Observation-based multi-decadal time series of dense shelf water properties (Deliverable D1.11)

Ocean Cryosphere Exchanges in Antarctica: Impacts on Climate and the Earth System Observation-based multi-decadal time series of dense shelf water properties Deliverable D1.11 https://ocean-ice.eu/ OCEAN:ICE – GA 101060452 Deliverable D1.11 About this document Deliverable: D1.11 Observation-based multi-decadal time series of dense shelf water properties Work Package: WP 1 Subpolar circulation, heat delivery and water mass Export, Task 1.4: Shelf Seas as a dense water source for the global thermohaline circulation Delivery date: 31 October 2025 Type of document: Report Dissemination level: Public Lead beneficiary and author: UKRI-BAS (partner 13), Pierre Dutrieux, [email protected], Shenjie Zhou, [email protected]k Review: PP1: Danish Meteorological Institute (DMI), Chiara Bearzotti ([email protected]) Cover sheet: Polygons of different shelf seas within which the T-S anomaly timeseries is constructed. Disclaimer: Funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or European Research Executive Agency (REA). Neither the European Union nor the granting authority can be held responsible for them. 2 OCEAN:ICE – GA 101060452 Deliverable D1.11 Table of contents 1. Publishable summary...........................................................................................................................4 2. Work performed and main achievements........................................................................................... 5 2.1 Description of the work performed.............................................................................................5 2.2 Open Science...............................................................................................................................8 3. Results..................................................................................................................................................9 3.1 Achieved results.......................................................................................................................... 9 3.2 Datasets.......................................................................................................................................9 4. Impact..................................................................................................................................................9 3 OCEAN ICE – GA 101060452 Deliverable D1.11 1. Publishable summary In this deliverable, we present a unique dataset that captures the temporal variability of temperature and salinity across the entire water column on Antarctic continental shelves. The temporal evolution of shelf sea water properties is evaluated based on a series of profiles of temperature and salinity anomaly computed by subtracting the hydrography profiles compiled in D1.1 from the gridded 3-D climatology constructed with the same profile compilation. As such, it is possible to evaluate the temporal variation of water mass properties in specific regions by combining anomaly profiles sampled in the vicinity, assuming that the deviation from the mean field in the selected region is spatially coherent. The dataset can facilitate the direct comparison of the dense shelf water formed on the continental shelf, where data is applicable to the abyssal ocean observation of the bottom water properties. The dataset has retained the vertical dimension of the source dataset; therefore, it can also be used to understand the temporal variability of temperature and salinity of different water masses throughout the water column, along with the background mean temperature/salinity profiles. Figure 1 below showcases the information that one can extract from the dataset, exemplified as Hovmöller diagram of the temperature and salinity anomaly over time and depth in the Ross Sea continental shelf. The dataset is published on SEANOE and is freely accessible: https://doi.org/10.17882/109454. Fig.1: Ross Sea continental shelf T-S anomaly Hovmöller diagram. 4 OCEAN ICE – GA 101060452 Deliverable D1.6 2. Work performed and main achievements 2.1 Description of the work performed Climatology In D1.1, we gathered over 800,000 temperature and salinity profiles south of 45°S collected with ship-based CTD casts, Argo floats and Seal-borne profilers. These profiles were then objectively mapped onto a 3-D regular spatial grid to estimate the climatological mean state of the Southern Ocean and the Antarctic continental shelf water mass properties (Figure 2). The methodology for constructing the climatology is included in an ongoing data release paper manuscript that will be submitted soon to Earth System Science Data. The climatology data product is available at https://doi.org/10.17882/103946, and the source profile database is available at https://doi.org/10.17882/99787. Fig. 2: The climatology field of potential temperature at 1500 m depth and the bottom conservative temperature on the Antarctic continental shelf. Constructing a pan-Antarctic continental shelf time-varying T-S anomaly field Due to the inhomogeneity of the observational data in space and time, it is difficult to assemble scattered data in time without contaminating the spatial gradients of temperature and salinity. As a first step, to eliminate the aliasing from the spatial gradient, which can be quite sharp in Antarctic shelf seas (see Figure 2, right panel), we remove the climatology mean value from each individual T-S profile. This allows us to alleviate the potential that samples from nearby locations harboring different water masses are misconstrued as a representation of temporal variability. An associated caveat, however, is that when we aggregate temporal variations from a regional setting, we necessarily assume that the deviations of these observational profiles from the climatology at its nearest grid point are spatially coherent within the aggregation/spatial averaging range. In a second step, to best avoid data gaps in space and time, we arbitrarily define a spatial grid, and a search box is applied to each grid point. Data falling into each box is then placed in ascending time order to reflect the T-S evolution at each grid point. Figure 3 illustrates the grid on which we assess 5 OCEAN ICE – GA 101060452 Deliverable D1.6 the anomaly field. Red points are the central grid points with 20 km horizontal spacing. The black squares around them depict the searching box (200 by 200 km) centred around two cyan example dots. The grid spacing and box size were chosen to smooth the data over space and time. For the purpose of assessing water mass properties on the continental shelf, we use only profiles collected inland of the continental shelf break defined by the 1000 m isobath. Fig. 3: The spatial grid on which the T-S anomaly profiles are binned. Two cyan example dots - Points a and b are showcasing the size of the searching box, illustrated as a black square around Points a and b. Note that no further mapping is performed during this process; therefore, the temporal coverage of the Hovmöller diagram at each grid point depends heavily on the data abundance. Figure 4 presents the local climatology and Hovmöller diagrams of temperature and salinity at Points a and b in Figure 3. 6 OCEAN ICE – GA 101060452 Deliverable D1.6 Fig. 4: The Hovmöller diagrams of temperature and salinity anomalies at Point a (Filchner Trough, Weddell) and Point b (Jodies Trough, Ross) in Figure 3. Panels on the left show the local climatological T-S profiles with the grey shading indicating the spread of anomalies throughout the water column. In the published dataset, we performed further binning on the shelf-sea scale to obtain temporal variations in temperature and salinity across the entire shelf sea region around Antarctica (Figure 5). Figure 5 shows the polygons in red used for shelf-sea scale binning overlying the grid points of base data illustrated as black dots (same as in Figure 4) that we described above. The smoothed blue contour is used to represent the broad orientation of the shelf break around Antarctica, which is then used to determine the orientation of each polygon that extends from the smoothed contour inland toward the coastline shown as a cyan contour in Figure 5. Only the shelf sea data were used for the 7 OCEAN ICE – GA 101060452 Deliverable D1.6 shelf-sea binning practice. Figure 1 shows an example of the Ross continental shelf Hovmöller diagram, constructed by selecting and binning grid points that fall within polygon #5 in Figure 5. Fig. 5: Map of polygons (red) of different shelf seas within which the T-S anomaly timeseries is constructed. The blue contour shows a smoothed representation of the shelf break (grey) to guide the orientation of the polygons that extend inland toward the coastline (cyan). Blue circles denote the central location on the smoothed contour for each polygon. In total, 15 polygons are constructed. A clear seasonal variation of the temperature and salinity field can be resolved in the Hovmöller diagram along with a long-term freshening signal seen in the mid-water column depth range between 200 to 700 m accompanied by a full water column ‘convective’-like flushing of salinification signal in 2019, consistent with previously reported salinity rebound in the Ross Sea Dense Shelf Water associated with the anomalous sea ice formation driven by the wind anomalies and relevant climatic modes. But there is always a balance between the temporal coverage and spatial searching range, as the assumption of spatially-coherent varying anomaly fields will fail once the selecting regions cover a broad range where different dynamical processes control the temporal variability in the data. 2.2 Open Science The dataset generated and described in this deliverable has been published at SEANOE Data Repository, https://doi.org/10.17882/109454 and is freely available upon the submission of this deliverable report. 8 OCEAN ICE – GA 101060452 Deliverable D1.6 3. Results 3.1 Achieved results The main result of this deliverable is a collection of temperature and salinity anomaly fields changing over time and the corresponding local T-S climatology at different grid points all over the continental shelf of Antarctica, as illustrated in Figures 3 and 4 in the previous section. The dataset is prepared in NetCDF format and contains the following variables. double XC(nshelf, scalar) ; XC:standard\ name = "x coordinate of the polygon central location" ; XC:unit = "m" ; double YC(nshelf, scalar) ; YC:standard\ name = "y coordinate of the polygon central location" ; YC:unit = "m" ; double PRES(nz, scalar) ; PRES:standard\ name = "water pressure" ; PRES:unit = "dbar" ; PRES:range = "[5 3005]" ; PRES:_FillValue = NaN ; double TIME(nt, scalar) ; TIME:standard\ name = "decimal year" ; TIME:unit = "year" ; TIME:range = "[1970 2025]" ; double CT(nt, nz, nshelf) ; CT:standard\ name = "conservative temperature anomaly" ; CT:unit = "°C" ; CT:_FillValue = NaN ; double SA(nt, nz, nshelf) ; SA:standard\ name = "absolute salinity anomaly" ; SA:unit = "g/kg" ; SA:_FillValue = NaN ; double CT_CLIM(nz, nshelf) ; CT_CLIM:standard\ name = "conservative temperature climatology" ; CT_CLIM:unit = "°C" ; CT_CLIM:_FillValue = NaN ; double SA_CLIM(nz, nshelf) ; SA_CLIM:standard\ name = "absolute salinity climatology" ; SA_CLIM:unit = "°C" ; SA_CLIM:_FillValue = NaN ; 3.2 Datasets PID (type of persistent identifier) Type of PID Brief description of the dataset URL to repository 10.17882/109454 DOI Temporally-varying T-S anomaly field on the Antarctic continental shelf https://www.seanoe.org/da ta/00983/109454/ 4. Impact This deliverable contributes to the following project objectives, as described in the action description. O1: Reduce the spatial and knowledge gaps in ocean observations around Antarctica. The work provides a carefully curated observational dataset depicting the temporal variation of the temperature and salinity fields over seasonal-to-multidecadal timescales. It can help pin down 9