Change (IPCC), in its fifth assessment report (AR5), concluded with high confidence that climate change would lead to increases in the number of days with very high and extreme fire weather. The greatest increase is expected in regions where fire is not limited by the availability of fuel, such as in southern Australia. The IPCC identified increased fire weather, along with complex impacts on vegetation and biodiversity changes, as a key risk from climate change to people, property, infrastructure, ecosystems and native species. The aim of the Future Fire Danger Project was to understand the changing risk of fire danger in Tasmania. This study builds on the scientific knowledge, fine–resolution climate simulations and the communication network generated by Climate Futures for Tasmania. The study used observations and models to examine changes in fire danger over recent decades, then used climate model projections, including high–resolution simulations for Tasmania, to assess projected changes of fire risk in the future.
Climate warming has large implications for rainfall patterns, and identifying the most plausible pattern of rainfall change over the next century among various model projections would be valuable for future planning. The spatial pattern of projected sea surface temperature change has a key influence on rainfall changes in the tropical Pacific Ocean. Here it is shown that simple indices of the size of the equatorial peak in the spatial pattern of warming and to a lesser extent the hemispheric asymmetry in warming are useful for classifying the surface temperature change in different models from phase 5 of the Coupled Model Intercomparison Project (CMIP5). Models with a more pronounced equatorial warming show a fairly distinct rainfall response compared to those with more uniform warming, including a greater "warmer-get-wetter" or dynamical response, whereby rainfall increases follow the surface warming anomaly. Models with a more uniform warming pattern project a smaller rainfall increase at the equator and a rainfall increase in the southern tropical Pacific, a pattern that is distinct from the multimodel mean of CMIP5. Thus, the magnitude of enhanced equatorial warming and to some extent the hemispheric asymmetry in warming provides a useful framework for constraining rainfall projections. While there is not a simple emergent constraint for enhanced equatorial warming in models in terms of past trends or bias in the current climate, further understanding of the various feedbacks involved in these features could lead to a useful constraint of rainfall for the Pacific region.
A set of 27 global climate models from the Coupled Model Inter-comparison Project Phase 5 (CMIP5) ensemble are assessed for their performance for the purpose of making future climate projection studies in the western tropical Pacific and differences to Coupled Model Inter-comparison Project Phase 3 (CMIP3) are assessed. The CMIP5 models show some improvements upon CMIP3 in the simulation of the climate in the western tropical Pacific in the late 20th century. There are fewer CMIP5 models with very poor skill scores than in CMIP3 for some measures and a small group of the well-performing models in CMIP5 have lower biases than in an equivalent group from CMIP3. These best-performing models could be particularly informative for studying certain climate sensitivities and feedbacks in the region. There is evidence to reject one model as unsuitable for making regional climate projections in the region, and another two models unsuitable for analysis of the South Pacific Convergence Zone (SPCZ). However, while there have been improvements, many of the systematic model biases in the mean climate in CMIP3 are also present in the CMIP5 models. They are primarily related to the shape of the transition between the Indo-Pacific warm pool and equatorial cold tongue, and the associated biases in the position and orientation of the SPCZ and Inter-Tropical Convergence Zone, as well as in the spatial pattern, variability and teleconnections of the West Pacific monsoon, and the simulation of El Niño Southern Oscillation. Overall, the results show that careful interpretation and consideration of biases is required when using CMIP5 outputs for generating regional climate projections for the western tropical Pacific, particularly at the country scale, just as there was with CMIP3.
Climate projections are essential for studying ecological responses to climate change, and their use is now common in ecology. However, the lack of integration between ecology and climate science has restricted understanding of the available climate data and their appropriate use. We provide an overview of climate model outputs and issues that need to be considered when applying projections of future climate in ecological studies. We outline the strengths and weaknesses of available climate projections, the uncertainty associated with future projections at different spatial and temporal scales, the differences between available downscaling methods (dynamical, statistical downscaling, and simple scaling of global circulation model output), and the implications these have for ecological models. We describe some of the changes in the Fifth Assessment Report (AR5) of the Intergovernmental Panel on Climate Change (IPCC), including the new representative concentration pathways. We highlight some of the challenges in using model projections in ecological studies and suggest how to effectively address them.For further resources related to this article, please visit the WIREs website.Conflict of interest: The authors have declared no conflicts of interest for this article.
We describe the method and performance of a bias-correction applied to high-resolution (˜10 km) simulations from a stretched-grid Regional Climate Model (RCM) over Tasmania, Australia. The bias-correction is a quantile mapping of empirical cumulative frequency distributions. Corrections are applied at a daily time step to five variables: rainfall, potential evaporation (PE), solar radiation, maximum temperature and minimum temperature. Corrections are calculated independently for each season.We show that quantile mapping of empirical distributions can be highly effective in correcting biases in RCM outputs. Cross-validation shows biases are effectively reduced across the range of cumulative frequency distributions, with few exceptions. The bias-correction is not as effective at correcting biases for values at or near zero (e.g. in rainfall simulations), although even here the bias-correction improves biases evident in the uncorrected simulations. In addition, the bias-correction improves frequency characteristics of variables such as the number of rain days.We use a detrending technique to apply the bias-correction to 140-year time series of RCM variables. We show that the bias-correction effectively preserves long-term changes (e.g. to the mean and variance) to variables projected by the uncorrected RCM simulations. Correlations between key variables are also largely preserved, thus the bias-corrected outputs reflect the dynamics of the underlying RCM. However, the bias-corrected simulations still exhibit some of the deficiencies of the RCM simulations, e.g. the tendency to underestimate the magnitude and duration of large, multi-day rain events, and the tendency to underestimate the duration of dry spells.The bias-corrected simulations for six downscaled GCMs for the A2 and B1 emissions scenarios are available to researchers from http://www.tpac.org.au.
Projected changes to the global climate system have great implications for the incidence of large infrequent fires in many regions. Here we examine the synoptic-scale and local-scale influences on the incidence of extreme fire weather days and consider projections of the large-scale mean climate to explore future fire weather projections. We focus on a case study region with periodic extreme fire dangers; southeast Tasmania, Australia. We compare the performance of a dynamically downscaled regional climate model with Global Climate Model outputs as a tool for examining the local-scale influences while accounting for high regional variability. Many of the worst fires in Tasmania and the southeast Australian region are associated with deep cold fronts and strong prefrontal winds. The downscaled simulations reproduce this synoptic type with greater fidelity than a typical global climate model. The incidence of systems in this category is projected to increase through the century under a high emission scenario, driven mainly by an increase in the temperature of air masses, with little change in the strength of the systems. The regional climate model projected increase in frequency is smaller than for the global climate models used as input, with a large model range and natural variability. We also demonstrate how a blocking Foehn effect and topographic channelling contributed to the extreme conditions during an extreme fire weather day in Tasmania in January 2013. Effects such as these are likely to contribute to high fire danger throughout the century. Regional climate models are useful tools that enable various meteorological drivers of fire danger to be considered in projections of future fire danger.
Daily values of McArthur Forest Fire Danger Index were generated at ~10-km resolution over Tasmania, Australia, from six dynamically downscaled CMIP3 climate models for 1961–2100, using a high (A2) emissions scenario. Multi-model mean fire danger validated well against observations for 2002–2012, with 99th percentile fire dangers having the same distribution and largely similar values to those observed over the same time. Model projections showed a broad increase in fire danger across Tasmania, but with substantial regional variation – the increase was smaller in western Tasmania (district mean cumulative fire danger increasing at 1.07 per year) compared with parts of the east (1.79 per year), for example. There was also noticeable seasonal variation, with little change occurring in autumn, but a steady increase in area subject to springtime 99th percentile fire danger from 6% in 1961–1980 to 21% by 2081–2100, again consistent with observations. In general, annually accumulated fire danger behaved similarly. Regional mean sea level pressure patterns resembled observed patterns often associated with days of dangerous fire weather. Days of elevated fire danger displaying these patterns increased in frequency during the simulated twenty-first century: in south-east Tasmania, for example, the number of such events detected rose from 101 (across all models) in 1961–1980 to 169 by 2081–2100. Correspondence of model output with observations and the regional detail available suggest that these dynamically downscaled model data are useful projections of future fire danger for landscape managers and the community.
Coupled General Circulation Models simulate broad-scale climate patterns and are standard tools for understanding potential changes to the climate system. Climate change projection information is required at the local scale for adaptation planning. Model skill can be limited at this scale due to model biases and uncertainties. Pacific Islands are one such example, particularly Nauru which lies at the intersection of the South Pacific Convergence Zone (SPCZ), the Western Pacific Monsoon, the oceanic warm pool and the ENSO zone of influence. In light of such constraints, a qualitative climate projection approach is presented whereby projections are based on expected changes to the broad-scale climate features, drawing on a review of current literature and new research. Nauru sits in the region subject to seasonal and interannual migrations of the SPCZ. Climate models simulate an overly zonal SPCZ with Nauru lying too close to the dry cold tongue zone. This sets an erroneous base state for Nauru by locating it in a very different climate zone. The climate model changes in the SPCZ can be used to argue how the climate of Nauru would change were it and the SPCZ correctly collocated. Similar approaches are applied to the set of climate features influencing Nauru to generate modified climate projections.
The ability of an ensemble of six GCMs, downscaled to a 0.1° lat/lon grid using the Conformal Cubic Atmospheric Model over Tasmania, Australia, to simulate observed extreme temperature and precipitation climatologies and statewide trends is assessed for 1961–2009 using a suite of extreme indices. The downscaled simulations have high skill in reproducing extreme temperatures, with the majority of models reproducing the statewide averaged sign and magnitude of recent observed trends of increasing warm days and warm nights and decreasing frost days. The warm spell duration index is however underestimated, while variance is generally overrepresented in the extreme temperature range across most regions. The simulations show a lower level of skill in modelling the amplitude of the extreme precipitation indices such as very wet days, but simulate the observed spatial patterns and variability. In general, simulations of dry extreme precipitation indices are underestimated in dryer areas and wet extremes indices are underestimated in wetter areas. Using two SRES emissions scenarios, the simulations indicate a significant increase in warm nights compared to a slightly more moderate increase in warm days, and an increase in maximum 1- and 5-day precipitation intensities interspersed with longer consecutive dry spells across Tasmania during the twenty-first century.
 In this study we develop methods for dynamically downscaling output from six general circulation models (GCMs) for two emissions scenarios using a variable-resolution atmospheric climate model. The use of multiple GCMs and emissions scenarios gives an estimate of model range in projected changes to the mean climate across the region. By modeling the atmosphere at a very fine scale, the simulations capture processes that are important to regional weather and climate at length scales that are subgrid scale for the host GCM. We find that with a multistaged process of increased resolution and the application of bias adjustment methods, the ability of the simulation to reproduce observed conditions improves, with greater than 95% of the spatial variance explained for temperature and about 90% for rainfall. Furthermore, downscaling leads to a significant improvement for the temporal distribution of variables commonly used in applied analyses, reproducing seasonal variability in line with observations. This seasonal signal is not evident in the GCMs. This multistaged approach allows progressive improvement in the skill of the simulations in order to resolve key processes over the region with quantifiable improvements in the correlations with observations.
The great majority of dendroclimatological work in Australia has thus far relied on ring-width chronologies only. We report novel results from a pilot study that show the potential to develop density-based climatically sensitive chronologies from two long-lived conifers endemic to Tasmania: Pencil Pine and Celery Top Pine. Cross-dating of average ring density profiles within each of the two sites examined was comparable with the better replicated ring-width chronologies from the sites. Cross-dating potential for maximum density was also indicated. Correlations between density and climate for both species were stronger and more persistent across a window of several months than correlations between ring width and climate. These stronger correlations suggest that temperature reconstructions based on average density may be possible. The ability to develop high resolution temperature-sensitive chronologies would allow for spatial comparisons across regions such as Tasmania that are affected by multiple broad-scale climate systems. A particularly novel result was the finding that maximum density was significantly related to stream-flow at the end of the growing season. Further work is required to assess the potential to reconstruct temperature, and to reconstruct stream-flow for important Tasmanian catchments over the past 500–800 years.
Cutoff lows are an important source of rainfall in the mid-latitudes that climate models need to simulate accurately to give confidence in climate projections for rainfall. Coarse-scale general circulation models used for climate studies show some notable biases and deficiencies in the simulation of cutoff lows in the Australian region and important aspects of the broader circulation such as atmospheric blocking and the split jet structure observed over Australia. The regional climate model conformal cubic atmospheric model or CCAM gives an improvement in some aspects of the simulation of cutoffs in the Australian region, including a reduction in the underestimate of the frequency of cutoff days by more than 15 % compared to a typical GCM. This improvement is due at least in part to substantially higher resolution. However, biases in the simulation of the broader circulation, blocking and the split jet structure are still present. In particular, a northward bias in the central latitude of cutoff lows creates a substantial underestimate of the associated rainfall over Tasmania in April to October. Also, the regional climate model produces a significant north–south distortion of the vertical profile of cutoff lows, with the largest distortion occurring in the cooler months that was not apparent in GCM simulations. The remaining biases and presence of new biases demonstrates that increased horizontal resolution is not the only requirement in the reliable simulation of cutoff lows in climate models. Notwithstanding the biases in their simulation, the regional climate model projections show some responses to climate warming that are noteworthy. The projections indicate a marked closing of the split jet in winter. This change is associated with changes to atmospheric blocking in the Tasman Sea, which decreases in June to November (by up to 7.9 m s−1), and increases in December to May. The projections also show a reduction in the number of annual cutoff days by 67 % over the century, together with an increase in their intensity, and these changes are strongest in spring and summer.
Coupled ocean–atmosphere general circulation models (GCMs) lack sufficient resolution to model the regional detail of changes to mean circulation and rainfall with projected climate warming. In this paper, changes in mean circulation and rainfall in GCMs are compared to those in a variable resolution regional climate model, the Conformal Cubic Atmospheric Model (CCAM), under a high greenhouse gas emissions scenario. The study site is Tasmania, Australia, which is positioned within the mid-latitude westerlies of the southern hemisphere. CCAM projects a different response in mean sea level pressure and mid-latitude westerly circulation to climate warming to the GCMs used as input, and shows greater regional detail of the boundaries between regions of increasing and decreasing rainfall. Changes in mean circulation dominate the mean rainfall response in western Tasmania, whereas changes to rainfall in the East Coast are less related to mean circulation changes. CCAM projects an amplification of the dominant westerly circulation over Tasmania and this amplifies the seasonal cycle of wet winters and dry summers in the west. There is a larger change in the strength than in the incidence of westerly circulation and rainfall events. We propose the regional climate model displays a more sensitive atmospheric response to the different rates of warming of land and sea than the GCMs as input. The regional variation in these results highlight the need for dynamical downscaling of coupled general circulation models to finely resolve the influence of mean circulation and boundaries between regions of projected increases and decreases in rainfall.
Changes to streamflows caused by climate change may have major impacts on the management of water for hydro-electricity generation and agriculture in Tasmania, Australia. We describe changes to Tasmanian surface water availability from 1961–1990 to 2070–2099 using high-resolution simulations. Six fine-scale (∼10 km<sup>2</sup>) simulations of daily rainfall and potential evapotranspiration are generated with the CSIRO Conformal Cubic Atmospheric Model (CCAM), a variable-resolution regional climate model (RCM). These variables are bias-corrected with quantile mapping and used as direct inputs to the hydrological models AWBM, IHACRES, Sacramento, SIMHYD and SMAR-G to project streamflows. The performance of the hydrological models is assessed against 86 streamflow gauges across Tasmania. The SIMHYD model is the least biased (median bias = −3%) while IHACRES has the largest bias (median bias = −22%). We find the hydrological models that best simulate observed streamflows produce similar streamflow projections. There is much greater variation in projections between RCM simulations than between hydrological models. Marked decreases of up to 30% are projected for annual runoff in central Tasmania, while runoff is generally projected to increase in the east. Daily streamflow variability is projected to increase for most of Tasmania, consistent with increases in rainfall intensity. Inter-annual variability of streamflows is projected to increase across most of Tasmania. This is the first major Australian study to use high-resolution bias-corrected rainfall and potential evapotranspiration projections as direct inputs to hydrological models. Our study shows that these simulations are capable of producing realistic streamflows, allowing for increased confidence in assessing future changes to surface water variability.
Changes to streamflows caused by climate change may have major impacts on the management of water for hydro-electric generation and agriculture in Tasmania, Australia. We present high-resolution projections of Tasmanian surface water availability between 1961–1990 and 2070–2099. Six fine-scale (10 km) simulations of daily rainfall and potential evapotranspiration are generated with the CSIRO Conformal Cubic Atmospheric Model (CCAM), a variable-resolution regional climate model (RCM). These variables are bias-corrected with quantile mapping and used as direct inputs to the hydrological models AWBM, IHACRES, Sacramento, SIMHYD and SMAR-G to project streamflows. The performance of the hydrological models is assessed against 86 streamflow gauges across Tasmania. The SIMHYD model is the least biased (median bias = −3%) while IHACRES has the largest bias (median bias = −22%). We find the hydrological models that best simulate observed streamflows produce similar streamflow projections. In contrast, the poorly performing IHACRES model amplifies changes more than the other hydrological models. There is much more variation in projections between RCM simulations than between hydrological models. This shows that it is more important to consider the range of RCM simulations than the range of hydrological models used here to adequately describe uncertainty in the projections. We use the SIMHYD model to describe future changes to streamflow in eight rivers. Changes to streamflows are projected to vary by region. Marked decreases of up to 30% are projected for annual runoff in central Tasmania, while runoff is generally projected to increase in the east. Daily streamflow variability is projected to increase for most of Tasmania, consistent with increases in rainfall intensity. Inter-annual variability of streamflows is projected to increase across most of Tasmania. This is the first major Australian study to use high-resolution bias-corrected rainfall and potential evapotranspiration projections as direct inputs to hydrological models. Our study shows that these simulations are capable of producing realistic streamflows, allowing for increased confidence in assessing future changes to surface water variability.
Many key landscape processes for human land use and natural ecosystems are regulated by surface climatic conditions, including an important influence from incoming shortwave radiation. Surface radiation has a key influence on landscape ecosystems through effects on surface thermal regimes, photosynthesis and evapotranspiration, which in turn affects soil water. The radiation environment varies on all spatial and temporal scales, from site to landscape. Incoming shortwave radiation is a key component of the global radiation budget in climate models and is examined in studies of the climate system. However, gauging the impact of changes to the climate on surface radiation at a scale relevant to landscape processes is not commonly done, and requires new methods and focus. For much of the Australian continent, surface radiation can be readily and accurately modeled to fine scales as a function of time of year, latitude, and a broad estimate of atmospheric attenuation. However in rugged terrain, surface radiation is significantly modified by surface slope, aspect, and topographic shading (a stark example is deep ravines). Here we examine the local radiation environment in Tasmania and the Australian Alps, where topography effects are important. The Solar Radiation Attributes program (SRAD) uses a digital elevation model (DEM) and a set of ancillary climatic and environmental parameters to estimate topographically modified surface radiation. We present a set of regional climate model outputs run at 10-20 km resolution coupled directly to the SRAD system with a 250 m DEM to describe changes to the local radiation environment under projected climate change scenarios. This process illustrates the effect of projected climate change on radiation at the surface, where many landscapes processes occur. We propose this method as a useful tool to account for local attributes and improve understanding of climate change impacts at a scale meaningful for natural ecosystems.
Climate change is a global phenomenon whose impacts will be most keenly felt at a local level. General Circulation Models (GCMs) provide the best estimates for assessing potential changes to our climate on a global. However GCMs do not provide much direction for decision-makers (like local governments, state governments or industries) in deciding on appropriate adaptive responses to the local effects of climate change. Such decision makers often require output from conceptual models, such as biophysical or hydrological models, or knowledge of changes to specific operational information or the likelihood of extreme events; GCMs are largely not suitable for this purpose. Biophysical and hydrological models require inputs such as rainfall, minimum and maximum temperature, evaporation and solar radiation, often in daily timesteps. Future climate input for these models are commonly created by perturbing historical datasets with climate anomalies based on output from climate models. This approach means that the shape of the probability distribution function (pdf) in observed data is maintained in the future data. This technique ignores the likelihood that climate change will result in changes to the synoptic drivers of climate variables which in turn result in changes to the pdfs of these variables. An alternative approach is to use high-resolution regional climate models to dynamically downscale GCM projections. These regional climate models produce output that is at appropriate spatial and temporal resolution so that it can be used as input into conceptual models. Unfortunately all climate models contain biases and parameterisations that render their output unsuitable for direct use into conceptual models. Conceptual models are calibrated against observations, and so a bias in a climate model needs only to be relatively small to have a significant effect on output metrics that are calibrated against observed datasets. In the Climate Futures for Tasmania project, climate variables were bias-adjusted against Australian Water Availability Project data over the period 1961-2007. Daily rainfall, minimum and maximum temperature, evaporation and solar radiation were adjusted using differences in one percentile bins for six models and four seasons for each 0.1° grid cell. The adjustments (1961-2007) were then applied to the climate model variables over the period 1961-2100. This technique assumes the 1961-2007 adjustments are maintained into the future. The technique aims to preserve the statistical characteristics of the original model output (the pdf) without changing the trends and variability (the climate change signature) of the data. Bias-adjusted model output can then be input directly into conceptual models. We present the bias-adjustment method and example results where bias-adjusted climate model output has been used as input into biophysical and hydrological models that have been calibrated against an observational record.
Translating meteorological projections from global climate models (GCMs) into useful information for water managers and industry involves addressing a combination of technical and communication challenges. The Climate Futures for Tasmania project has projected water yield in Tasmania, Australia to 2100. This paper describes how the Climate Futures for Tasmania project successfully translated climate projections into useable information for water managers and industry. From its inception, the Climate Futures for Tasmania project has maintained a dialogue with the two major water managers in the Tasmania: the Department of Primary Industry, Parks, Water and Environment (DPIPWE), the government body with statutory responsibility for water management in Tasmania, and Hydro Tasmania, Australia's largest hydropower generator. Frequent discussions with these two organisations directed the technical research into future water yields. Tasmania is a difficult region for climate change-hydrology studies. Tasmania's complex rainfall patterns are not replicated by GCMs, and hence GCMs produce information that is too general to be useful to Tasmanian water managers. To overcome this problem, GCM projections were downscaled to a finer spatial resolution. Downscaling greatly improved the spatial correlation of modelled rainfall with observations, and accordingly the usefulness of the projections to water managers. The downscaled climate projections were fed into hydrological models to produce projections of streamflow. The hydrological modelling involved two steps: 1. Runoff modelling - calculating statewide, gridded natural runoff at a resolution of 0.05 degrees 2. River system modelling - aggregating the gridded natural runoff to 65 Tasmanian river basins and then accounting for human activities in rivers including dams, irrigation and hydropower generation. Splitting the hydrological modelling into these two steps allows the effects of climate and human activity to be differentiated. This is important for water managers, as it separates elements outside of their control (climate) from those under their control (e.g. irrigation). While changes in human water use are not considered in the Climate Futures for Tasmania study, Tasmanian water managers will be able to adapt the river systems models to quantify changes in water management policies. Finally, projections of runoff were adapted to run through the Hydro Tasmania Systems model Temsim. Temsim uses hydrological inputs in conjunction with projected power demand and energy prices to simulate the Hydro Tasmania power generation system. The Temsim runs translate CFT climate projections into metrics such as storage levels, power generation, and revenue - metrics that can inform the future operation of the Hydro Tasmania system. The result is climate information tailored to the needs of water managers and industry, ensuring the research will be understandable and useable. This paper presents the communication strategy implemented by Climate Futures for Tasmania, and provides a case study of how interaction with government and industry directed the technical research.
Understanding and modelling Tasmanian rainfall variability and making future projections of Tasmanian rainfall are challenging tasks. Tasmania has spatially and temporally complex rainfall patterns. Rainfall variability is influenced by a complex suite of remote drivers and these influences vary by season. The Climate Futures for Tasmania high-resolution model simulations project small changes to annual rainfall averaged over Tasmania, but larger changes to the spatial patterns and seasonality of rainfall. A case study of changes to summer rainfall under a high greenhouse gas emission scenario is shown here. The projected summer decrease in rainfall in the western rainfall region is consistent with the southerly movement and intensification of the subtropical ridge as well as an enhancement of the high phase of the Southern Annular Mode. The increase along the east coastal strip is consistent with an increase in blocking in the Tasman Sea as well as an increase in sea surface temperature, relative humidity and convective rainfall. We propose that projections of rainfall for places like Tasmania are strengthened through dynamical downscaling and also the analysis of the rainfall mechanisms within the model at all length scales.
Coupled Ocean-Atmosphere General Circulation Models (GCMs) provide the best estimates for assessing potential changes to our climate on a global scale out to the end of this century. Because coupled GCMs have a fairly coarse resolution they do not provide a detailed picture of climate (and climate change) at the local scale. Tasmania, due to its diverse geography and range of climate over a small area is a particularly difficult region for drawing conclusions regarding climate change when relying solely on GCMs. The foundation of the Climate Futures for Tasmania project is to take the output produced by multiple GCMs, using multiple climate change scenarios, and use this output as input into the Conformal Cubic Atmospheric Model (CCAM) to downscale the GCM output. CCAM is a full atmospheric global general circulation model, formulated using a conformal-cubic grid that covers the globe but can be stretched to provide higher resolution in the area of interest (Tasmania). By modelling the atmosphere at a much finer scale than is possible using a coupled GCM we can more accurately capture the processes that drive Tasmania's weather/climate, and thus can more clearly answer the question of how Tasmania's climate will change in the future. We present results that show the improvements in capturing the local-scale climate and climate drivers that can be achieved through downscaling, when compared to a gridded observational data set. The underlying assumption of this work is that a better simulated current climatology will also produce a more credible climate change signal.
The ability of regional dynamically-downscaled general circulation models (GCMs) to assess changes to future extreme climatic events was investigated by comparing hindcast model outputs with observations. Projections were generated on a 0.1 grid across Tasmania using the CSIRO Conformal Cubic Atmospheric Model (CCAM). Two future SRES emission scenarios (A2 and B1) and multiple boundary conditions from GCMs were used for the period 1961-2100. A bias-adjustment procedure was employed to spatially correct extreme magnitudes. Events were fitted to a Generalised Pareto Distribution (GPD) using an automated threshold selection procedure developed for gridded precipitation datasets. Estimates of precipitation average recurrence intervals (ARIs) were calculated using extreme value analysis and compared to gridded observations. Spatial patterns were found in gridded precipitation extremes that closely matched observations. Projections of future changes to precipitation extremes were found to vary spatially between models, correlating with projected changes to regional climate drivers. Results demonstrate that dynamical downscaling captures regional climate variability (particularly relevant for precipitation) and displays significant ability in modelling future changes to the intensity, magnitude and frequency of extreme events at the local scale for use in adaptation and emergency planning applications.
Modelling future runoff by running meteorological projections from global climate models (GCMs) directly through hydrological models presents considerable technical challenges, but promises several advantages over the so-called `perturbation method'. The Climate Futures for Tasmania project has projected water yield in Tasmania, Australia to 2100. This paper describes how the Climate Futures for Tasmania project used dynamically downscaled climate projections directly in hydrological models to produce useable information for water managers and industry. Tasmania is a difficult region for climate change hydrology studies. Tasmanian rainfall is generated by complex regional weather systems such as atmospheric blocking that are not always well-represented in GCM-scale projections. Further, the spatial resolution of GCMs is too coarse to represent the complex distribution of Tasmanian rainfall. Rainfall changes caused by changes in these regional weather systems may not be predicted by GCMs. Previous studies of climate change impacts on Tasmanian rivers have used the `perturbation method', where historical rainfall and evaporation data are modified to reflect changes predicted by GCMs. In this method rainfall events occur exactly as often as in the historical record - only the magnitude of events changes. This can mask long-term effects on runoff caused by changes in the timing or duration of rainfall events due to climate change. We avoided this problem by dynamically downscaling six GCMs with the regional climate model CCAM to a spatial resolution of 0.1 degrees under the A2 SRES emissions scenario. Dynamical downscaling is computing-intensive, but can simulate changes to rain-bearing weather systems (e.g. increases in convective storms). Downscaled hindcasts generally showed excellent spatial and temporal agreement with climate observations. However, some spatial biases were still evident. To account for these biases, modelled rainfall and evaporation were bias-adjusted by percentile to observations for 1961-2007. A premise of bias-adjustment is that discrepancies between observed and modelled data are constant through time. A leave-one-decade-out test was devised to demonstrate that the biases were constant through time. Existing statewide hydrologic models were adapted to accept the bias-adjusted dynamically downscaled GCM projections. Five runoff models were available: AWBM, Ihacres, Sacramento, Simhyd, and SMARG. Each of these models reacts differently to climate inputs. The uncertainty in projected changes to runoff due to the choice of hydrological model was assessed for one GCM. Downscaled projections from all six GCMs were run through the Simhyd model to produce runoff at a 0.05 degree statewide grid. Runoff was aggregated to river basins, and human activities such as irrigation and hydropower generation were accounted for. Model hindcasts of river flows showed very good agreement with observed flows. Impacts on future runoff were highly spatially heterogeneous, demonstrating the value of high resolution downscaling for hydrologic projections. There was some evidence that changes in rain-bearing systems - such as the incidence of convective storms - will influence water yields in certain catchments. The result is one of the most detailed regional climate change hydrology studies in Australia. The hydrologic projections have been tailored to the needs of water managers and industry, ensuring the research will be understandable and useable.
Executive Summary Climate Futures for Tasmania has produced sophisticated hydrological projections for Tasmania to 2100: Climate Futures for Tasmania has combined state-of-the-art regional climate modelling and hydrological models to project future catchment yields for Tasmania. The project has produced runoff projections from an ensemble of six dynamically downscaled global climate models and five runoff models to 2100. River flows were projected for more than 1900 subcatchments in 78 river catchments that cover more than 70% of the state by area. Only changes caused by increases in greenhouse gases were considered and changes in land use or water use were not investigated. The future operations of Tasmania’s hydro-electric system and 14 major irrigation storages were also simulated to 2100. Runoff across Tasmania is projected to increase slightly by 2100: The statewide annual runoff shows significant decadal variations throughout the 21st century. On average, statewide annual runoff is likely to increase by 559 GL (1.1%) by 2100. Individual climate models project increases in statewide annual runoff of up to 7085 GL (14.6%) or decreases up to 2110 GL (4.2%) by 2100. In a changing climate, patterns of runoff will differ from the current climate: Changes to runoff by 2100 vary between different regions. These patterns of change are more important than the relatively small statewide changes. Annual runoff is likely to decrease significantly in Tasmania’s central highlands, with 30% less runoff in some areas. On average, annual runoff in eastern areas of the state are generally projected to increase, particularly in the lowlands. Runoff in the lower Derwent Valley is likely to increase, with increases of more than 50% in some areas. Annual runoff is likely to increase in the lower South Esk River and lower Macquarie River catchments, increasing by more than 15% in most areas. In a changing climate, seasonal runoff is likely to differ markedly from the current climate: Marked seasonal changes to runoff are likely to occur over the coming century. Annual runoff on the west coast is not projected to change greatly by 2100, however west coast runoff is likely to increase in winter and decrease markedly in summer and autumn. Increases in runoff in the lower South Esk River and the lower Macquarie River are projected to be greatest in winter, increasing by more than 15% in most areas. In a changing climate, some river flows will decrease and some will increase by 2100: Of the 78 rivers modelled, on average 32 are projected to have changes to mean annual flows of more than ±10% by 2100. Changes of this size may have implications for water management and infrastructure development. On average, 28 of the 78 rivers modelled are projected to have decreased flows by 2100, while 50 rivers are projected to have increased flows. However, in one climate projection as many as 55 of 78 rivers have decreased flows, while in another climate projection 77 of 78 rivers will have increased flows. Large irrigation storages fed from runoff from the central highlands are likely to have reduced inflows by 2100: The likely reduction in runoff to the central highlands will mean reduced inflows to irrigation storages. For example, the mean inflows to Lake Crescent/Sorell and Meander Dam are projected to fall by 20% and 13% respectively. The driest model projections indicated that inflows to Lake Crescent could fall by up to 48%, while inflows to Meander Dam could fall by up to 21% by 2100. Declines in inflows to these storages could affect the reliability of supply to downstream water users who rely on releases from these storages. Large irrigation storages supplying the Macquarie River and Coal River catchments are projected to experience increased inflows by 2100: The irrigation storages Lake Leake and Tooms Lake in the Macquarie River catchment are both projected to have increased inflows. Mean inflows to Lake Leake and Tooms Lake are projected to increase by 23% and 25% respectively. Mean inflows to Craigbourne Dam in the Coal River catchment are projected to rise by 24%, although one climate model projected an increase of 83%. Climate change is likely to reduce inflows to catchments used for hydro-electricity generation throughout the 21st century, and this could reduce the power-generation capacity of the Hydro Tasmania hydro-electric system: Reduced inflows to catchments supplying hydro-electric power stations could lead to a gradual and continuous reduction in overall power generation capacity throughout the 21st century. Power generation capacity could also be reduced by seasonal and spatial changes to runoff. Declines to inflows in the central plateau catchments are likely to have a marked impact on power generation, because these catchments feed a large capacity, highly efficient power station. More strongly seasonally delineated inflows in the western catchments are likely to result in lost power generation in run-of-river hydro-electric schemes. Some catchments experience both increased and decreased runoff in different parts of the catchment: The fine-resolution modelling allows projections of changes within catchments. Catchments that are fed by the central highlands and flow east generally experience decreased runoff in the upper part of the catchment. These decreases are partially offset by increased runoff in the lower parts. For example, the Derwent catchment is projected to experience an average decline in flows of 5.2% by 2100. Here the increased runoff in the lower Derwent catchment is outweighed by the larger decrease in runoff in the upper parts of the catchment. Conversely, the decrease in runoff in the upper part of the Clyde River catchment is offset by the larger increase in runoff in the lower Clyde catchment, and the Clyde River catchment is projected to experience an average increase in flows of 17%.
The Climate Futures for Tasmania (CFT) project has undertaken a series of dynamical downscaling simulations using CSIRO's Conformal Cubic Atmospheric Model (CCAM). These simulations provide high resolution (10 km) output over the Australian state of Tasmania. The simulations use as boundary conditions output from six GCMs and two SRES emission scenarios, giving a total of twelve runs. By modeling the atmosphere at a much finer scale than is possible using a coupled GCM we can more accurately capture the processes that operate on Tasmania's weather/climate at the regional level and thus can more clearly answer the question of how Tasmania's climate will change in the future. We present results that show the improvements in resolving the local-scale climate and climate drivers with increasing resolution that can be achieved through downscaling, when compared to a gridded observational data set (AWAP). Changes in rainfall patterns are one of the key uncertainties surrounding climate change. The spatial pattern of rainfall in the 1961 to 1990 climate in the GCMs has no correlation with the observed rainfall climate over Tasmania, however the CFT downscaled simulations have a correlation of 0.69 with the observed annual rainfall over this period. For temperature the correlation between the CFT simulations and AWAP data is 0.99. A feature of the projections is the increased summer and autumn rainfall along Tasmania's east coast, in contrast to the current (15 year) dry spell that has seen drying in the east and wetting in the west. This seasonal increase in rainfall is not present in global climate model projections, which all predict a decreasing rainfall trend across Tasmania. We demonstrate that the increased rainfall over the east coast is due to a localised coupled ocean-atmosphere response. The high pressure belt moves southward and increases in strength, deflecting the dominant westerly winds further south. This wind increase spins up the South Pacific Gyre, causing the East Australian Current to extend southwards resulting in an anomaly of the mean sea level pressure that enhances the moisture flux over the east coast of Tasmania. This coupled ocean-atmosphere response for rainfall is an example of large scale climate change driving a local change in rainfall through increased frequency of low pressure systems over the Tasman Sea.
Tasmania is a challenging and rigorous test case for studies attempting to understand and project changes to rainfall. Tasmania is the island to the southeast of Australia, positioned in the roaring 40s of the Southern Ocean with a temperate maritime climate. Tasmania is topographically complex featuring rugged mountain ranges, a high plateau and lowland plains. Tasmanian rainfall distribution and variability is spatially complex, varying from high rainfall (>3000 mm) in a strong seasonal cycle near the west coast, to low rainfall (~600 mm) with no seasonal cycle on the east coast. A complex suite of large-scale climatic features drives rainfall in Tasmania. The effects of these drivers vary with location and season. Global-scale modelling indicates that Tasmania lies on the boundary between zones of increasing and decreasing rainfall into the future. We have produced fine-scale (~10 km) model projections of Tasmanian climate to 2100 that retain global-scale features present in global climate models using a process of dynamical downscaling. We use these model projections to examine changes to rainfall in Tasmania to 2100. Whilst annual rainfall for the whole state shows no marked changes to 2100 in the projections, changes for the regional districts and for the four seasons are much larger. These changes include a steady decrease in rainfall in the highlands, an increase in summer and autumn rain on the east coast and a significant change in the seasonality of rainfall in the west coast region after 2050. The use of fine-scale dynamical downscaling modelling allows us to examine projected changes at individual districts within the state, and examine the relevant mechanisms involved at all spatial scales. This includes processes at the hemispheric scale relevant to the region, such as dominant pressure patterns, Hadley Circulation, and the Southern Annular Mode (SAM). It also includes processes at the regional scale such as atmospheric blocking, significant synoptic events (such as cutoff lows), and processes at the fine scale such as the effect of topography. For the case of the west coast, the projected rainfall trends can be linked to the plausible continuation of current trends to the Hadley Circulation, the high phase of SAM and atmospheric blocking that alter the dominant westerly flow. In contrast, changes to rainfall near the east coast are dependent on a change in the frequency and character of low pressure systems in the in the Tasman Sea resulting in a mean low pressure anomaly driving an onshore circulation.
Evaluation and analysis of synoptic climatology is useful for quantifying the uncertainties in the simulation of rainfall processes by climate models, and then to determine the drivers behind projected changes to rainfall. The frequency of different rain-bearing system types was examined in fine-scale dynamically downscaled global climate model (GCM) simulations using software that automates expert knowledge of these phenomena. The incidence and rain produced by these systems in the downscaled GCMs is evaluated in comparison to NCEP reanalysis datasets, and changes in these types and the rain they bring is examined over the 21st century. The study site is Tasmania, the island in southeast Australia. Tasmania has a temperate maritime climate and complex rainfall distribution and variability across its small area. A complex suite of large-scale remote drivers influences rainfall variability in Tasmania, and these vary with location and season. These drivers affect the rainfall in any location through a change in the frequency or nature of the dominant synoptic systems. Rain bearing systems relevant to Tasmania can be placed into three categories: cutoff lows, cold frontal systems and other. Cutoff lows are significantly associated with blocking in the region, and are an especially important source of rain to the northeast of Tasmania. Onshore cold frontal systems contribute a large proportion of the >3000 mm annual rainfall over the western district of Tasmania. The `other' category includes those systems within pre- and postfrontal airstreams. An automated analysis package, `synview', has been developed by CAWCR to detect and attribute cutoff systems from other types in reanalysis or model data. The synview algorithms are a synthesis of knowledge gained from considerable manual analysis. The synoptic climatology was evaluated in a set of six fine-scale (~10 km grid) dynamically downscaled GCM projections of Tasmanian climate to 2100. These simulations are run at a finer spatial scale than coarse scale GCMs, and therefore improve the resolution of cutoff lows. These fine-scale simulations also show an improved rainfall distribution through a more finely resolved topography and land-ocean boundary. However, we outline limitations of the simulations to reproduce the synoptic climatology of the recent climate when assessed against reanalysis. We conclude that fine-scale dynamical downscaling better accounts for synoptic climatology over Tasmania than coarse scale GCM simulations, but there are still notable limitations. We advocate that an explicit check of the synoptic climatology in model simulations provides useful information about the errors and uncertainties involved in climate modelling. We also find that an analysis of changes to the synoptic climatology in model simulations can help identify the drivers behind changes to rainfall and better inform the use of model projections.
General Circulation Models (GCMs) are our best tool for assessing potential changes to our climate on a global scale into the future. However, certain physical processes that influence rainfall at any particular location operate at spatial scales finer than GCMs can simulate. Regional dynamical downscaling of GCMs addresses this problem by simulating relevant processes at a finer scale, whilst retaining the important global scale features simulated in the original GCM. Thus this process simulates relevant climate mechanisms at several length scales. Here we present fine-scale dynamical downscaling model results of rainfall changes and the drivers behind that change for the challenging test case of Tasmania, Australia. Tasmania is an island with a temperate maritime climate. It is positioned in the roaring 40s of the Southern Ocean and has a temperate maritime climate. It is topographically rugged and has complex rainfall distribution and variability across a small area. Tasmanian rainfall is influenced by a complex suite of large-scale climatic drivers that varies with location and season. Remote drivers of Tasmanian rainfall include the southern annular mode (SAM), El Niño Southern Oscillation (ENSO), Indian Ocean dipole (IOD), the position and intensity of the subtropical ridge (STR) of high pressure, and atmospheric blocking in the adjacent Tasman Sea. We have produced a set of fine-scale (~10 km) projections of Tasmanian climate to 2100 using a process of dynamical downscaling. This process retains global-scale features present in GCMs, and then simulates processes at a finer scale in the area of interest. Projected changes to rainfall to 2100 are distinct in the different districts of the state, and vary greatly by season. Each of these changes is driven by a unique combination of drivers and processes. This includes the local expression of large-scale drivers such as ENSO, IOD, SAM and the position of the STR. It also includes the influence of processes that are poorly resolved or completely lacking in GCM simulations, such as atmospheric blocking, the atmospheric response to changes in local sea surface temperature (SST), convection, and the interaction of synoptic systems with rugged topography. In this paper we present a complete view of rainfall changes in the districts of Tasmania, in which we link large-scale drivers and finer scale processes to spatial and temporal changes in rainfall. Such analyses are only possible with fine-scale dynamical downscaling. We conclude that fine-scale dynamical downscaling provides more useful projections of rainfall changes at a local scale into the future, and that an analysis of all the relevant rainfall mechanisms also adds confidence in the use of climate simulations as a tool for understanding future changes to rainfall.
The ability of regional dynamically-downscaled global circulation models (GCMs) to assess changes to future extreme climatic events was investigated. Extreme value methods were employed for the analysis of precipitation extremes. Projections were generated on a 0.1 grid across Tasmania using the CSIRO Conformal Cubic Atmospheric Model (CCAM) model. Two future SRES emission scenarios (A2 and B1) and multiple GCMs were used for the period 1961-2100. A bias-adjustment procedure was developed to spatially correct extreme magnitudes. Events were correlated against mean trends and compared to gridded and station observations using a suite of indices to demonstrate evolving changes to extremes. Estimates of precipitation return periods were calculated using extreme value analysis. Events were fitted to a Generalised Pareto Distribution (GPD) using an automated threshold selection procedure developed for gridded precipitation datasets. Spatial patterns were found in gridded precipitation extremes that closely matched observations. Projections of future changes to precipitation extremes were found to vary spatially between models, correlating with projected changes to regional climate drivers. Results demonstrate that dynamical downscaling captures regional climate variability (particularly relevant for precipitation) and displays significant ability in modelling future changes to the intensity, magnitude and frequency of extreme events at the local scale for use in adaptation and emergency planning applications.
Environmental context. Understanding the role of clouds in assessing the impact of climate change is a challenging issue. It is thought that plankton and seaweed contribute to the formation of clouds by emitting gases that lead to the particle production necessary for cloud formation. Macroalgae (kelp) at Mace Head, Ireland, produce large quantities of iodine when exposed to sunlight at low tide and this iodine results in the rapid production of particles. Cape Grim, Tasmania, also has large colonies of kelp and the role of Bull Kelp (Durvillaea potatorum) in particle production was assessed. Abstract. Iodine emissions from coastal macroalgae have been found to be important initiators for nucleation events at Mace Head, Ireland. The source of this iodine is the large beds of the brown kelp Laminaria digitata, which are significantly exposed at low tide. On the coast around Cape Grim, Tasmania, there are beds of the brown kelp Durvillaea potatrum. The Precursors to Particles 2006 (P2P 2006) campaign at the Cape Grim Baseline Air Pollution Station in late summer (February) 2006 focused on the role of this local kelp in providing precursor gases to particle formation. Durvillaea potatorum does not produce iodated precursor gases at the levels observed at Mace Head. IO was measured at 0.5 ± 0.3 ppt, while OIO was below detection limits (9 ppt). The dominant atmospheric iodated species was methyl iodide and the average concentration measured at the Cape Grim Station was 1.5 ± 0.3 pptv in baseline conditions, but showed significant variation in discrete samples collected immediately above the ocean surface. Nucleation events were not detected at the Cape Grim Station, except for one period where the plume of a local bushfire interacted with air of marine origin. The passage of four fronts did not result in nucleation bursts and measurements on the beach 94 m below the Cape Grim Station suggested that Durvillaea potatorum was only a weak source of new particles.
Brown kelp, in particular Laminaria digitata at Mace Head, Ireland, has been shown to emit iodine when under stress, resulting in new particle formation. The Cape Grim Baseline Air Pollution Station, Tasmania, is surrounded by rocky reefs that support large colonies of the brown kelp Durvillaea potatorum. During an intensive campaign in February 2006 at Cape Grim, levels of IO, OIO and methyl iodide remained at background levels and no particle formation events could be associated with locally generated precursor iodine species. In order to better understand the limitations of the local kelp to provide a source of precursor species, samples of Durvillaea potatorum were collected from the beach below the Cape Grim Station and tested for their capacity to initiate particle formation using a flux chamber technique. Particles were observed only when the kelp was exposed to both very high levels (>100 ppb) of ozone and natural solar radiation. There was a high correlation between ozone level and particles produced. The particles resulting from exposure to high levels of ozone were aromatic and volatile. Durvillaea potatorum appears to plays a very limited role in contributing to particle formation at Cape Grim, but it does represent a source of atmospheric iodine under photo-oxidative stress, of 18 pmol g−1 (fresh weight) min−1 and is
Environmental context. Emissions of methyl iodide of a biological origin from inshore and coastal waters can be an important component of the atmospheric budget of iodine. Iodine from this and other sources is important in the natural ozone cycle in the troposphere and stratosphere, and may play a role in the formation of new small particles that can then grow to seed clouds. The specific coastal ecology at each location is important to the magnitude and characteristics of this methyl iodide source. Abstract. Methyl iodide concentration in seawater and in the air directly above the sea was measured at an inshore site adjacent to the Cape Grim Baseline Air Pollution Station (Cape Grim BAPS) near a bed of Bull Kelp (Durvillaea potatorum) over daylight cycles and along a transect out to 5 km offshore. Most inshore samples had low and variable methyl iodide concentrations in seawater (14.8–57.7 pM) and in air immediately above the sea (2.1–3.8 parts per trillion by volume), with a partial tidal influence. A period of elevated methyl iodide concentration in the water (144.5 pM) and in air above the sea surface (5.5 pptv) was immediately followed by a measurement of new particles at the Cape Grim BAPS. This correlation provided indirect evidence that emission of methyl iodide from kelp is connected to the new particle formation pathway, but there was no evidence of a direct causal link. Elevated levels of atmospheric methyl iodide were not detected at the station (adjacent to the site but on top of a 94-m cliff) at the same time, which suggests the effect was localised above the sea surface. A rapid decrease of methyl iodide out to 5 km suggested that a source at the coastal reef was greater than from pelagic phytoplankton; this source could be the intertidal kelp beds.