the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Future changes in seasonal drought in Australia
Abstract. Climate change is expected to exacerbate the frequency and intensity of drought in many water-limited regions. However, future drought changes in Australia –the driest inhabited continent on Earth– have remained stubbornly uncertain due to a lack of model agreement in projected precipitation changes in most regions. We use an ensemble of future projections from the National Hydrological Projections to assess future drought changes in Australia. The ensemble of 32 simulations was created using the Australian Landscape Water Balance model (AWRA-L) forced by four global climate models (GCMs) from the Coupled Model Intercomparison Project phase 5 that were downscaled and bias-corrected using four alternative methods. This ensemble provided an opportunity to analyse multiple sources of uncertainty on the future projections and quantify changes across multiple drought types (meteorological, hydrological and agricultural). We show future increases in the time spent under drought for all three drought types, with largest increases projected in winter and spring. The future changes are particularly robust in the highly populated and agricultural regions of Australia, suggesting potential impacts on agricultural activities, ecosystems and urban water supply. We attributed uncertainty in future drought changes to GCMs, downscaling/bias correction methods and emissions scenarios. GCMs represent the largest source of uncertainty but the choice of downscaling method is also important. The emissions scenarios were the lowest source of uncertainty but influenced the magnitude and spatial extent of robust future changes. Overall, the projections suggest likely future increases in drought in Australia with little evidence for ameliorating drought risk with climate change despite ongoing uncertainty in future changes in parts of the country.
Status: closed (peer review stopped)
-
RC1: 'Comment on egusphere-2024-1925', Anonymous Referee #1, 05 Nov 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2024-1925/egusphere-2024-1925-RC1-supplement.pdf
-
RC2: 'Comment on egusphere-2024-1925', Anonymous Referee #2, 20 Nov 2024
Ukolla et al. analyse an important topic, the impact of climate change on three drought types in Australia, including their seasonal changes. They use an ensemble of 32 members (4 x 4 x 2) based on the Australian Landscape Water Balance model (AWRA-L), forced with downscaled and bias-corrected 4 CMIP5 models. They used three statistical bias correction approaches and one combined downscaling and bias correction approach for two RCPs (4.5 and 8.5). Their results suggest an overall increase in all three drought types (meteorological, hydrological, and agricultural), particularly in winter and spring. They also attempt to quantify associated uncertainties. The topic of this manuscript is relevant to HESS readership and it is a nice contribution to the community. I find the current manuscript is very suitable for publication in HESS after addressing some minor comments listed below. The paper is mostly very clearly written and well-referenced.
Here are my minor comments:
- The abstract does not clearly address, which drought characteristics you quantify. You mention only one aspect in the abstract, L24: “time spent under drought”, which could be better referred to as “drought duration”? From the abstract should be clear whether you also considered other characteristics, such as severity or spatial extent. If not, then the title of the manuscript should be adjusted accordingly. Also, L30 “ future increases in drought” doesn’t specify which aspect of drought is analysed. Also, in L98 in the Introduction, they say “across different indicators of drought”, but they don’t explicitly mention them. You mention them for the first time on L231ff. Results on drought intensity should be also mentioned in the abstract.
- Line 109: “model that is calibrated towards observed river streamflow, satellite soil moisture and evapotranspiration across the continent.” It is not clear, whether it was done in this study, or they refer to any other previous work,
- Lines 121-124: did you do any of these evaluations, which you could possibly include in the supporting information file?
- Section 2.2.1. on historical observations should include some of the model’s evaluations. I fully understand that “gridded runoff and soil moisture observations are not available”, but still, you could compare the routed runoff against observed streamflow observations to assess the credibility of your simulations in the historical period.
- Lines 161-179: was this done by the authors, or taken from other authors, i.e., Peters et al.?
- Line 197: are you sure that you are able to obtain steady-state conditions of your model’s states in 10 years, given the very arid region? I guess in the dry regions, it can take longer… Did you check the time series of selected pixels? I can imagine the disagreement in results (grey in Fin 3) in central Australia could be driven by this factor of too short initialization.
- Section 2.2.3 You run the model at a daily time step, and then drought analysis is done at a monthly time scale. I guess, you need to state this somewhere explicitly, possibly in this section. And then you apply 3 months averaging. This sequence needs to be stated clearly here in the section. Then, I would suggest moving L206-214 elsewhere because they are a bit distracting where they are. I would start directly with L203-205 and then move directly to L223 onwards. L205-208 could go to discussion.
- L 231: you could have also analyse spatial drought extent? Do you see distinct results for behaviours in duration and time under drought? If not, then I would suggest keeping just one.
- In results, the results quantify the drought types. It would be interesting, which aspects lead to the runoff droughts, which seem to be by 20% longer, it’s not only because of precipitation deficits, but surely from evaporative increases due to increased temperature? Also, in Fig.2, there are the reference values missing, to better relate the percentage increase to a reference. The 20% would be different from 2 months or from 4 months … ?
- It might be useful to rearrange the sequence of the results. How about starting with 3.4, where observations are compared, and show basic characteristics of individual realization (Fig 5), then showing aggregated characteristics of drought for the full ensemble (Figs 2 and 3)…
- Why does the GFDL model stand so much apart? Is it because of precipitation or temperature differences?
- Nicely written discussion section, but could the conclusions be taken apart into one paragraph section at the end?
Citation: https://doi.org/10.5194/egusphere-2024-1925-RC2
Status: closed (peer review stopped)
-
RC1: 'Comment on egusphere-2024-1925', Anonymous Referee #1, 05 Nov 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2024-1925/egusphere-2024-1925-RC1-supplement.pdf
-
RC2: 'Comment on egusphere-2024-1925', Anonymous Referee #2, 20 Nov 2024
Ukolla et al. analyse an important topic, the impact of climate change on three drought types in Australia, including their seasonal changes. They use an ensemble of 32 members (4 x 4 x 2) based on the Australian Landscape Water Balance model (AWRA-L), forced with downscaled and bias-corrected 4 CMIP5 models. They used three statistical bias correction approaches and one combined downscaling and bias correction approach for two RCPs (4.5 and 8.5). Their results suggest an overall increase in all three drought types (meteorological, hydrological, and agricultural), particularly in winter and spring. They also attempt to quantify associated uncertainties. The topic of this manuscript is relevant to HESS readership and it is a nice contribution to the community. I find the current manuscript is very suitable for publication in HESS after addressing some minor comments listed below. The paper is mostly very clearly written and well-referenced.
Here are my minor comments:
- The abstract does not clearly address, which drought characteristics you quantify. You mention only one aspect in the abstract, L24: “time spent under drought”, which could be better referred to as “drought duration”? From the abstract should be clear whether you also considered other characteristics, such as severity or spatial extent. If not, then the title of the manuscript should be adjusted accordingly. Also, L30 “ future increases in drought” doesn’t specify which aspect of drought is analysed. Also, in L98 in the Introduction, they say “across different indicators of drought”, but they don’t explicitly mention them. You mention them for the first time on L231ff. Results on drought intensity should be also mentioned in the abstract.
- Line 109: “model that is calibrated towards observed river streamflow, satellite soil moisture and evapotranspiration across the continent.” It is not clear, whether it was done in this study, or they refer to any other previous work,
- Lines 121-124: did you do any of these evaluations, which you could possibly include in the supporting information file?
- Section 2.2.1. on historical observations should include some of the model’s evaluations. I fully understand that “gridded runoff and soil moisture observations are not available”, but still, you could compare the routed runoff against observed streamflow observations to assess the credibility of your simulations in the historical period.
- Lines 161-179: was this done by the authors, or taken from other authors, i.e., Peters et al.?
- Line 197: are you sure that you are able to obtain steady-state conditions of your model’s states in 10 years, given the very arid region? I guess in the dry regions, it can take longer… Did you check the time series of selected pixels? I can imagine the disagreement in results (grey in Fin 3) in central Australia could be driven by this factor of too short initialization.
- Section 2.2.3 You run the model at a daily time step, and then drought analysis is done at a monthly time scale. I guess, you need to state this somewhere explicitly, possibly in this section. And then you apply 3 months averaging. This sequence needs to be stated clearly here in the section. Then, I would suggest moving L206-214 elsewhere because they are a bit distracting where they are. I would start directly with L203-205 and then move directly to L223 onwards. L205-208 could go to discussion.
- L 231: you could have also analyse spatial drought extent? Do you see distinct results for behaviours in duration and time under drought? If not, then I would suggest keeping just one.
- In results, the results quantify the drought types. It would be interesting, which aspects lead to the runoff droughts, which seem to be by 20% longer, it’s not only because of precipitation deficits, but surely from evaporative increases due to increased temperature? Also, in Fig.2, there are the reference values missing, to better relate the percentage increase to a reference. The 20% would be different from 2 months or from 4 months … ?
- It might be useful to rearrange the sequence of the results. How about starting with 3.4, where observations are compared, and show basic characteristics of individual realization (Fig 5), then showing aggregated characteristics of drought for the full ensemble (Figs 2 and 3)…
- Why does the GFDL model stand so much apart? Is it because of precipitation or temperature differences?
- Nicely written discussion section, but could the conclusions be taken apart into one paragraph section at the end?
Citation: https://doi.org/10.5194/egusphere-2024-1925-RC2
Viewed
Since the preprint corresponding to this journal article was posted outside of Copernicus Publications, the preprint-related metrics are limited to HTML views.
HTML | XML | Total | BibTeX | EndNote | |
---|---|---|---|---|---|
150 | 0 | 0 | 150 | 0 | 0 |
- HTML: 150
- PDF: 0
- XML: 0
- Total: 150
- BibTeX: 0
- EndNote: 0
Viewed (geographical distribution)
Since the preprint corresponding to this journal article was posted outside of Copernicus Publications, the preprint-related metrics are limited to HTML views.
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1