Author Contributions
Conceptualization, M.V.-Y., F.M. and M.C.G.-M.; methodology, M.V.-Y., M.J. and G.J.; software, M.V.-Y., M.J. and G.J.; investigation, S.P., E.B. and C.A.; resources and data curation, S.P., R.S., R.B., M.S., J.S., J.P., E.T. and V.M.; writing—original draft preparation, M.V.-Y. and M.J.; writing—review and editing, F.M., E.B., J.S., E.T. and C.A.; visualization, M.V.-Y.; project administration, M.C.G.-M.; funding acquisition, M.C.G.-M. All authors have read and agreed to the published version of the manuscript.
Figure 1.
Red dots are the position of the RADMED project stations. Letters close to the transects stand for Cape Pino (P), Málaga (M), Vélez (V), Cape Sacratif (S), Cape Gata (CG), Cape Palos (CP), Balearic Channels (C), Mallorca (B), Tarragona (T), Barcelona (BNA), and Mahón (MH). Cabrera deep station is named as “Cabrera”. Blue dots and blue labels indicate the position of coastal tide-gauges: Alegiras (Alg.), Tarifa, Ceuta, Málaga (M), Alicante, and Palma. Black dots show the positions of the Fuengirola temperature time series, L’Estartit oceanographic station and tide-gauge, and the Blanes Bay Microbiological Observatory (BBMO). The base of the triangles in the Balearic Channels also correspond to the glider lines and the seasonal surveys maintained by SOCIB.
Figure 1.
Red dots are the position of the RADMED project stations. Letters close to the transects stand for Cape Pino (P), Málaga (M), Vélez (V), Cape Sacratif (S), Cape Gata (CG), Cape Palos (CP), Balearic Channels (C), Mallorca (B), Tarragona (T), Barcelona (BNA), and Mahón (MH). Cabrera deep station is named as “Cabrera”. Blue dots and blue labels indicate the position of coastal tide-gauges: Alegiras (Alg.), Tarifa, Ceuta, Málaga (M), Alicante, and Palma. Black dots show the positions of the Fuengirola temperature time series, L’Estartit oceanographic station and tide-gauge, and the Blanes Bay Microbiological Observatory (BBMO). The base of the triangles in the Balearic Channels also correspond to the glider lines and the seasonal surveys maintained by SOCIB.
Figure 2.
(
A–
C) show the number of data available for each month of the year at 0, 600, and 1200 m depth within the square corresponding to the Alboran Sea. (
D–
F) show the same results for the Northern Sector (see red rectangles in
Figure 1).
Figure 2.
(
A–
C) show the number of data available for each month of the year at 0, 600, and 1200 m depth within the square corresponding to the Alboran Sea. (
D–
F) show the same results for the Northern Sector (see red rectangles in
Figure 1).
Figure 3.
Black dots in (A) are the surface temperature values from a three-monthly sampling from satellite SST data in Málaga Bay for the years 1982 and 1983. Blue dots are the annual mean values for those years. (B) shows the same temperature values as in (A), eliminating the summer 1982 and the winter 1983 values. Blue dots are the annual mean values presented in (A), and red dots are the annual means calculated after the elimination of summer 1982 and winter 1983. In (C), the missing values have been substituted with the climatological averages calculated for the period 1982–2021 (blue triangles). Red dots are the new annual values after substitution using the climatologies. (D) illustrates the representative anomalies method. The climatological values corresponding to each season are plotted with blue triangles. Black arrows show the anomalies for each season of the year.
Figure 3.
Black dots in (A) are the surface temperature values from a three-monthly sampling from satellite SST data in Málaga Bay for the years 1982 and 1983. Blue dots are the annual mean values for those years. (B) shows the same temperature values as in (A), eliminating the summer 1982 and the winter 1983 values. Blue dots are the annual mean values presented in (A), and red dots are the annual means calculated after the elimination of summer 1982 and winter 1983. In (C), the missing values have been substituted with the climatological averages calculated for the period 1982–2021 (blue triangles). Red dots are the new annual values after substitution using the climatologies. (D) illustrates the representative anomalies method. The climatological values corresponding to each season are plotted with blue triangles. Black arrows show the anomalies for each season of the year.
Figure 4.
Frequency distribution of the dates corresponding to all the RADMED cruises in the Málaga Bay from 1992 for each season of the year: (A), winter, (B), spring, (C), summer, and (D), autumn. Inserts in each figure show the average date for each season.
Figure 4.
Frequency distribution of the dates corresponding to all the RADMED cruises in the Málaga Bay from 1992 for each season of the year: (A), winter, (B), spring, (C), summer, and (D), autumn. Inserts in each figure show the average date for each season.
Figure 5.
(A) shows all the potential temperature time series corresponding to different data processing methods for the surface layer (0–150 m) averaged for the four geographical areas analyzed in this work: Alboran, Cape Palos, Balearic Islands, and Northern Sector. Black, blue, red, and green symbols correspond to representative anomalies, zero anomalies, zero anomalies + trend, and ignorance, respectively. Filled symbols correspond to the use of zero anomaly for the vertical and horizontal average, and open symbols to representative anomalies. Black line in (B) is the mean value of all the time series shown in (A). The gray-shaded area shows the range of values corresponding to the different methodologies. (C,D) show similar results, but for the intermediate layer (150–600 m), and (E,F) are the same, but for the deep layer (600 m–bottom).
Figure 5.
(A) shows all the potential temperature time series corresponding to different data processing methods for the surface layer (0–150 m) averaged for the four geographical areas analyzed in this work: Alboran, Cape Palos, Balearic Islands, and Northern Sector. Black, blue, red, and green symbols correspond to representative anomalies, zero anomalies, zero anomalies + trend, and ignorance, respectively. Filled symbols correspond to the use of zero anomaly for the vertical and horizontal average, and open symbols to representative anomalies. Black line in (B) is the mean value of all the time series shown in (A). The gray-shaded area shows the range of values corresponding to the different methodologies. (C,D) show similar results, but for the intermediate layer (150–600 m), and (E,F) are the same, but for the deep layer (600 m–bottom).
Figure 6.
(A) shows all salinity time series corresponding to different data processing methods for the surface layer (0–150 m) averaged for the four geographical areas analyzed in this work: Alboran, Cape Palos, Balearic Islands, and Northern Sector. Black, blue, red, and green symbols correspond to representative anomalies, zero anomalies, zero anomalies + trend, and ignorance, respectively. Filled symbols correspond to the use of zero anomaly for the vertical and horizontal average, and open symbols to representative anomalies. Black line in (B) is the mean value of all the time series shown in (A). The gray-shaded area shows the range of values corresponding to the different methodologies. (C,D) show similar results, but for the intermediate layer (150–600 m), and (E,F) are the same, but for the deep layer (600 m–bottom).
Figure 6.
(A) shows all salinity time series corresponding to different data processing methods for the surface layer (0–150 m) averaged for the four geographical areas analyzed in this work: Alboran, Cape Palos, Balearic Islands, and Northern Sector. Black, blue, red, and green symbols correspond to representative anomalies, zero anomalies, zero anomalies + trend, and ignorance, respectively. Filled symbols correspond to the use of zero anomaly for the vertical and horizontal average, and open symbols to representative anomalies. Black line in (B) is the mean value of all the time series shown in (A). The gray-shaded area shows the range of values corresponding to the different methodologies. (C,D) show similar results, but for the intermediate layer (150–600 m), and (E,F) are the same, but for the deep layer (600 m–bottom).
Figure 7.
(A) shows the heat content time series corresponding to the different data processing methods for the surface layer (0–150 m) averaged for the four geographical areas analyzed in this work: Alboran, Cape Palos, Balearic Islands, and Northern Sector. (B,C) show similar results, but for the intermediate (150–600 m) and deep layers (600 m–bottom). (D) shows the heat content time series for the whole water column. Heat content is expressed in J × 1021.
Figure 7.
(A) shows the heat content time series corresponding to the different data processing methods for the surface layer (0–150 m) averaged for the four geographical areas analyzed in this work: Alboran, Cape Palos, Balearic Islands, and Northern Sector. (B,C) show similar results, but for the intermediate (150–600 m) and deep layers (600 m–bottom). (D) shows the heat content time series for the whole water column. Heat content is expressed in J × 1021.
Figure 8.
(A) shows the position of the four areas of study. Seasonal cycles for the SST of the four areas are presented in (B). The color of the curves coincides with the color used for the rectangles in (A). (C–F) are the time series of temperature anomalies for the four regions.
Figure 8.
(A) shows the position of the four areas of study. Seasonal cycles for the SST of the four areas are presented in (B). The color of the curves coincides with the color used for the rectangles in (A). (C–F) are the time series of temperature anomalies for the four regions.
Figure 9.
Time series of temperature anomalies at L’Estartit oceanographic station at 0 (A), 20 (B), 50 (C), and 80 m (D) depth (black lines). Linear trends expressed in °C/100 year have been inserted in these plots. The uncertainty corresponds to the 95% confidence interval taking into account the auto-correlation of the time series. The red lines are the straight lines fitted by means of least squares for estimating the linear trends.
Figure 9.
Time series of temperature anomalies at L’Estartit oceanographic station at 0 (A), 20 (B), 50 (C), and 80 m (D) depth (black lines). Linear trends expressed in °C/100 year have been inserted in these plots. The uncertainty corresponds to the 95% confidence interval taking into account the auto-correlation of the time series. The red lines are the straight lines fitted by means of least squares for estimating the linear trends.
Figure 10.
Monthly (black line) and annual (blue line) sea surface temperature time series from Fuengieola Becah. Red line shows the linear trend. The uncertainty is expressed by the 95% confidence interval taking into account the auto-correlation of the time series.
Figure 10.
Monthly (black line) and annual (blue line) sea surface temperature time series from Fuengieola Becah. Red line shows the linear trend. The uncertainty is expressed by the 95% confidence interval taking into account the auto-correlation of the time series.
Figure 12.
Coastal sea level trends (in mm/yr) estimated for the period 1993–2019 from the reconstruction based on tide-gauge data by [
7].
Figure 12.
Coastal sea level trends (in mm/yr) estimated for the period 1993–2019 from the reconstruction based on tide-gauge data by [
7].
Figure 13.
Black line in (A) is the time series of sea level at L’Estartit tide-auge. The red line is the reconstruction of sea level using a linear regression on atmospheric pressure, zonal and meridional components of the wind, and the thermosteric and halosteric contributions calculated from time series of temperature and salinity profiles. (B) shows similar results for the altimetry sea level in a grid point close to L’Estartit tide-gauge. Inserts in (A,B) show the multiple correlation coefficient and the sea level linear trend corrected for the effect of GIA.
Figure 13.
Black line in (A) is the time series of sea level at L’Estartit tide-auge. The red line is the reconstruction of sea level using a linear regression on atmospheric pressure, zonal and meridional components of the wind, and the thermosteric and halosteric contributions calculated from time series of temperature and salinity profiles. (B) shows similar results for the altimetry sea level in a grid point close to L’Estartit tide-gauge. Inserts in (A,B) show the multiple correlation coefficient and the sea level linear trend corrected for the effect of GIA.
Figure 14.
(A) shows the monthly seasonal cycles obtained from daily (black dots and line) and monthly (blue crosses) time series extending from 1981 to 2021. (B) shows the monthly seasonal cycle obtained from daily time series (black dots and line) and from three-monthly time series sampled at fixed dates from 1981 to 2021 (blue crosses). Red circles are the seasonal cycle made of 12-monthly values interpolated from the three-monthly seasonal one. (C) shows the monthly seasonal cycle obtained from daily time series (black dots and line) and from three-monthly time series sampled randomly at dates with the same distribution as dates from RADMED (including the possible existence of gaps) from 1981 to 2021 (blue crosses). Red circles show the interpolated seasonal cycle. (D) shows the monthly seasonal cycle obtained from daily time series (black dots and line) extending from 1981 to 2021, and from three-monthly time series sampled randomly at dates with the same distribution as dates from RADMED (including the possible existence of gaps) and extending from 1993 to 2021 (blue crosses). Red circles show the interpolated seasonal cycle.
Figure 14.
(A) shows the monthly seasonal cycles obtained from daily (black dots and line) and monthly (blue crosses) time series extending from 1981 to 2021. (B) shows the monthly seasonal cycle obtained from daily time series (black dots and line) and from three-monthly time series sampled at fixed dates from 1981 to 2021 (blue crosses). Red circles are the seasonal cycle made of 12-monthly values interpolated from the three-monthly seasonal one. (C) shows the monthly seasonal cycle obtained from daily time series (black dots and line) and from three-monthly time series sampled randomly at dates with the same distribution as dates from RADMED (including the possible existence of gaps) from 1981 to 2021 (blue crosses). Red circles show the interpolated seasonal cycle. (D) shows the monthly seasonal cycle obtained from daily time series (black dots and line) extending from 1981 to 2021, and from three-monthly time series sampled randomly at dates with the same distribution as dates from RADMED (including the possible existence of gaps) and extending from 1993 to 2021 (blue crosses). Red circles show the interpolated seasonal cycle.
Figure 15.
Blue dots correspond to all the temperature and salinity profiles available from Argo profilers in those areas analyzed in this work and represented by red squares. (A) corresponds to the Alboran Sea, (B) to Cape Palos, (C) to the Balearic Islands and (D) to the Northern Sector. The 200 and 500 m isobaths are included.
Figure 15.
Blue dots correspond to all the temperature and salinity profiles available from Argo profilers in those areas analyzed in this work and represented by red squares. (A) corresponds to the Alboran Sea, (B) to Cape Palos, (C) to the Balearic Islands and (D) to the Northern Sector. The 200 and 500 m isobaths are included.
Table 1.
Linear trends for the potential temperature of the surface, intermediate, and deep layers and for the whole water column averaged for the four geographical areas of Alboran, Cape Palos, Balearic Islands, and Northern Sector. Columns 3 and 5 show the 95% confidence intervals considering both the uncertainty associated with the natural variability of the system and that associated with the choice of the data processing method, i.e., [Lmin, Umax] (see
Section 2.2.2). Columns 2 and 4 show the mid-points of the 95% confidence intervals. Trends are expressed in °C/100 years.
Table 1.
Linear trends for the potential temperature of the surface, intermediate, and deep layers and for the whole water column averaged for the four geographical areas of Alboran, Cape Palos, Balearic Islands, and Northern Sector. Columns 3 and 5 show the 95% confidence intervals considering both the uncertainty associated with the natural variability of the system and that associated with the choice of the data processing method, i.e., [Lmin, Umax] (see
Section 2.2.2). Columns 2 and 4 show the mid-points of the 95% confidence intervals. Trends are expressed in °C/100 years.
Potential Temperature Trends (°C/100 Years) |
---|
| 1900–2020 | 1945–2020 |
---|
0–150 | 0.34 | [−0.09, 0.76] | 0.50 | [−0.22, 1.22] |
150–600 | 0.08 | [0.00, 0.15] | 0.23 | [0.10, 0.36] |
600–bottom | 0.13 | [0.06, 0.19] | 0.25 | [0.16, 0.35] |
Water column | 0.33 | [0.08, 0.58] | 0.12 | [−0.14, 0.39] |
Table 2.
Same as in
Table 1, except for salinity. Trends are expressed in (100 years)
−1.
Table 2.
Same as in
Table 1, except for salinity. Trends are expressed in (100 years)
−1.
Salinity Trends (100 Years)−1 |
---|
| 1900–2020 | 1945–2020 |
---|
0–150 | 0.11 | [0.06, 0.16] | 0.23 | [0.08, 0.39] |
150–600 | 0.03 | [0.01, 0.05] | 0.09 | [0.04, 0.14] |
600–bottom | 0.05 | [0.04, 0.07] | 0.11 | [0.09, 0.13] |
Water column | 0.08 | [0.04, 0.12] | 0.19 | [0.10, 0.29] |
Table 3.
Same as in
Table 1, except for density. Trends are expressed in kg m
−3/100 years.
Table 3.
Same as in
Table 1, except for density. Trends are expressed in kg m
−3/100 years.
Potential Density Trends (kg m−3/100 Years) |
---|
| 1900–2020 | 1945–2020 |
---|
0–150 | −0.01 | [−0.13, 0.11] | 0.03 | [−0.17, 0.24] |
150–600 | 0.01 | [−0.02, 0.03] | −0.01 | [−0.08, 0.06] |
600–bottom | 0.01 | [0.01, 0.02] | 0.03 | [0.02, 0.04] |
Water column | −0.08 | [−0.19, 0.02] | −0.02 | [−0.13, 0.09] |
Table 4.
Same as in
Table 1, except for heat content. Trends are expressed in W/m
2.
Table 4.
Same as in
Table 1, except for heat content. Trends are expressed in W/m
2.
Absorbed Heat (W/m2) |
---|
| 1900–2020 | 1945–2020 |
---|
0–150 | 0.03 | [−0.02, 0.08] | 0.05 | [−0.08, 0.17] |
150–600 | 0.03 | [0.00, 0.07] | 0.11 | [0.03, 0.18] |
600–bottom | 0.15 | [0.07, 0.22] | 0.34 | [0.22, 0.45] |
Water column | 0.21 | [0.05, 0.37] | 0.46 | [0.20, 0.72] |
Table 5.
Linear trends and 95% confidence intervals for the evolution of the MLD in the four selected geographical areas: Alboran Sea, Cape Palos, Balearic Islands, and Northern Sector. Trends are expressed in m/yr. Positive values indicate deepening of the mixed layer, and negative values indicate shallowing of the mixed layer. Columns 2 and 3 show the initial and final years of the time series analyzed. Columns 4 and 5 are the trends corresponding to MLD time series calculated by means of the temperature and density thresholds, respectively. This table is taken from [
17].
Table 5.
Linear trends and 95% confidence intervals for the evolution of the MLD in the four selected geographical areas: Alboran Sea, Cape Palos, Balearic Islands, and Northern Sector. Trends are expressed in m/yr. Positive values indicate deepening of the mixed layer, and negative values indicate shallowing of the mixed layer. Columns 2 and 3 show the initial and final years of the time series analyzed. Columns 4 and 5 are the trends corresponding to MLD time series calculated by means of the temperature and density thresholds, respectively. This table is taken from [
17].
| Initial Year | Final Year | t-Threshold | d-Threshold |
---|
Alboran Sea | 2006 | 2021 | 0.1 ± 0.9 | 0.1 ± 0.8 |
Cape Palos | 2004 | 2021 | 0.1 ± 0.5 | 0.3 ± 0.5 |
Balearic Islands | 2004 | 2021 | −4.5 ± 2.1 | −1.5 ± 1.1 |
Northen Sector | 2003 | 2021 | −4.8 ± 5.4 | −4 ± 5 |
Table 6.
Linear trends from tide-gauge data for the period 1948–2019 (column 3) and for the period 1993–2019 (column 4). These trends are corrected for the effect of GIA. Column 2 shows the contribution of GIA to relative sea level. Column 5 corresponds to the trends calculated from altimetry data at the grid points closest to the tide-gauge locations. * Palma results correspond to the shorter period 1997–2019.
Table 6.
Linear trends from tide-gauge data for the period 1948–2019 (column 3) and for the period 1993–2019 (column 4). These trends are corrected for the effect of GIA. Column 2 shows the contribution of GIA to relative sea level. Column 5 corresponds to the trends calculated from altimetry data at the grid points closest to the tide-gauge locations. * Palma results correspond to the shorter period 1997–2019.
Trend (mm/yr) | Tide-Gauge | Tide-Gauge | Altimetry |
---|
Location | GIA | 1948–2019 | 1993–2019 | 1993–2019 |
Tarifa | −0.2 | 1.4 ± 0.2 | 4.7 ± 0.7 | 2.5 ± 0.3 |
Algeciras | −0.2 | 1.0 ± 0.1 | 2.3 ± 0.6 | 2.4 ± 0.4 |
Ceuta | −0.2 | 0.9 ± 0.1 | 1.9 ± 0.6 | 2.4 ± 0.4 |
Málaga | −0.2 | 1.4 ± 0.2 | 3.7 ± 0.7 | 4.1 ± 0.4 |
Alicante | −0.2 | 0.8 ± 0.2 | 2.0 ± 0.8 | 3.0 ± 0.3 |
L’Estartit | 0.1 | | 2.7 ± 0.8 | 2.7 ± 0.3 |
Palma | 0.3 | | 2.0 ± 1.1 * | 1.8 ± 0.5 * |