Limits...
Understanding the causes of recent warming of mediterranean waters. How much could be attributed to climate change?

Macias D, Garcia-Gorriz E, Stips A - PLoS ONE (2013)

Bottom Line: These SST results show a clear sinusoidal tendency that follows the Atlantic Multidecadal Oscillation (AMO) during the simulation period.Our results reveal that 58% of recent warming in Mediterranean waters could be attributed to this AMO-like oscillation, being anthropogenic-induced climate change only responsible for 42% of total trend.It has been proposed that this change in the AMO phase will mask the effect of global warming in the forthcoming decades, and our results indicate that the same could also be applicable to the Mediterranean Sea.

View Article: PubMed Central - PubMed

Affiliation: European Commission, Joint Research Center, Institute for Environment and Sustainability, Water Research Unit, Ispra, Italy.

ABSTRACT
During the past two decades, Mediterranean waters have been warming at a rather high rate resulting in scientific and social concern. This warming trend is observed in satellite data, field data and model simulations, and affects both surface and deep waters throughout the Mediterranean basin. However, the warming rate is regionally different and seems to change with time, which has led to the question of what causes underlie the observed trends. Here, we analyze available satellite information on sea surface temperature (SST) from the last 25 years using spectral techniques and find that more than half of the warming tendency during this period is due to a non-linear, wave-like tendency. Using a state of the art hydrodynamic model, we perform a hindcast simulation and obtain the simulated SST evolution of the Mediterranean basin for the last 52 years. These SST results show a clear sinusoidal tendency that follows the Atlantic Multidecadal Oscillation (AMO) during the simulation period. Our results reveal that 58% of recent warming in Mediterranean waters could be attributed to this AMO-like oscillation, being anthropogenic-induced climate change only responsible for 42% of total trend. The observed acceleration of water warming during the 1990s therefore appears to be caused by a superimposition of anthropogenic-induced warming with the positive phase of the AMO, while the recent slowdown of this tendency is likely due to a shift in the AMO phase. It has been proposed that this change in the AMO phase will mask the effect of global warming in the forthcoming decades, and our results indicate that the same could also be applicable to the Mediterranean Sea. Henceforth, natural multidecadal temperature oscillations should be taken into account to avoid underestimation of the anthropogenic-induced warming of the Mediterranean basin in the future.

Show MeSH
Satellite data – model simulation comparison for the period from 1985 to 2009.a) Taylor Diagram of the model-satellite data comparison. The red star corresponds to the spatial distribution comparison. The yellow square corresponds to the temporal evolution comparison. b) Time series of the observed (from satellite) monthly mean SST (blue line) and simulated monthly mean SST (red line) during the period from 1985 to 2009.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3842300&req=5

pone-0081591-g002: Satellite data – model simulation comparison for the period from 1985 to 2009.a) Taylor Diagram of the model-satellite data comparison. The red star corresponds to the spatial distribution comparison. The yellow square corresponds to the temporal evolution comparison. b) Time series of the observed (from satellite) monthly mean SST (blue line) and simulated monthly mean SST (red line) during the period from 1985 to 2009.

Mentions: The results of the model corresponding to the years between 1985 and 2009 were used to validate our model against satellite values (Fig. 2). The climatological SST distribution during this time period was reasonably simulated by the model, showing a pearson-correlation coefficient with the observed patterns of r = 0.927 (p<0.001), a mean difference of 0.18°C and a very similar standard deviation (representing spatial heterogeneity) in the model and data (Fig. 2a). For the purpose of the analysis presented here, however, we are more interested in how the model reproduces the temporal evolution of the mean SST observed in the satellite data (Fig. 2b). In this case, mean SST values are very similar, with a mean difference between the model and data of only 0.0096°C being detected throughout the analyzed time-span. The pearson-correlation coefficient is r = 0.9935 (p<0.001), with the standard deviation (now representing temporal variability) being almost identical in the model and data (Fig. 2A).


Understanding the causes of recent warming of mediterranean waters. How much could be attributed to climate change?

Macias D, Garcia-Gorriz E, Stips A - PLoS ONE (2013)

Satellite data – model simulation comparison for the period from 1985 to 2009.a) Taylor Diagram of the model-satellite data comparison. The red star corresponds to the spatial distribution comparison. The yellow square corresponds to the temporal evolution comparison. b) Time series of the observed (from satellite) monthly mean SST (blue line) and simulated monthly mean SST (red line) during the period from 1985 to 2009.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3842300&req=5

pone-0081591-g002: Satellite data – model simulation comparison for the period from 1985 to 2009.a) Taylor Diagram of the model-satellite data comparison. The red star corresponds to the spatial distribution comparison. The yellow square corresponds to the temporal evolution comparison. b) Time series of the observed (from satellite) monthly mean SST (blue line) and simulated monthly mean SST (red line) during the period from 1985 to 2009.
Mentions: The results of the model corresponding to the years between 1985 and 2009 were used to validate our model against satellite values (Fig. 2). The climatological SST distribution during this time period was reasonably simulated by the model, showing a pearson-correlation coefficient with the observed patterns of r = 0.927 (p<0.001), a mean difference of 0.18°C and a very similar standard deviation (representing spatial heterogeneity) in the model and data (Fig. 2a). For the purpose of the analysis presented here, however, we are more interested in how the model reproduces the temporal evolution of the mean SST observed in the satellite data (Fig. 2b). In this case, mean SST values are very similar, with a mean difference between the model and data of only 0.0096°C being detected throughout the analyzed time-span. The pearson-correlation coefficient is r = 0.9935 (p<0.001), with the standard deviation (now representing temporal variability) being almost identical in the model and data (Fig. 2A).

Bottom Line: These SST results show a clear sinusoidal tendency that follows the Atlantic Multidecadal Oscillation (AMO) during the simulation period.Our results reveal that 58% of recent warming in Mediterranean waters could be attributed to this AMO-like oscillation, being anthropogenic-induced climate change only responsible for 42% of total trend.It has been proposed that this change in the AMO phase will mask the effect of global warming in the forthcoming decades, and our results indicate that the same could also be applicable to the Mediterranean Sea.

View Article: PubMed Central - PubMed

Affiliation: European Commission, Joint Research Center, Institute for Environment and Sustainability, Water Research Unit, Ispra, Italy.

ABSTRACT
During the past two decades, Mediterranean waters have been warming at a rather high rate resulting in scientific and social concern. This warming trend is observed in satellite data, field data and model simulations, and affects both surface and deep waters throughout the Mediterranean basin. However, the warming rate is regionally different and seems to change with time, which has led to the question of what causes underlie the observed trends. Here, we analyze available satellite information on sea surface temperature (SST) from the last 25 years using spectral techniques and find that more than half of the warming tendency during this period is due to a non-linear, wave-like tendency. Using a state of the art hydrodynamic model, we perform a hindcast simulation and obtain the simulated SST evolution of the Mediterranean basin for the last 52 years. These SST results show a clear sinusoidal tendency that follows the Atlantic Multidecadal Oscillation (AMO) during the simulation period. Our results reveal that 58% of recent warming in Mediterranean waters could be attributed to this AMO-like oscillation, being anthropogenic-induced climate change only responsible for 42% of total trend. The observed acceleration of water warming during the 1990s therefore appears to be caused by a superimposition of anthropogenic-induced warming with the positive phase of the AMO, while the recent slowdown of this tendency is likely due to a shift in the AMO phase. It has been proposed that this change in the AMO phase will mask the effect of global warming in the forthcoming decades, and our results indicate that the same could also be applicable to the Mediterranean Sea. Henceforth, natural multidecadal temperature oscillations should be taken into account to avoid underestimation of the anthropogenic-induced warming of the Mediterranean basin in the future.

Show MeSH