Statistically downscaled CMIP6 ocean variables for European waters
Abstract Climate change impact studies need climate projections for different scenarios and at scales relevant to planning and management, preferably for a variety of models and realizations to capture the uncertainty in these models. To address current gaps, we statistically downscaled (SD) 3–7 CMI...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2024-01-01
|
| Series: | Scientific Reports |
| Online Access: | https://doi.org/10.1038/s41598-024-51160-1 |
| _version_ | 1797355777615200256 |
|---|---|
| author | Trond Kristiansen Momme Butenschön Myron A. Peck |
| author_facet | Trond Kristiansen Momme Butenschön Myron A. Peck |
| author_sort | Trond Kristiansen |
| collection | DOAJ |
| description | Abstract Climate change impact studies need climate projections for different scenarios and at scales relevant to planning and management, preferably for a variety of models and realizations to capture the uncertainty in these models. To address current gaps, we statistically downscaled (SD) 3–7 CMIP6 models for five key indicators of marine habitat conditions: temperature, salinity, pH, oxygen, and chlorophyll across European waters for three climate scenarios SSP1-2.6, SSP2-4.5, and SSP5-8.5. Results provide ensemble averages and uncertainty estimates that can serve as input data for projecting the potential success of a range of Nature-based Solutions, including the restoration of habitat-forming species such as seagrass in the Mediterranean and kelp in coastal areas of Portugal and Norway. Evaluation of the ensemble with observations from four European regions (North Sea, Baltic Sea, Bay of Biscay, and Mediterranean Sea) indicates that the SD projections realistically capture the climatological conditions of the historical period 1993–2020. Model skill (Liu-mean efficiency, Pearson correlation) clearly improves for both surface temperature and oxygen across all regions with respect to the original ESMs demonstrating a higher skill for temperature compared to oxygen. Warming is evident across all areas and large differences among scenarios fully emerge from the background uncertainties related to internal variability and model differences in the second half of the century. Scenario-specific differences in acidification significantly emerge from model uncertainty and internal variability leading to distinct trajectories in surface pH starting before mid-century (in some cases starting from present day). Deoxygenation is also present across all domains, but the climate signal was significantly weaker compared to the other two indicators when compared to model uncertainty and internal variability, and the impact of different greenhouse gas trajectories is less distinct. The substantial regional and local heterogeneity in these three abiotic indicators underscores the need for highly spatially resolved physical and biogeochemical projections to understand how climate change may impact marine ecosystems. |
| first_indexed | 2024-03-08T14:17:01Z |
| format | Article |
| id | doaj.art-406eecbef59d48a999973a8deb77850a |
| institution | Directory Open Access Journal |
| issn | 2045-2322 |
| language | English |
| last_indexed | 2024-03-08T14:17:01Z |
| publishDate | 2024-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj.art-406eecbef59d48a999973a8deb77850a2024-01-14T12:21:26ZengNature PortfolioScientific Reports2045-23222024-01-0114112010.1038/s41598-024-51160-1Statistically downscaled CMIP6 ocean variables for European watersTrond Kristiansen0Momme Butenschön1Myron A. Peck2Farallon InstituteCMCC Foundation—Euro-Mediterranean Center on Climate ChangeDepartment of Coastal Systems, Royal Netherlands Institute for Sea ResearchAbstract Climate change impact studies need climate projections for different scenarios and at scales relevant to planning and management, preferably for a variety of models and realizations to capture the uncertainty in these models. To address current gaps, we statistically downscaled (SD) 3–7 CMIP6 models for five key indicators of marine habitat conditions: temperature, salinity, pH, oxygen, and chlorophyll across European waters for three climate scenarios SSP1-2.6, SSP2-4.5, and SSP5-8.5. Results provide ensemble averages and uncertainty estimates that can serve as input data for projecting the potential success of a range of Nature-based Solutions, including the restoration of habitat-forming species such as seagrass in the Mediterranean and kelp in coastal areas of Portugal and Norway. Evaluation of the ensemble with observations from four European regions (North Sea, Baltic Sea, Bay of Biscay, and Mediterranean Sea) indicates that the SD projections realistically capture the climatological conditions of the historical period 1993–2020. Model skill (Liu-mean efficiency, Pearson correlation) clearly improves for both surface temperature and oxygen across all regions with respect to the original ESMs demonstrating a higher skill for temperature compared to oxygen. Warming is evident across all areas and large differences among scenarios fully emerge from the background uncertainties related to internal variability and model differences in the second half of the century. Scenario-specific differences in acidification significantly emerge from model uncertainty and internal variability leading to distinct trajectories in surface pH starting before mid-century (in some cases starting from present day). Deoxygenation is also present across all domains, but the climate signal was significantly weaker compared to the other two indicators when compared to model uncertainty and internal variability, and the impact of different greenhouse gas trajectories is less distinct. The substantial regional and local heterogeneity in these three abiotic indicators underscores the need for highly spatially resolved physical and biogeochemical projections to understand how climate change may impact marine ecosystems.https://doi.org/10.1038/s41598-024-51160-1 |
| spellingShingle | Trond Kristiansen Momme Butenschön Myron A. Peck Statistically downscaled CMIP6 ocean variables for European waters Scientific Reports |
| title | Statistically downscaled CMIP6 ocean variables for European waters |
| title_full | Statistically downscaled CMIP6 ocean variables for European waters |
| title_fullStr | Statistically downscaled CMIP6 ocean variables for European waters |
| title_full_unstemmed | Statistically downscaled CMIP6 ocean variables for European waters |
| title_short | Statistically downscaled CMIP6 ocean variables for European waters |
| title_sort | statistically downscaled cmip6 ocean variables for european waters |
| url | https://doi.org/10.1038/s41598-024-51160-1 |
| work_keys_str_mv | AT trondkristiansen statisticallydownscaledcmip6oceanvariablesforeuropeanwaters AT mommebutenschon statisticallydownscaledcmip6oceanvariablesforeuropeanwaters AT myronapeck statisticallydownscaledcmip6oceanvariablesforeuropeanwaters |