Poster (2017, September 04)
Many studies show that the surface solar radiation has underwent large variations over the second half of the 20th century as a result of variations in cloud cover and aerosol loading in the atmosphere. However, it is difficult to build strong conclusions before the 1950s because of the observations scarcity. The evolution of the surface solar radiation has been reconstructed over 1900-2014 using the regional model MAR (« Modèle Atmosphérique Régional ») which has recently been chosen to be part of the EURO-CORDEX project, thanks to the CORDEX.be project. Simulations were performed at a horizontal resolution of 5 km over a domain of 600 x 550 km² covering Belgium. Boundary conditions were provided by four reanalysis products: ERA-interim (1979-2014) completed by the ERA40 (1958-1978), NCEP/NCAR-v1 (1948-2014), ERA-20C (1900-2010) and 20CRV2C (1900-2010). Surface solar radiation measurements from the Global Energy Balance Archive and cloud cover observations from Belgocontrol covering 1966-2007 were used for the evaluation of the MAR model and the forcing reanalyses. Results show that MAR produces much better results than the reanalyses. The driving reanalyses can generate divergent trends while they assimilate observations and are supposed to represent the same climate.
in Bulletin de la Société Géographique de Liège (2017), 68
On December 2010, several snow events allowed an exceptional snow cover over Belgium. 27 days with snow cover were observed at Uccle and snow depths of 20, 30 and 70 cm were measured on Christmas 2010 respectively at Uccle, Bierset and Mont Rigi in the Hautes-Fagnes. On December 20, while the entire Belgium was covered by a thick blanket of snow, warmer air invaded the country on December 21. This air was quickly replaced by polar air in Lower and Central Belgium (including Bierset). Heavy snowfalls were observed on December 22 and 23, except in the Upper Ardennes where rainfalls occurred under positive temperature which then dropped to -5°C. This event was due to a strong thermal inversion in the lower layers with warm air at 850 hPa above the Ardennes only. This paper aims to explain this atypical extreme event using the regional climate model MAR developed at the University of Liège
in Dahech, Salem; Charfi, Sami (Eds.) Actes du XXXe colloque de l'Association Internationale de Climatologie : Climat, ville et environnement (2017, July)
Many studies show that the surface solar radiation has underwent large variations over the second half of the 20th century as a result of variations in cloud cover and aerosol loading in the atmosphere. However, it is difficult to build strong conclusions before the 1950' because of the observations scarcity. The evolution of the surface solar radiation has been reconstructed over 1900-2014 using the regional model MAR (« Modèle Atmosphérique Régional ») in Belgium. Boundary conditions were provided by four reanalysis products : the ERA-interim (1979-2014) completed by the ERA40 (1958-1978), the NCEP/NCAR-v1 (1948-2014), the ERA-20C (1900-2010) and the 20CRV2C (1900-2010). Results show that the reanalyses can generate divergent trends while they assimilate observations and are supposed to represent the same climate.
in Dahech, Salem; Charfi, Sami (Eds.) Actes du XXXème colloque de l'Association Internationale de Climatologie - Climat, ville et environnement (2017, July)
The evolution of the snow height over the Alps can strongly impact tourism, but also the water availability of the region. In this study, we have reproduced the evolution of the climate in the Alps over the 20th century with the help of the regional atmospheric model MAR forced by three reanalyses (ERA-20C, NCEP/NCAR, and ERA-Interim). MAR shows that the snow height has increased since the beginning of the 20th century, first only at higher altitudes, then also at lower levels, before knowing a strong and abrupt decrease between 1985 and 1990. This evolution, which is consistent with observations given in the literature, is directly linked with the trends of NAO and AO. In fact, the atmospheric circulation changes highlighted by NAO and AO induce temperature and precipitation changes that directly determine the snow height in the Alps.
in Dahech, Salem; Charfi, Sami (Eds.) Actes du XXXe colloque de l'Association Internationale de Climatologie : CLIMAT, VILLE ET ENVIRONNEMENT (2017, July)
In Intertropical Africa, climate is essentially characterized by the amount of precipitation and its annual regime. These precipitations and their evolution during the period 1970-1999 are simulated thanks to the Regional Atmospheric Model (MAR), developed at the ULg, and forced by the NCEP1 reanalyses and by the outputs of three global models (GCM) of the CMIP5 database. These MAR simulations are compared to the gridded data of the Climate Research Unit (CRU). It is clear from our investigations that the simulation of the MAR model forced by the NCEP1 reanalyses is better reproducing the quantities as well as the annual rainfall regime in the semi-arid regions than in equatorial regions. On the other hand, simulations of the MAR forced by the outputs of the GCMs are globally unsatisfactory throughout the intertropical domain in terms of quantities as well as the seasonality of precipitation.
Doctoral thesis (2015)
Future projections of the atmospheric circulation over the Northern Hemisphere high latitudes, especially the North Atlantic, have high uncertainties and some of the projected changes are opposed to circulation changes that have been observed since the 2000s. In this thesis, we focus on three particular aspects of the past and projected future summertime atmospheric circulation over the broader North Atlantic region. First, we analyse whether the 2007-2012 summertime anticyclonic anomaly over the Beaufort Sea, the Canadian Arctic Archipelago, and Greenland might rather be due to global warming or to the internal variability of the atmospheric circulation by putting it in perspective with the circulation variability over the last 150 years given by five reanalysis datasets. Then, this analysis is extended for the future circulation projected towards 2100 by CMIP3 and CMIP5 General Circulation Models (GCMs) over Greenland. Finally, we evaluate the impact of the uncertainties of the future atmospheric circulation projections on the mitigating or enhancing influence of the summertime circulation changes on temperature and precipitation over Europe. We use an automatic circulation type classification to analyse in detail the atmospheric circulation changes by grouping similar daily SLP (mean sea level pressure) or Z500 (500 hPa geopotential height) fields into homogeneous circulation types. It appears that the choice of the index, on the basis of which the days are grouped together, strongly influences the characteristics of the circulation types and the kinds of circulation changes that can be detected. In comparison with Euclidean distance and pressure gradient-based indices, correlation-based indices, especially the Spearman rank correlation, are the most suitable indices when focusing on the circulation pattern. Over the Arctic region, four periods with circulation anomalies similar to that of 2007-2012 (i.e. a summertime anticyclonic anomaly over the western Arctic region) have been detected over the last 150 years, despite a higher uncertainty of the circulation given by the reanalyses due to the scarcity of observational data before 1940. Nevertheless, the 2007-2012 anomaly appears to be exceptional and several connexions with other variables, such as the North Atlantic Oscillation index and sea ice loss, suggest that it could be part of a major climatic anomaly extending beyond the Arctic region. However, the occurrence of similar periods in the past and the influence of several external and internal forcings do not allow us to attribute it to global warming. The future summertime atmospheric circulation projected by GCMs over Greenland confirms this conclusion. In fact, no significant circulation pattern changes are simulated towards 2100, besides a generalised Z500 increase caused by the projected warming. Since GCMs are able to simulate atmospheric circulation changes over other regions and since the atmospheric circulation itself is influenced by other variables, such as sea ice or snow extent, which are already impacted by long-term changes, we conclude that the 2007-2012 anomaly could rather be attributed to the internal variability of the climatic system. Finally, we evidence that projected future atmospheric circulation changes impact on the SLP and precipitation changes simulated over Europe towards 2100 for summer. Over north-western Europe, these circulation changes could mitigate the SLP decrease by around 50 % and cancel out the precipitation increase. Nevertheless, high uncertainties among the GCMs on the magnitude and even on the sign of these changes cast doubt on the reliability of these projections. On the other hand, future atmospheric circulation changes are not projected to affect significantly the warming and drying simulated for the next decades over the Mediterranean region and eastern Europe.
in Cryosphere (2015), 9
A significant increase in the summertime occurrence of a high pressure area over the Beaufort Sea, the Canadian Arctic Archipelago, and Greenland has been observed since the beginning of the 2000s, and particularly between 2007 and 2012. These circulation anomalies are likely partly responsible for the enhanced Greenland ice sheet melt as well as the Arctic sea ice loss observed since 2007. Therefore, it is interesting to analyse whether similar conditions might have happened since the late 19th century over the Arctic region. We have used an atmospheric circulation type classification based on daily mean sea level pressure and 500 hPa geopotential height data from five reanalysis data sets (ERA-Interim, ERA-40, NCEP/NCAR, ERA-20C, and 20CRv2) to put the recent circulation anomalies in perspective with the atmospheric circulation variability since 1871. We found that circulation conditions similar to 2007–2012 have occurred in the past, despite a higher uncertainty of the reconstructed circulation before 1940. For example, only ERA-20C shows circulation anomalies that could explain the 1920–1930 summertime Greenland warming, in contrast to 20CRv2. While the recent anomalies exceed by a factor of 2 the interannual variability of the atmospheric circulation of the Arctic region, their origin (natural variability or global warming) remains debatable.
in Camberlin, Pierre; Richard, Yves (Eds.) Actes du XXVIIème colloque de l'Association de Climatologie - Climat : système & interactions (2014, June)
Over the five last decades, the reanalyses (ERA and NCEP/NCAR) show a strengthening of the pressure gradient between the southern hemisphere subtropical anticyclone belt and the southern circumpolar lows during summer. With the help of an automatic circulation type classification, we show that the strengthening of the pressure gradient is generalised to all circulation types and, paradoxically, it does not cause circulation changes. It is probably implied by the strengthening of the temperature gradient between the tropics and the South Pole, without consequences on the general circulation. Our classification also allows a successful comparison between the two reanalyses in a region where the observation data are rare.
in International Journal of Climatology (2014)
Future climate change projections are not limited to a simple warming, but changes in precipitation and sea level pressure (SLP) are also projected. The SLP changes and the associated atmospheric circulation changes could directly mitigate or enhance potential projected changes in temperature and precipitation associated with rising temperatures. With the aim of analysing the projected circulation changes and their possible impacts on temperature and precipitation over Europe in summer [June–July–August (JJA)], we apply an automatic circulation type classification method, based on daily SLP, on general circulation model (GCM) outputs from the Coupled Model Intercomparison Project phase 5 (CMIP5) database over the historical period (1951–2005) and for climate under two future scenarios (2006–2100). We focus on summer as it is the season when changes in temperature and precipitation have the highest impact on human health and agriculture. Over the historical observed reference period (1960–1999), our results show that most of the GCMs have significant biases over Europe when compared to reanalysis data sets, both for simulating the observed circulation types and their frequencies, as well as for reproducing the intraclass means of the studied variables. The future projections suggest a decrease of circulation types favouring a low centred over the British Isles for the benefit of more anticyclonic conditions. These circulation changes mitigate the projected precipitation increase over north-western Europe in summer, but they do not significantly affect the projected temperature increase and the precipitation decrease over the Mediterranean region and eastern Europe. However, the circulation changes and the associated precipitation changes are tarnished by a high uncertainty among the GCM projections.
in Cryosphere (2013), 7
Since 2007, there has been a series of surface melt records over the Greenland ice sheet (GrIS), continuing the trend towards increased melt observed since the end of the 1990's. The last two decades are characterized by an increase of negative phases of the North Atlantic Oscillation (NAO) favouring warmer and drier summers than normal over GrIS. In this context, we use a circulation type classification based on daily 500 hPa geopotential height to evaluate the role of atmospheric dynamics in this surface melt acceleration for the last two decades. Due to the lack of direct observations, the interannual melt variability is gauged here by the summer (June–July–August) mean temperature from reanalyses at 700 hPa over Greenland; analogous atmospheric circulations in the past show that ~70% of the 1993–2012 warming at 700 hPa over Greenland has been driven by changes in the atmospheric flow frequencies. Indeed, the occurrence of anticyclones centred over the GrIS at the surface and at 500 hPa has doubled since the end of 1990's, which induces more frequent southerly warm air advection along the western Greenland coast and over the neighbouring Canadian Arctic Archipelago (CAA). These changes in the NAO modes explain also why no significant warming has been observed these last summers over Svalbard, where northerly atmospheric flows are twice as frequent as before. Therefore, the recent warmer summers over GrIS and CAA cannot be considered as a long-term climate warming but are more a consequence of NAO variability affecting atmospheric heat transport. Although no global model from the CMIP5 database projects subsequent significant changes in NAO through this century, we cannot exclude the possibility that the observed NAO changes are due to global warming.
in Climate Dynamics (2013), 41(7-8),
The Greenland ice sheet is projected to be strongly affected by global warming. These projections are either issued from downscaling methods (such as Regional Climate Models) or they come directly from General Circulation Models (GCMs). In this context, it is necessary to evaluate the accuracy of the daily atmospheric circulation simulated by the GCMs, since it is used as forcing for downscaling methods. Thus, we use an automatic circulation type classification based on two indices (Euclidean distance and Spearman rank correlation using the daily 500 hPa geopotential height) to evaluate the ability of the GCMs from both CMIP3 and CMIP5 databases to simulate the main circulation types over Greenland during summer. For each circulation type, the GCMs are compared to three reanalysis datasets on the basis of their frequency and persistence differences. For the current climate (1961–1990), we show that most of the GCMs do not reproduce the expected frequency and the persistence of the circulation types and that they simulate poorly the observed daily variability of the general circulation. Only a few GCMs can be used as reliable forcings for downscaling methods over Greenland. Finally, when applying the same approach to the future projections of the GCMs, no significant change in the atmospheric circulation over Greenland is detected, besides a generalised increase of the geopotential height due to a uniform warming of the atmosphere.
in Bigot, Sylvain; Rome, Sandra (Eds.) XXVème colloque de l'Association Internationale de Climatologie - Les climats régionaux : observation et modélisation (2012, September)
The IPCC projects more frequent and longer heat waves and droughts during summer for future over Western Europe. These extreme events occur during anticyclonic blocking events. We use atmospheric circulation type classifications to determine if the models project an increase of the number and the persistence of these anticyclonic blockings. For recent climate, the number of blocking events depends on the ability of the models to reproduce the observed general circulation. The future projections do not show any systematic evolution of the number of anticyclonic blockings over Western Europe. Nevertheless, other changes like an increase of the temperature will lead to more frequent heat waves and droughts.
Poster (2012, April 26)
Some studies show that most General Circulation Models (GCMs) have difficulties to simulate the main observed circulation patterns and their frequencies. However, this does not only impact the GCM based projections for future climate, but also the results of downscaling methods using the circulation simulated by GCMs as forcing. Indeed, the downscaling methods are not able to correct the biases introduced by the GCM simulations in the free atmosphere. Here, we focus on the anticyclonic blocking situations over western Europe for summer (June, July and August). Indeed, these blocking situations, which are often related to droughts and heat waves, could become more frequent due to global warming. Moreover, their frequency and persistence depend on the variability of the circulation, which is known to be difficult to reproduce by the GCMs. In order to evaluate the ability of the GCMs to reproduce the observed frequency and persistence of blocking situations, we compare them with two reanalysis datasets (NCEP-NCAR 1 and ECMWF ERA-40) by using an automatic circulation type classification. The daily geopotential height at 500 hPa over the last 30 years of the current climate simulation (Historical experiment, 1976-2005) of all available CMIP5 GCMs prepared for the upcoming IPCC report AR5 is used here. The circulation type classification groups similar daily circulation situations together on basis of a leader-algorithm to obtain a few homogeneous circulation types describing the general circulation of the region. Thus, the frequency and the persistence of each circulation type can be analysed on a daily timescale. We show that the ability of the GCMs to reproduce the observed frequency and persistence of blocking situations is influenced by the anomalies in their circulation type frequency repartition. So, the GCMs which underestimate the frequency of the anticyclonic types tend to simulate less and shorter blocking situations. The contrary is observed for GCMs that overestimate the frequency of these circulation types. This rises questions about the reliability of the future projections for events related to blocking situations. Indeed, when applying the same approach as for the current climate to the future projections (experiments RCP4.5 and RCP8.5), it seems that the blocking situations become more frequent and persistent. However, when considering only the circulation patterns by removing the mean geopotential height increase due to global warming, there is no significant circulation change till 2100. This means that the GCMs conserve their circulation biases in spite of climate change and so, the frequency and the persistence of the blocking situations are projected to remain almost the same as those simulated for the current climate.
in Fazzini, Massimiliano; Beltrando, Gérard (Eds.) XXIVème colloque de l'Association Internationale de Climatologie - Climat montagnard et risques (2011, September)
Atmospheric circulation simulations from general circulation models are used as forcing for downscaling methods and for future projections. Thus, it is essential to evaluate them. An automatic circulation type classification is applied to daily 500 hPa geopotential height data. Firstly, the classification is done for the NCEP-NCAR 1 reanalysis, and then the main circulation types are imposed to the simulations of six general circulation models. For recent climate (20C3M scenario), it appears that most models are not able to simulate well the circulation over western Europe, due to biases in the mean geopotential height and an underestimation of the circulation variability. For future climate (A1B scenario), a general increase of the geopotential height is projected, leading to the emergence of new circulation types.
in Blanco, Juan; Kheradmand, Houshang (Eds.) Climate Change - Geophysical Foundations and Ecological Effects (2011)
We present here future projections of the Greenland climate performed by the regional climate model MAR coupled with a snow model and forced by two scenarios of greenhouse gas emissions from the global model CanESM2 of the next IPCC assessment report (AR5). Knowing that MAR forced by CanESM2 over the current climate (1970-1999) compares well with the reference MAR simulation performed by using the ERA-40 reanalysis as forcing, this gives us confidence in our future projections. For the RCP4.5 scenario (optimistic) and respectively RCP8.5 scenario (pessimistic), MAR projects a sea level rise in 2100 of 6.5 +/- 1.5 cm and respectively 14+/-2 cm as result of increasing surface melt of the Greenland ice sheet over 2000-2100. It is true that MAR projects a small increase of snowfall in the winter because the atmosphere will be warmer and therefore can contain more water vapor. But this is not sufficient to offset the acceleration of melt, notably for the scenario RCP8.5 which projects an increase of 10 °C in 2100 above the ice sheet. This work ﬁts in the ICE2SEA project (http://www.ice2sea.eu) of the 7th Framework Program (FP7) which aims to improve the projections of the continental ice melting contribution to sea level rise.
Conference (2011, April 08)
Downscaling methods forced by General Circulation Model (GCM) simulations are not able to correct the biases in the general circulation simulated by the GCMs. Moreover, since the GCMs have a coarse spatial resolution, they have difficulties to simulate reliably ground variables like temperature and precipitation which are affected by topography, land use and local features. So, we can attempt that they simulate better the large-scale atmospheric circulation. That is why it is of special interest to evaluate the GCM simulations of atmospheric circulation for current climate by comparing them with the NCEP-NCAR 1 and the ECMWF reanalysis data over 1961-1990. This analysis is done over western Europe for summer (JJA) and winter (DJF) for the GCMs (available on http://cmippcmdi.llnl.gov/cmip5/) proposed by the IPCC for its upcoming report (AR5). The method used is an automated circulation type classification based on the daily geopotential height at 500 hPa. It is a leader-algorithm correlation based method taking part of the COST733CAT classification catalogue. Unlike the usually used methods based on the monthly mean circulation, this approach allows a precise analysis of each circulation type. So, it gives much more information on the ability of the GCMs to simulate the different circulation types and consequently the climatic variability of a region. In order to allow a direct comparison between the GCM simulations and the reanalysis data, the classification is done first only for the reanalysis dataset over 1961-1990. Then, the main types individualised here are imposed for the classification of the GCM outputs. Since the circulation types are the same, the comparison between the datasets can be made on the basis of the differences of the frequency distribution throughout the classes. Moreover, the mean intraclass repartition of the circulation situations may differ from one dataset to another. So, the study of this mean and its standard deviation gives an idea of the differences between the reanalysis and the GCMs within each class. Firstly, this approach is applied to current climate (1961-1990) for evaluating the ability of the GCMs driven by the historical experiment to simulate the climate of the last decades over western Europe. In fact, if one GCM is not able to reproduce reliably the main characteristics of the current climate, its future projections may be questionable. Then, the best matching GCMs are retained and the same approach is applied to the future simulations driven by RCP concentrations or emissions (2011-2040, 2041-2070 and 2071-2100). So, the evolution of the frequency of the circulation types and maybe the appearance of new types can be analysed under climate change conditions. Moreover, it is interesting to compare the uncertainty of the current climate simulations to the projected changes for future climate. If the uncertainty is of the same order or higher than the projected changes, the reliability of the simulations for future climate may be very questionable.
Conference (2011, April 06)
The regional climate models MAR and RCMO show that the surface mass balance (SMB) rate of the whole Greenland ice sheet (GrIS) is the lowest in 2010 since 50 years. This record is a combination of an abnormal dry year and an exceptional melt in summer confirmed by ground measurements and satellite-derived observations. An automated circulation type classification (CTC) is used for detecting anomalies in the daily atmospheric circulation at 500hPa over the Greenland ice sheet during 2010. The CTC reveals that the low snow accumulation is due to the general circulation (negative NAO index) while the record melt in summer is rather a consequence of the well known surface albedo-temperature feedback induced by - a warmer and thinner than normal snowpack above the bare ice at the end of the spring. - an earlier beginning of the melt season. - a drier summer. - an exceptional persistence of atmospheric circulations inducing warm and dry conditions over the GrIS. All these anomalies induced in summer 2010 an exceptional time exposure of bare ice areas (with a lower albedo than snow) over the GrIS which impacts the surface melt. Sensitivity experiments carried out by the MAR model allow to estimate the importance of each anomaly in the record simulated melt of summer 2010.
Poster (2011, April 05)
Future projections of the Greenland ice sheet melt are based on General Circulation Model (GCM) simulations. In particular, the reliability of downscaling methods forced by these simulations depends on the quality of the atmospheric circulation simulated by GCMs. Therefore, it is essential to analyse and evaluate the GCMs modelled general circulation for current climate (1961-1990). Atmospheric circulation type classifications offer a very interesting approach for evaluating the GCM-based circulation at a daily time scale compared to the most used methods based only on monthly means. Indeed, the circulation type classification allows a precise and detailed analysis of each circulation type and so, it gives much more information on the ability of GCMs to simulate the different circulation types and consequently the climatic variability of a region. In fact, exceptional circulation events over Greenland, which cannot be taken into account by the monthly mean approach, have much more impact on the melt than the mean atmospheric state. Thus, an automated correlation-based atmospheric circulation type classification (CTC) is used for evaluating the new GCM outputs (available on http://cmip-pcmdi.llnl.gov/cmip5/) computed for the upcoming IPCC report (AR5). The daily geopotential height at 500 hPa simulations of the GCMs for current climate are compared to the NCEP-NCAR 1 and the ECMWF reanalysis data for the summer months (JJA), when melt is the most important. To achieve this, the classification is first done for the reanalysis data over 1961-1990 and afterwards, the types of the reanalysis based CTC are imposed for classifying the GCM datasets over 1961-1990 (from the historical experiment) to allow a direct type per type comparison based on the frequency distribution of each dataset. This approach also gives the opportunity to study the intraclass repartition differences between the reanalysis and the GCMs. After the evaluation of the GCM simulations for current climate, the future projections driven by RCP concentrations or emissions (2011-2040, 2041-2070 and 2071-2100) from the best matching GCMs are analysed in the same way. For current climate, it clearly appears that only a few GCMs are able to reproduce reliably the variability of the atmospheric circulation over Greenland during summer. The differences of frequency between the GCMs and the reanalysis are mainly due to biases of the geopotential height which is systematically over or underestimated by most GCMs and to the underestimation of the variability of the circulation by most GCMs. For future projections, no new circulation types are detected, but rather a general increase of the mean geopotential height regardless of the circulation type. It is also important to note that for many GCMs, the uncertainty of the current climate simulations (given by the differences of the classification results between the GCM simulations for current climate and the reanalysis data for the same time) are of the same order than the projected changes for future climate. Therefore, these projections may be questionable.
Conference (2011, April 05)
Abstract. As part of the ICE2SEA project, the regional climate model MAR was forced by the general circulation model ECHAM5 for making future projections of the Greenland Ice Sheet (GrIS) Surface Mass Balance (SMB) over 1980-2099 at a resolution of 25km. For the A1B scenario, MAR projects a highly negative (-500 GT/yr) SMB rate at the end of this century and a induced mass loss corresponding to a sea level rise of ~7 cm over 2000-2100. However, the comparison with MAR forced by the ERA-40 reanalysis over 1980-1999 shows that MAR forced by the 20C3M scenario is not able to represent reliably the current SMB due to biases in the general circulation and in the free atmosphere summer temperature modeled by ECHAM5 around the GrIS. These biases induce in MAR an underestimation of the snow accumulation and an overestimation of the surface melt. Therefore, this questions the reliability of these ECHAM5-forced future projections, knowing that i) these biases could be amplified in future and that ii) the MAR outputs are used to force ice sheets models for the ICE2SEA project. That is why, by waiting the outputs from the next generation of GCMs (CMIP5), we investigate the impacts of current climate biases over the future projections and we suggest corrections of ECHAM5 forcing files for having a better agreement with the ERA-40 forced simulation. This is useful for the ice sheet model wanting to use the absolute values of MAR future projections instead of anomalies.
Conference (2010, November 23)
Future projections of the Greenland ice sheet melt are mainly based on General Circulation Model (GCM) outputs. The atmospheric circulation type classification offers a unique opportunity for validating the GCM-based circulations. Six GCMs used in the last IPCC report are analysed here. A correlation-based classification is constructed for each model using daily geopotential height at 500 hPa over Greenland. It is applied to a dataset combining the GCM-based outputs (20C3M scenario) for the current climate and the NCEP-NCAR 1 reanalysis data over the period 1961-1990 allowing a direct comparison for each circulation type. Most of the analysed models are able to reproduce the main circulation types, but they fail to reproduce their frequencies because they underestimate the climate variability. In addition, some biases in the mean geopotential height remain. However, we use our atmospheric circulation type classification for analysing future projections made by GCMs. As for the 20th century climate, a combined classification is made integrating reanalysis data over 1971-1990, GCM-based outputs over 1971-1990 (using 20C3M scenario) and GCM-based outputs over 2046-2065 and 2081-2100 (using A1B scenario). No new circulation types are individualised knowing that the main changes are just a general increase of the geopotential height. Furthermore, the changes in frequency observed between the 20th century climate and the first future period (2046-2065) are of the same order than the uncertainties of the models for simulating the current climate by comparison with the reanalysis data. Therefore, the circulation type classification is a useful tool to give a precise analysis of the atmospheric circulation simulated by GCMs knowing that most of downscaling techniques are dependent on the general circulation simulated by the GCMs.
Conference (2010, November 05)
The variability of the geopotential height at 500 hPa simulated by General Circulation Models (GCMs) over Greenland is evaluated using an atmospheric circulation type classification. The GCM outputs for the current climate (20C3M) are first compared to reanalysis data over 1961-1990. The comparison shows that most of them simulate well the main circulation types but fail to reproduce their frequencies because of underestimations of circulation variability and biases in the mean geopotential height. GCM-based future projections do not individualise new circulation types but show a general increase of the geopotential height. Based on this approach, the correlation between surface temperature and atmospheric circulation offers a new way for estimating the Greenland ice sheet melt.