Guidelines for Use of Climate Scenarios Developed from Statistical Downscaling Methods. Wilby, R. L., Charles, S. P., Zorita, E., Timbal, B., Whetton, P., & Mearns, L. O. . ![Guidelines for Use of Climate Scenarios Developed from Statistical Downscaling Methods [pdf]](https://bibbase.org/img/filetypes/pdf.svg "pdf")
Paper abstract bibtex [Excerpt: introduction] The climate change information required for many impact studies is of a spatial scale much finer than that provided by global or regional climate models. The ensuing problems for impact assessment have been recognised for a long time (Kim et al., 1984; Gates, 1985; Robinson and Finkelstein, 1989; Lamb, 1987; Smith and Tirpak, 1989; Cohen, 1990). Global climate models (GCMs) have resolutions of hundreds of kilometres whilst regional climate models (RCMs) may be as fine as tens of kilometres. However, many impact applications require the equivalent of point climate observations and are highly sensitive to fine-scale climate variations that are parameterised in coarse-scale models. This is especially true for regions of complex topography, coastal or island locations, and in regions of highly heterogeneous land-cover. [\n] The most straightforward means of obtaining higher spatial resolution scenarios is to apply coarse-scale climate change projections to a high resolution observed climate baseline - the change factor method. This method is often used when RCM output are unavailable, for sensitivity studies, or whenever rapid assessments of multiple climate change scenarios (and/or GCM experiments) are required. Fine resolution climate change information for use in impact studies can also be obtained via more sophisticated statistical downscaling (SD) methods but such studies have, to date, largely restricted themselves to the use of a single driving GCM. The purpose of this report is to provide background information and guidance on the application of SD methods for climate scenario development. Guidance material on the use of regional modelling for climate scenario development is provided in a companion document (Mearns et al., 2003). [\n] Statistical downscaling is based on the view that the regional climate is conditioned by two factors: the large scale climatic state, and regional/local physiographic features (e.g. topography, land-sea distribution and land use; von Storch, 1995, 1999). From this perspective, regional or local climate information is derived by first determining a statistical model which relates large-scale climate variables (or ” predictors”) to regional and local variables (or ” predictands”). Then the large-scale output of a GCM simulation is fed into this statistical model to estimate the corresponding local and regional climate characteristics. One of the primary advantages of these techniques is that they are computationally inexpensive, and thus can be easily applied to output from different GCM experiments. Another advantage is that they can be used to provide site-specific information, which can be critical for many climate change impact studies. The major theoretical weakness of SD methods is that their basic assumption is not verifiable, i.e., that the statistical relationships developed for the present day climate also hold under the different forcing conditions of possible future climates - a limitation that also applies to the physical parameterizations of dynamical models. [\n] To date, most of the SD approaches described in this document are practiced by climatologists rather than by impact analysts undertaking fully fledged, policy orientated impact assessments. This is because the scenarios have largely been regarded as unreliable, too difficult to interpret, or do not embrace the range of uncertainties in GCM projections in the same way that simpler interpolation methods do. This means that downscaled scenarios based on single GCMs or emission scenarios, when translated into an impact study, can give the misleading impression of increased resolution equating to increased confidence in the projections. However, it is increasingly recognized that comprehensive impact studies must be founded on multiple GCM outputs. [\n] Part 2 of this guidance note provides researchers of climate impacts with background material and descriptions of the main SD techniques. We also outline some of the key assumptions and limitations applying to their usage. Part 3 continues by advising researchers to consider some broader questions about downscaling approaches, whether statistical downscaling or regional modelling. For example, is high-resolution information really needed for the application? Is the effort involved in producing high-resolution climate information appropriate in the context of all of the uncertainties associated with the project? Many of these wider issues are addressed by the companion document on regional climate modelling (Mearns et al., 2003). They are not repeated here so we recommend that both guides be read in tandem. Rather, the focus of Part 3 will be on the practical aspects of statistical downscaling implementation, beginning with the study objectives themselves. Finally, a worked case study is provided in Part 4, and a summary of the key recommendations attached to the proper implementation of SD methods is given in Part 5.
@report{wilbyGuidelinesUseClimate2004,
title = {Guidelines for Use of Climate Scenarios Developed from Statistical Downscaling Methods},
author = {Wilby, R. L. and Charles, S. P. and Zorita, E. and Timbal, B. and Whetton, P. and Mearns, L. O.},
date = {2004},
institution = {{IPCC / IPCC Data Distribution Centre}},
url = {http://www.ipcc-data.org/guidelines/dgm_no2_v1_09_2004.pdf},
abstract = {[Excerpt: introduction] The climate change information required for many impact studies is of a spatial scale much finer than that provided by global or regional climate models. The ensuing problems for impact assessment have been recognised for a long time (Kim et al., 1984; Gates, 1985; Robinson and Finkelstein, 1989; Lamb, 1987; Smith and Tirpak, 1989; Cohen, 1990). Global climate models (GCMs) have resolutions of hundreds of kilometres whilst regional climate models (RCMs) may be as fine as tens of kilometres. However, many impact applications require the equivalent of point climate observations and are highly sensitive to fine-scale climate variations that are parameterised in coarse-scale models. This is especially true for regions of complex topography, coastal or island locations, and in regions of highly heterogeneous land-cover.
[\textbackslash n] The most straightforward means of obtaining higher spatial resolution scenarios is to apply coarse-scale climate change projections to a high resolution observed climate baseline - the change factor method. This method is often used when RCM output are unavailable, for sensitivity studies, or whenever rapid assessments of multiple climate change scenarios (and/or GCM experiments) are required. Fine resolution climate change information for use in impact studies can also be obtained via more sophisticated statistical downscaling (SD) methods but such studies have, to date, largely restricted themselves to the use of a single driving GCM. The purpose of this report is to provide background information and guidance on the application of SD methods for climate scenario development. Guidance material on the use of regional modelling for climate scenario development is provided in a companion document (Mearns et al., 2003).
[\textbackslash n] Statistical downscaling is based on the view that the regional climate is conditioned by two factors: the large scale climatic state, and regional/local physiographic features (e.g. topography, land-sea distribution and land use; von Storch, 1995, 1999). From this perspective, regional or local climate information is derived by first determining a statistical model which relates large-scale climate variables (or ” predictors”) to regional and local variables (or ” predictands”). Then the large-scale output of a GCM simulation is fed into this statistical model to estimate the corresponding local and regional climate characteristics. One of the primary advantages of these techniques is that they are computationally inexpensive, and thus can be easily applied to output from different GCM experiments. Another advantage is that they can be used to provide site-specific information, which can be critical for many climate change impact studies. The major theoretical weakness of SD methods is that their basic assumption is not verifiable, i.e., that the statistical relationships developed for the present day climate also hold under the different forcing conditions of possible future climates - a limitation that also applies to the physical parameterizations of dynamical models.
[\textbackslash n] To date, most of the SD approaches described in this document are practiced by climatologists rather than by impact analysts undertaking fully fledged, policy orientated impact assessments. This is because the scenarios have largely been regarded as unreliable, too difficult to interpret, or do not embrace the range of uncertainties in GCM projections in the same way that simpler interpolation methods do. This means that downscaled scenarios based on single GCMs or emission scenarios, when translated into an impact study, can give the misleading impression of increased resolution equating to increased confidence in the projections. However, it is increasingly recognized that comprehensive impact studies must be founded on multiple GCM outputs.
[\textbackslash n] Part 2 of this guidance note provides researchers of climate impacts with background material and descriptions of the main SD techniques. We also outline some of the key assumptions and limitations applying to their usage. Part 3 continues by advising researchers to consider some broader questions about downscaling approaches, whether statistical downscaling or regional modelling. For example, is high-resolution information really needed for the application? Is the effort involved in producing high-resolution climate information appropriate in the context of all of the uncertainties associated with the project? Many of these wider issues are addressed by the companion document on regional climate modelling (Mearns et al., 2003). They are not repeated here so we recommend that both guides be read in tandem. Rather, the focus of Part 3 will be on the practical aspects of statistical downscaling implementation, beginning with the study objectives themselves. Finally, a worked case study is provided in Part 4, and a summary of the key recommendations attached to the proper implementation of SD methods is given in Part 5.},
keywords = {*imported-from-citeulike-INRMM,~INRMM-MiD:c-8861447,climate,climate-projections,downscaling,ipcc,statistical-downscaling}
}
Downloads: 0
{"_id":"fLAnyMD26Rk2KQcR9","bibbaseid":"wilby-charles-zorita-timbal-whetton-mearns-guidelinesforuseofclimatescenariosdevelopedfromstatisticaldownscalingmethods","authorIDs":[],"author_short":["Wilby, R. L.","Charles, S. P.","Zorita, E.","Timbal, B.","Whetton, P.","Mearns, L. O."],"bibdata":{"bibtype":"report","type":"report","title":"Guidelines for Use of Climate Scenarios Developed from Statistical Downscaling Methods","author":[{"propositions":[],"lastnames":["Wilby"],"firstnames":["R.","L."],"suffixes":[]},{"propositions":[],"lastnames":["Charles"],"firstnames":["S.","P."],"suffixes":[]},{"propositions":[],"lastnames":["Zorita"],"firstnames":["E."],"suffixes":[]},{"propositions":[],"lastnames":["Timbal"],"firstnames":["B."],"suffixes":[]},{"propositions":[],"lastnames":["Whetton"],"firstnames":["P."],"suffixes":[]},{"propositions":[],"lastnames":["Mearns"],"firstnames":["L.","O."],"suffixes":[]}],"date":"2004","institution":"IPCC / IPCC Data Distribution Centre","url":"http://www.ipcc-data.org/guidelines/dgm_no2_v1_09_2004.pdf","abstract":"[Excerpt: introduction] The climate change information required for many impact studies is of a spatial scale much finer than that provided by global or regional climate models. The ensuing problems for impact assessment have been recognised for a long time (Kim et al., 1984; Gates, 1985; Robinson and Finkelstein, 1989; Lamb, 1987; Smith and Tirpak, 1989; Cohen, 1990). Global climate models (GCMs) have resolutions of hundreds of kilometres whilst regional climate models (RCMs) may be as fine as tens of kilometres. However, many impact applications require the equivalent of point climate observations and are highly sensitive to fine-scale climate variations that are parameterised in coarse-scale models. This is especially true for regions of complex topography, coastal or island locations, and in regions of highly heterogeneous land-cover. [\\n] The most straightforward means of obtaining higher spatial resolution scenarios is to apply coarse-scale climate change projections to a high resolution observed climate baseline - the change factor method. This method is often used when RCM output are unavailable, for sensitivity studies, or whenever rapid assessments of multiple climate change scenarios (and/or GCM experiments) are required. Fine resolution climate change information for use in impact studies can also be obtained via more sophisticated statistical downscaling (SD) methods but such studies have, to date, largely restricted themselves to the use of a single driving GCM. The purpose of this report is to provide background information and guidance on the application of SD methods for climate scenario development. Guidance material on the use of regional modelling for climate scenario development is provided in a companion document (Mearns et al., 2003). [\\n] Statistical downscaling is based on the view that the regional climate is conditioned by two factors: the large scale climatic state, and regional/local physiographic features (e.g. topography, land-sea distribution and land use; von Storch, 1995, 1999). From this perspective, regional or local climate information is derived by first determining a statistical model which relates large-scale climate variables (or ” predictors”) to regional and local variables (or ” predictands”). Then the large-scale output of a GCM simulation is fed into this statistical model to estimate the corresponding local and regional climate characteristics. One of the primary advantages of these techniques is that they are computationally inexpensive, and thus can be easily applied to output from different GCM experiments. Another advantage is that they can be used to provide site-specific information, which can be critical for many climate change impact studies. The major theoretical weakness of SD methods is that their basic assumption is not verifiable, i.e., that the statistical relationships developed for the present day climate also hold under the different forcing conditions of possible future climates - a limitation that also applies to the physical parameterizations of dynamical models. [\\n] To date, most of the SD approaches described in this document are practiced by climatologists rather than by impact analysts undertaking fully fledged, policy orientated impact assessments. This is because the scenarios have largely been regarded as unreliable, too difficult to interpret, or do not embrace the range of uncertainties in GCM projections in the same way that simpler interpolation methods do. This means that downscaled scenarios based on single GCMs or emission scenarios, when translated into an impact study, can give the misleading impression of increased resolution equating to increased confidence in the projections. However, it is increasingly recognized that comprehensive impact studies must be founded on multiple GCM outputs. [\\n] Part 2 of this guidance note provides researchers of climate impacts with background material and descriptions of the main SD techniques. We also outline some of the key assumptions and limitations applying to their usage. Part 3 continues by advising researchers to consider some broader questions about downscaling approaches, whether statistical downscaling or regional modelling. For example, is high-resolution information really needed for the application? Is the effort involved in producing high-resolution climate information appropriate in the context of all of the uncertainties associated with the project? Many of these wider issues are addressed by the companion document on regional climate modelling (Mearns et al., 2003). They are not repeated here so we recommend that both guides be read in tandem. Rather, the focus of Part 3 will be on the practical aspects of statistical downscaling implementation, beginning with the study objectives themselves. Finally, a worked case study is provided in Part 4, and a summary of the key recommendations attached to the proper implementation of SD methods is given in Part 5.","keywords":"*imported-from-citeulike-INRMM,~INRMM-MiD:c-8861447,climate,climate-projections,downscaling,ipcc,statistical-downscaling","bibtex":"@report{wilbyGuidelinesUseClimate2004,\n title = {Guidelines for Use of Climate Scenarios Developed from Statistical Downscaling Methods},\n author = {Wilby, R. L. and Charles, S. P. and Zorita, E. and Timbal, B. and Whetton, P. and Mearns, L. O.},\n date = {2004},\n institution = {{IPCC / IPCC Data Distribution Centre}},\n url = {http://www.ipcc-data.org/guidelines/dgm_no2_v1_09_2004.pdf},\n abstract = {[Excerpt: introduction] The climate change information required for many impact studies is of a spatial scale much finer than that provided by global or regional climate models. The ensuing problems for impact assessment have been recognised for a long time (Kim et al., 1984; Gates, 1985; Robinson and Finkelstein, 1989; Lamb, 1987; Smith and Tirpak, 1989; Cohen, 1990). Global climate models (GCMs) have resolutions of hundreds of kilometres whilst regional climate models (RCMs) may be as fine as tens of kilometres. However, many impact applications require the equivalent of point climate observations and are highly sensitive to fine-scale climate variations that are parameterised in coarse-scale models. This is especially true for regions of complex topography, coastal or island locations, and in regions of highly heterogeneous land-cover.\n\n[\\textbackslash n] The most straightforward means of obtaining higher spatial resolution scenarios is to apply coarse-scale climate change projections to a high resolution observed climate baseline - the change factor method. This method is often used when RCM output are unavailable, for sensitivity studies, or whenever rapid assessments of multiple climate change scenarios (and/or GCM experiments) are required. Fine resolution climate change information for use in impact studies can also be obtained via more sophisticated statistical downscaling (SD) methods but such studies have, to date, largely restricted themselves to the use of a single driving GCM. The purpose of this report is to provide background information and guidance on the application of SD methods for climate scenario development. Guidance material on the use of regional modelling for climate scenario development is provided in a companion document (Mearns et al., 2003).\n\n[\\textbackslash n] Statistical downscaling is based on the view that the regional climate is conditioned by two factors: the large scale climatic state, and regional/local physiographic features (e.g. topography, land-sea distribution and land use; von Storch, 1995, 1999). From this perspective, regional or local climate information is derived by first determining a statistical model which relates large-scale climate variables (or ” predictors”) to regional and local variables (or ” predictands”). Then the large-scale output of a GCM simulation is fed into this statistical model to estimate the corresponding local and regional climate characteristics. One of the primary advantages of these techniques is that they are computationally inexpensive, and thus can be easily applied to output from different GCM experiments. Another advantage is that they can be used to provide site-specific information, which can be critical for many climate change impact studies. The major theoretical weakness of SD methods is that their basic assumption is not verifiable, i.e., that the statistical relationships developed for the present day climate also hold under the different forcing conditions of possible future climates - a limitation that also applies to the physical parameterizations of dynamical models.\n\n[\\textbackslash n] To date, most of the SD approaches described in this document are practiced by climatologists rather than by impact analysts undertaking fully fledged, policy orientated impact assessments. This is because the scenarios have largely been regarded as unreliable, too difficult to interpret, or do not embrace the range of uncertainties in GCM projections in the same way that simpler interpolation methods do. This means that downscaled scenarios based on single GCMs or emission scenarios, when translated into an impact study, can give the misleading impression of increased resolution equating to increased confidence in the projections. However, it is increasingly recognized that comprehensive impact studies must be founded on multiple GCM outputs.\n\n[\\textbackslash n] Part 2 of this guidance note provides researchers of climate impacts with background material and descriptions of the main SD techniques. We also outline some of the key assumptions and limitations applying to their usage. Part 3 continues by advising researchers to consider some broader questions about downscaling approaches, whether statistical downscaling or regional modelling. For example, is high-resolution information really needed for the application? Is the effort involved in producing high-resolution climate information appropriate in the context of all of the uncertainties associated with the project? Many of these wider issues are addressed by the companion document on regional climate modelling (Mearns et al., 2003). They are not repeated here so we recommend that both guides be read in tandem. Rather, the focus of Part 3 will be on the practical aspects of statistical downscaling implementation, beginning with the study objectives themselves. Finally, a worked case study is provided in Part 4, and a summary of the key recommendations attached to the proper implementation of SD methods is given in Part 5.},\n keywords = {*imported-from-citeulike-INRMM,~INRMM-MiD:c-8861447,climate,climate-projections,downscaling,ipcc,statistical-downscaling}\n}\n\n","author_short":["Wilby, R. L.","Charles, S. P.","Zorita, E.","Timbal, B.","Whetton, P.","Mearns, L. O."],"key":"wilbyGuidelinesUseClimate2004","id":"wilbyGuidelinesUseClimate2004","bibbaseid":"wilby-charles-zorita-timbal-whetton-mearns-guidelinesforuseofclimatescenariosdevelopedfromstatisticaldownscalingmethods","role":"author","urls":{"Paper":"http://www.ipcc-data.org/guidelines/dgm_no2_v1_09_2004.pdf"},"keyword":["*imported-from-citeulike-INRMM","~INRMM-MiD:c-8861447","climate","climate-projections","downscaling","ipcc","statistical-downscaling"],"downloads":0},"bibtype":"report","biburl":"https://tmpfiles.org/dl/58794/INRMM.bib","creationDate":"2020-07-02T22:41:33.551Z","downloads":0,"keywords":["*imported-from-citeulike-inrmm","~inrmm-mid:c-8861447","climate","climate-projections","downscaling","ipcc","statistical-downscaling"],"search_terms":["guidelines","use","climate","scenarios","developed","statistical","downscaling","methods","wilby","charles","zorita","timbal","whetton","mearns"],"title":"Guidelines for Use of Climate Scenarios Developed from Statistical Downscaling Methods","year":null,"dataSources":["DXuKbcZTirdigFKPF"]}