PROJECT PERIODIC REPORT

Transcription

1 PROJECT PERIODIC REPORT Grant Agreement number: Project acronym: EURO4M Project title: European Reanalysis and Observations for Monitoring Funding Scheme: collaborative project Date of latest version of Annex I against which the assessment will be made: 27 Jan 2010 Periodic report: 1 st 2 nd 3 rd 4 th Period covered: from 1 April 2010 to 31 March 2011 Name, title and organisation of the scientific representative of the project's coordinator 1 : Dr. A.M.G. Klein Tank Royal Netherlands Meteorological Institute Tel: Fax: kleintan@knmi.nl Project website 2 address: 1 Usually the contact person of the coordinator as specified in Art of the Grant Agreement. 2 The home page of the website should contain the generic European flag and the FP7 logo which are available in electronic format at the Europa website (logo of the European flag: logo of the 7th FP: The area of activity of the project should also be mentioned.

2 Declaration by the scientific representative of the project coordinator I, as scientific representative of the coordinator of this project and in line with the obligations as stated in Article II.2.3 of the Grant Agreement declare that: The attached periodic report represents an accurate description of the work carried out in this project for this reporting period; The project (tick as appropriate) 3 : has fully achieved its objectives and technical goals for the period; has achieved most of its objectives and technical goals for the period with relatively minor deviations. has failed to achieve critical objectives and/or is not at all on schedule. The public website, if applicable is up to date is not up to date To my best knowledge, the financial statements which are being submitted as part of this report are in line with the actual work carried out and are consistent with the report on the resources used for the project (section 3.4) and if applicable with the certificate on financial statement. All beneficiaries, in particular non-profit public bodies, secondary and higher education establishments, research organisations and SMEs, have declared to have verified their legal status. Any changes have been reported under section (Project Management) in accordance with Article II.3.f of the Grant Agreement. Name of scientific representative of the Coordinator: Dr A.M.G. Klein Tank Data 14 October 2011 For most of the projects, the signature of this declaration could be done directly via the IT reporting tool through an adapted IT mechanism. 3 If either of these boxes below is ticked, the report should reflect these and any remedial actions taken. 2

3 3.1 Publishable summary Objective: The goal of EURO4M is to meet the need for timely and reliable monitoring information about the state of the climate in Europe. User consultations indicate that a clear need exists for long-term data, and particularly also for value-added data products which describe the evolution of the physical climate system on a European scale. Context: EURO4M develops regional reanalyses of past weather and user-oriented data products for monitoring climate variability and change in Europe. The project addresses the situation of fragmentation and scarcity of long-term climate change monitoring information. It does so by combining seamlessly the comprehensive data sets from model-based regional reanalyses and the Essential Climate Variable (ECV) data sets from satellites and ground-based stations. EURO4M will develop the capacity for, and deliver the best possible and most complete (gridded) climate change time series and monitoring services covering all of Europe. Reference historical databases: New regional reanalyses are produced as part of the project using the state-of-the-art regional weather models HIRLAM and NAE. The great benefit of a reanalysis is that it provides a complete picture of the atmosphere, covering the whole of the 3-dimensional domain, not only of the observed variables, but also of those that are not directly measured. In order to further enhance the km resolution of the regional reanalyses to the local scale, 2-dimensional downscaling is performed using the systems MESAN and SAFRAN. These high-resolution analysis systems employ regional variations given by observational statistics and physiographic factors such as landsea mask and orography. This results in downscaled gridded climate time series of surface variables at about 5 km resolution. Due to computational constraints and input data limitations, the regional reanalyses and downscaled data sets typically cover a time period up to 20 years. For climate change applications in risk management and science-based adaptation most users need information about longer-term changes. In particular, information about climate trends and changing probabilities of high impact extremes (such as flooding or heat waves) cannot be derived from relatively short reanalyses data sets only. Therefore, the reanalyses will be combined with multidecadal satellite data sets and century scale in situ observations. Long-term gridded climate time series based on satellite data and interpolated station observations include the updated and extended versions of the monthly GPCC precipitation data set, the monthly CRU temperature, precipitation and humidity data set, and the daily ECA&D data sets of multiple ECVs. Within EURO4M, the European domain consists of the area 25W-45E and 30N-75N (minimum required) or 30W-60E and 30N-85N (desired). Climate Indicator Bulletins (CIBs): The reference historical databases enable us to put the observed high-impact weather and climate extremes in a long-term historical context. To guide this process, so-called Climate Indicator Bulletins (CIBs) are being developed which consist of knowledge abstractions from different data sets including associated uncertainty estimates. By integrating the different data sources, these bulletins improve the climate change services for society. CIBs will also be issued in near-real-time during emerging extreme events. 3

4 Future GMES service: In conclusion, the EURO4M project will extend, in a cost effective manner, European capacity to systematically monitor climate variability and change (including extremes) on a range of space and time scales. EURO4M will reach out with innovative and integrated data products and climate change services to policy-makers, researchers, planners and citizens at European, national and local levels. This will directly address the needs of, for instance, the European Environment Agency for their environmental assessment reports - and even provide online reporting during emerging extreme events. As the primary source of timely, targeted and reliable information about the state of the climate in Europe, EURO4M is an important building block for GMES. The project will integrate and extend core services activities on ECVs, specifically developing the capacity required for state-of-the-art user-oriented products for monitoring climate change. EURO4M has the potential to evolve into a future GMES service on climate change monitoring that is fully complimentary and supporting the existing operational services. Year-1 progress: In year 1, all partners started to work on their tasks. Good progress has been made towards the general objectives of the project, and all DoW deliverables have been submitted. In particular, existing station-based gridded datasets for Europe (CRU, GPCC, and E-OBS) have been further developed, updated and published. Project partners have worked with existing data recovery and digitisation activities to improve the ECV databases for the European region (linking with the international ACRE initiative and the MEDARE initiative for the Mediterranean). In addition, model-based European regional reanalysis capabilities have been further developed. First evaluations with some of these systems against in situ observations show promising results. A comparison of existing ERAMESAN and SAFRAN methods for further downscaling has been completed and work has started to improve the input data for the regional reanalyses in the project. With respect to user-oriented information and products, the so-called Climate Liaison Team (CLT) has been set up. This team will implement a user feedback loop by compiling user requirements for all GEO areas (e.g. define the most crucial climatological parameters and how they should be summarized) and providing scientific guidance on the best ways to disseminate and use the EURO4M products and services. The CLT will further evolve when user feedback is provided to shape the first Climate Indicator Bulletin (CIB) which will contain user-oriented information and multi-purpose climate change products related to European temperature. This CIB is due for Month 24 and will serve the full range of climate users and applications sectors in Europe within the wider global community. Project consortium: EURO4M is a collaborative effort of 9 European partners. The Management Board of the EURO4M project is made up of the following persons: Albert Klein Tank (Royal Netherlands Meteorological Institute, Coordinator), Dale Barker (Met Office, United Kingdom), Roxana Bojariu (National Meteorological Administration, Romania), Phil Jones (Climatic Research, University of East Anglia, United Kingdom). The EU Project Officer is Stijn Vermoote (REA, Belgium). The Principle Investigators for the other project partners are: Manola Brunet (University Rovira i Virgili, Spain), Christoph Frei (Meteo Swiss, Switzerland), Richard Mueller (Deutscher Wetterdienst, Germany), Per Unden (Swedish Meteorological and Hydrological Institute, Sweden), and Eric Bazile (Météo France, France). The Advisory Board of EURO4M is made up of the following persons: Hans-Eduard Hauser (DG Climate, EU, Belgium), Andre Jol (EEA, Denmark), and Adrian Simmons, ECMWF, United Kingdom. The Project Office at KNMI (The Netherlands), which is responsible for the routine administration of the project and the scientific direction, is staffed by Karin van der Schaft, Gé Verver and Albert Klein Tank. 4

5 Project website: Core of the report for the period: Project objectives, work progress and achievements, project management Project objectives for the period The overall goal of EURO4M is to develop the capacity for, and deliver the best possible and most complete (gridded) climate change time series and monitoring services covering all of Europe. These will enable adequate descriptions of the status and evolution of the Earth system components. Specifically, the Year-1 objectives are to: WP1: REGIONAL OBSERVATION DATASETS further develop station-based gridded datasets for Europe in its entirety and selected sub-regions assess the capacity of additional satellite-derived data for climate monitoring (linked to the EUMETSAT-SAFs), examining the parameters to be monitored, drawbacks and strengths of the methods and approaches used, the capacity for monitoring patterns of anomalies in time and across space, the capacity for trend analysis, and the capacity for the appropriate merging of satellite data with ground-truth (i.e. in situ based observations); work with existing data recovery, rescue, imaging and digitisation activities to improve and temporally extend the ECV databases for the European region (linking with WG1 of the international ACRE initiative plus EuroCryoClim in the Arctic, EUMETNET-ECSN for Europe, and MEDARE for the Mediterranean), in order to coordinate and make accessible currently available historical climate data; WP2: REGIONAL REANALYSIS develop a state-of-the-art system capable of a comprehensive regional reanalysis and 2D downscaling; review available observations and prepare for a longer reanalysis period improve capability for reanalysis through better input data WP3: USER-ORIENTED INFORMATION AND CLIMATE CHANGE PRODUCTS produce innovative and integrated climate change products that are multi-purpose, but based on user requirements, and form the basis for future GMES implementation services manage and operate the European Climate Liaison Team (CLT) and implement a user feedback loop by compiling user requirements for all GEO areas (e.g. define the most crucial climatological parameters and how they should be summarized) and providing scientific guidance on the best ways to disseminate and use the EURO4M products and services; WP4: PROJECT MANAGEMENT, COORDINATION AND SUSTAINABILITY coordinate the consortium and all EURO4M activities; manage day to day operation of the project; ensure that deliverables are completed on time; 5

6 3.2.2 Work progress and achievements during the period WP1: REGIONAL OBSERVATION DATASETS This work package has further developed and updated some of the well-known gridded datasets which are based on in situ observations only and include the CRU, GPCC and E-OBS datasets. Also, the work on satellite datasets for climate change monitoring has continued. Cooperation between the partners working on regional observation datasets has been particularly intensive for the activities related to recovery, rescue, imaging and digitisation of old data. This has lead to a coordinated approach to improving and temporally extending the ECV databases for the European region (linking with similar activities in MEDARE for the Mediterranean and ACRE worldwide). There have been no deviations in this work package from the description in the DoW (Annex I). The actual use of resources (about 160 pm) is close to what has been planned for these activities. KNMI has released Version 4 of the E-OBS gridded observations dataset on March 7, E- OBS is a daily gridded observational dataset for precipitation, temperature and sea level pressure in Europe based on ECA&D station observations (see The full dataset covers the period (Figure 1). It has originally been developed as part of the ENSEMBLES project (EU-FP6) and is now maintained and elaborated as part of the EURO4M project. This dataset covers the period over the entire European area, including the Mediterranean and the Middle-East. E-OBS consists of daily minimum, mean and maximum temperature and daily precipitation amount. New since version 4 is the addition of a daily mean sea level pressure grid (Figure 2). Besides the new Version 4, monthly updates have been issued for each E-OBS grid, and this will continue in during the lifetime of EURO4M. Figure 1: Example of an E-OBS daily field. The maximum temperature in Europe on 4 August 2003 is shown here. This is the day with the highest European averaged maximum temperature in the 0.25 degree regular grid version 2.0 of the E-OBS dataset. The European average maximum temperature was 28.2 C with a range of 7-42 C. 6

7 Figure 2: E-OBS daily mean sea level pressure grid for the wind storm event over Western Europe on 28 February The contour lines (also from E-OBS) are shown in steps of 5 hpa rounded off to 5 and 10 hpa. Within the framework of EURO4M, KNMI will deliver data sets of surface solar incoming radiation, precipitation and cloud properties retrieved from SEVIRI satellite data too. So far, KNMI extended the algorithm used to estimate surface solar incoming radiation by decomposing the radiation into its direct and diffuse component and by considering the effect of aerosol variations. These extensions are needed for making accurate estimates of solar panel yield from the data. KNMI also validated the estimates of surface solar incoming radiation by comparison with groundbased measurements (see Figure 3) and we validated satellite-retrieved precipitation by comparison with rain gauge data and rain radar estimates. Figure 3: Comparison of satellite-derived surface solar incoming radiation (vertical axis) with ground-based measurements (horizontal axis) for a cloud-free atmosphere. In the panel on the left-hand side hourly aerosol properties were used as input for the calculations whereas in the panel on the right-hand side monthly-averaged aerosol properties were used. Different colours pertain to different seasons. MO has contributed to WP1 with their ACRE work. The international Atmospheric Circulation Reconstructions over the Earth (ACRE) initiative ( is led by six core partners - the Queensland State Government in Australia; the UK Met Office (UKMO) Hadley 7

8 Centre; the US NOAA Earth System Research Laboratory (ESRL) and the Climate Diagnostics Center (CDC) of the Cooperative Institute for Research in Environmental Sciences (CIRES) at the University of Colorado at Boulder; NOAA s National Climatic Data Center (NCDC); and the universities of Giessen in Germany and Bern in Switzerland. ACRE is imaging, digitising and providing historical terrestrial station weather observations and some full data series for various locations in the Mediterranean region to WG 1 of EURO4M. Some of this is being directed through EURO4M s European partner at the University Rovira i Virgili, Spain, and the remainder is going to that European partner and ERA-CLIM via the University of Giessen in Germany, who are funding a large block of ACRE s digitisation activities covering Europe and the Mediterranean. The number and distribution of stations involved is available through ERA-CLIM s Spanish partner at the University Rovira i Virgili. These surface data are routinely added into the International Surface Pressure Data bank (ISPD) at NOAA s NCDC in the US, and are also available to the SurfaceTemperatures.org ( project data base. In addition, remote sensing data are being prepared by the MO for Climate Information Bulletins. URV, in charge of WP1.3 on Data Coordination, has made substantial progress during the first year in order to carry out the planned activities as follows: to define the network of Mediterranean stations and variables for which records are to be recovered and developed (quality controlled and homogenised, if required) to gather relevant data in images or paper format from different international and national sources to coordinate this effort with other worldwide projects and NMHSs efforts in order to avoid duplication to digitise the gathered data to contract the required human and technical resources to carry out these tasks (a post-doc and seven digitisers have been contracted). The progress made so far is in agreement with the planned activities and there has not been any deviation from the DoW. In the context of increasing historical climate data availability over Europe, URV has focused on the geographical regions of the southern and eastern Mediterranean, which are poorly documented although there is a rich heritage of climate data recorded over these Mediterranean sectors. A large number of data sources, both in digital and in paper format, have been examined from the MEDARE network, the organisations of NOAA-NCDC/CDMP, AEMet (Spanish Met Office), Météo-France, the Italian NMS, the Ebro Observatory Library, the ECA&D, CIRCE and Salvá-Sinobas/MILLENIUM projects and the ACRE and ISPD initiatives. The primary aim was to construct lists (for each ECV to be developed: temperature, precipitation and air pressure) with targeted meteorological stations. The ancient data to be recovered once merged with current data of these stations should enable the development of long-term climatic series at the daily- and hourly-scale on a near-evenly distributed network of sites across North Africa and Middle East. The digitisation process of the old parts of the targeted records started timely, as the first step heading to the development of long and high-quality climate records (e.g. quality control and homogenisation), in order to make them usable for developing advanced-level climate products and services. Stations from 46 sites were selected (locations shown in Figure 4) as a first option for digitisation and segments of non-digitised/digitised meteorological records were compiled in order to define periods that the digitisation should focus on. This ensures that the old parts of the records can be 8

9 combined with more recent parts of the records which are already available from the ECA&D project. In this way, long records will be developed and the risk of failure in the case that some NMHS are not keen in taking part in exchanging the data are minimized. A more ambitious secondguess list of stations was also compiled and the corresponding imaged data were gathered. This next step will enlarge the first selection (about 20 sites more). The digitisation started from the stations in the west end of North Africa and progressively moved eastwards. Already 22 sites for daily temperatures (maximum and minimum records), 15 daily precipitation and 15 hourly air pressure records over Morocco, Algeria, Tunisia and Libya have been digitized, while more eastern stations are currently being digitised. Near 40% of the targeted data for the first option list has been completed. In parallel, the quality control is in progress using internationally developed specialised software (the ETCCDI s RClimDex software) with additional home-expanded extra-qc which builds upon the former software in R. Data from new sources are also being sought to fill in intervals with missing records; this ongoing data-searching activity will run in parallel with the data digitisation/processing, as long as WP1.3 lasts. Figure 4: Locations of the 46 stations with records for digitization in the initial selection. In parallel, and under the umbrella of the WMO MEDARE Initiative, the Directors of NMHSs and Permanent Representatives with the WMO are being contacted in order to make possible the exchange of the recovered/digitised historical climate data by the current/recent parts of the targeted records. Although this task is in progress, there are good perspectives for being able to proceed with the suggested data exchange. NMA-RO has identified additional sources for data rescue activities, and has worked on gridding surface observations. Sources for data rescue activities NMA-RO identified sources for data rescue activities (listed in deliverable D.1.14). Archived climatological records at Romanian stations covering time intervals prior to 1960 could be used to extend the time series for their entire functioning period. Monthly bulletins issued by the Romanian Meteorological Service (spanning roughly the interval ) are sound sources for some snow data and catchment average precipitation covering the Romanian territory and some of its neighbouring countries. These precipitation data could be extended to the present and used as climatic indices relevant for potential users and stakeholders in WP3. The monthly values of snow cover depths and number of days with snow cover allow us to add extra data to the digitized dataset for the Former Soviet Union Hydrological Snow Surveys for the period (Krenke, 1998). Daily bulletins issued by the Romanian Meteorological Service (from 1925) are alternative sources 9

10 to fill some gaps of digitized data (pressure, temperature, precipitation, wind speed and wind direction) and improve their quality over regions from Eastern Europe, the Balkan and Northern and Eastern Mediterranean areas. In this context, we have already identified at least two stations (Tanger and Tripoli) for which air pressure data is digitized in selected years to fill the gaps from the other sources. Gridded surface observations NMA-RO has investigated the influences of grid resolutions on representation of surface observation patterns over the Romanian territory. NMA-RO applied the Cressman interpolation procedure to monthly temperature and precipitation at 103 stations covering the Romanian territory for the interval (e.g. figure 5). The Cressman scheme leads to extremes in the grid values that are not realistic if the selected resolution is too high using a small number of stations. For the Romanian territory, the available number of stations (103) has the best representation of the observed point data if the resolution is 0.5 º in latitude and longitude (Figure 5). Figure 5. Station (colored squares) and interpolated (contour lines) anomalies of temperature (in º C) at 103 Romanian stations for the winter 2010/2011 at different resolutions; the resolutions are 1º in latitude and longitude (a) and 0.5º in latitude and longitude (b). The reference period is

11 Cressman, G. P., 1959: An Operational Objective Analysis System,. Monthly Weather Review, Vol. 87, No. 10, pp Krenke, A. 1998, updated Former Soviet Union Hydrological Snow Surveys, Edited by NSIDC. Boulder, CO: National Snow and Ice Data Center/World Data Center for Glaciology. Digital media. ( MS has made good progress towards developing the gridded dataset of daily precipitation for the entire Alpine region ( ) by following two main lines of activities during the first year: a) Compilation of rain gauge data Institutions in seven Alpine countries (domain of interest: E, N) have been contacted to conclude agreements for obtaining access to daily rain-gauge observations from high-resolution operational networks. The requests focussed on updates to the existing Alpine Precipitation Dataset ( , see Frei and Schär, 1998). In some cases, however, a complete recollection ( ) is envisaged to ensure an up-to-date quality status of the data. The following list shows the current status of data collection for each of the seven national territories: Austria: The data has been successfully collected and stored in the central archive. Croatia: The request for data has been sent. Response of institution is pending. France: Meteo France formally contacted. Negotiations ongoing. Data delivery likely in spring/summer Germany: DWD formally contacted. Negotiations ongoing. Data delivery likely in spring/summer Italy: Responsibilities for operational hydrometeorology are fragmented. We have contacted all the regional agencies for the protection of the environment (ARPA) and additional institutions from each of the nine regions in the domain (Aosta, Piemonte, Lombardia, Liguria, Trentino, Alto Adige, Emilia-Romagna, Veneto and Friuli-Venezia-Giulia). The responses to our requests were heterogeneous. Promising negotiations are ongoing with the ARCIS project ( ARchivio Climatologico per l Italia Settentrionale or climatological archive of Northern Italy) for the exchange of a part of the available precipitation time series of the nine regions. It is likely that additional efforts will be needed to obtain access to networks of higher spatial density. Slovenia: The data has been collected. Integration into central archive to be completed. Switzerland: The data has been successfully collected and stored in the archive. b) Establishment of technical facilities A central data base has been developed for archiving the Alpine Precipitation Dataset. A flexible interface for accessing/deploying station data via the R software has been implemented. A complex R-tool has been written, which permits to join the new data series with already existing time series, to detect/treat missing values and to identify changes in station location. Existing gridding and display tools have been adapted and modified to work with the Alpine Precipitation Dataset and grids thereof. During the first year six employees (Christof Appenzeller, Mark Liniger Kerland, Christof Frei, Christian Lukasczyk, Gaudenz Flury and Marco Willi) were involved in tasks for EURO4M. In addition, MS has a full-time employee (Francesco Isotta) working for EURO4M since August 2010 and an additional person, working part time for EURO4M is being engaged in April MS has 11

12 used a total of person months of work for EURO4M. This is in line with the planned person months in WP1 and WP2 and the estimated efforts for D1.1. DWD has developed CDO scripts for the delivery of data in the European window of the GPCC dataset (D1.3). In addition, for EURO4M an integrated gridded dataset has been constructed based on HOAPS over ice-free oceans and GPCC over land areas (D1.8). The new dataset will offer monthly resolution in time and a spatial resolution of 0.5. It is intended to serve the needs of climate monitoring. The new dataset will cover at least the years 1988 to A preliminary version of the new dataset has been generated in this period. Also, an evaluation process of the new precipitation dataset has now been started. The data is compared to a variety of existing gridded datasets and is evaluated with station data. This process in ongoing. The data over land comes from the GPCC (Global Precipitation Climatology Centre, The GPCC is lead by the Deutscher Wetterdienst (DWD) and is part of GCOS (Global Climate Observation System) and WCRP (World Climate Research Programme). The product used for the new dataset is the Full Data Reanalysis product in Version 5 with a spatial resolution of 0.5. The GPCC Full Data Reanalysis (GPCC_FDR) is based on quality controlled rain-gauge data from over stations from about 190 countries worldwide. The HOAPS (Hamburg Ocean Atmosphere Parameters and Fluxes from Satellite Data, database consists of several water-cycle relevant parameters, including the used precipitation data over oceans. HOAPS precipitation is derived by SSM/I brightness temperatures and is available at 0.5 spatial resolution. The full time series of SSM/I brightness temperatures is now reprocessed and will be extended, probably to The new integrated GPCC-HOAPS precipitation dataset is promising, as GPCC and HOAPS fit quite reasonable at the coast, which is a challenging part in the integrated dataset. This preliminary EURO4M precipitation dataset (Figure 6) has been evaluated by comparison with other satellite based datasets (e.g. GPCP, TRMM) and to Reanalysis data (Figure 7). Over the ocean the uncertainties are highest. That is why the dataset is evaluated over oceans additionally with PACRAIN atoll station data (Figure 8). Figure 6: Multiyear mean of precipitation [mm/d], of preliminary EURO4M dataset 12

13 Figure 7: Zonal averages of multiyear precipitation datasets and Reanalysis (EURO4M, GPCP, ERA-Interim) Figure 8: Comparison of absolute biases of preliminary EURO4M precipitation, GPCP and ERA-Interim to PACRAIN atoll station data A preliminary version of the EURO4M global precipitation dataset has been presented with a poster at the EGU 2011 conference in Vienna in April At the EURO4M 2nd General Assembly in Bucharest in March 2011 the current status of the dataset generation has been presented with a talk. UEA has worked on developing gridded datasets based on long-term station series (WP1.1). Although the best data products make best use of in situ and remotely-sensed information, it is important to develop some products from one or the other type of information. Provided trends and spatial patterns of change agree, the scientific certainty gained from totally independent measurements is immense. UEA has used the latest version of the CRU monthly high-resolution (0.5 by 0.5 latitude/longitude) datasets (CRU TS 3, an update of CRU TS 2.1, Mitchell and Jones, 2005) to assess the quality and accuracy of many of the gridded dataset products (including those based on ERA-40) available for Europe. In the course of development of the CRU high-resolution datasets (which include the following variables: maximum and minimum temperature, precipitation, rainday counts, vapour pressure and cloudiness), UEA has gained considerable experience in the best practice of gridding and interpolation of station data. For example, it is rarely optimum to use the data expressed in the original measured units. Additionally, many users take CRU products to derive other fields, but the approaches employed are generally sub-optimal. The CRU TS 3 grids cover the whole globe but it is possible to isolate the subset(s) relevant to EURO4M. Software has been developed which allows the frequent (approaching real-time) re- 13

14 running of the gridding operation and thus the ability to keep the grids up-to-date. A new paper on the updated dataset will be submitted in the next couple of months, which has been partly supported by EURO4M. Here we give some background to the work and some comparisons with other datasets for the European domain covered by EURO4M. The full listing of variables (all at the monthly mean/total level) is: 1. Minimum air temperature (Tmin) 2. Maximum air temperature (Tmax) 3. Mean air temperature (Tmean) 4. Daily air temperature range (DTR) 5. Precipitation (Precip) 6. Vapour pressure (hpa,vap) 7. Wet day count (> 0.1mm, Wet) 8. Cloud cover (percentage, Cld) 9. Potential evapo-transpiration (Pet) 10. Frost days (ground frost, Frs) See Figures 9-13 for annual mean time-series, showing the areal average for all variables listed above. All series are expressed as anomalies from the average. For precipitation and wet days, the anomalies are expressed as percentages (i.e. as ratios as opposed to differences). Figure 9: Northern Europe and Mediterranean Basin minimum temperature ( C) anomalies (left) and maximum temperature ( C) anomalies (right) 14

15 Figure 10: Northern Europe and Mediterranean Basin mean temperature ( C) anomalies (left) and diurnal temperature range ( C) anomalies (right) Figure 11: Northern Europe and Mediterranean Basin precipitation percentage anomalies (left) and vapour pressure anomalies (hpa)- (right) Figure 12: Northern Europe and Mediterranean Basin percentage wet days anomalies (left) and cloud anomalies (right) Figure 13: Northern Europe and Mediterranean Basin PET anomalies (left) and frost-day anomalies (right) In addition, some comparisons are shown between the CRU TS 3 output and output from alternative products (Figures 14 and 15). Two regions have been used for the time-series illustrations. This 15

16 effectively splits the EURO4M region into two basic climatic zones: Northern Europe (region) N, 10 W-40 E and Mediterranean basin (region) N, 10 W-40 E. The alternative gridded products used in the comparison exercises are: CRUTEM3 5 x5 latitude/longitude grids of monthly-mean temperature produced by the Climatic Research Unit (Brohan et al., 2006) GPCC -0.5 x 0.5 latitude/longitude grids of monthly-total precipitation produced by DWD (Legates and Willmott, 1990) UDEL x 0.5 latitude/longitude grids of monthly-mean temperature (Schneider et al., 2010) Figure 14: Regional comparison between CRU TS 3.00 and UDEL equivalents 16

17 Figure 15: Regional comparison between CRU TS 3.00 and GPCC equivalents Discussion of the regional time-series and the comparisons with other products: Regional temperatures show an upward trend in recent decades which is in accord with the trends seen in the larger scale illustrations using alternative gridded products. These trends are also reflected in the regional (falling) trends in frost-day anomalies. The upward trend in precipitation seen in the northern European annual time-series is not evident in the Mediterranean Basin time-series, where a slight downward trend is apparent. This situation is also reflected in the wet-day count anomalies with increasing trends in northern Europe and falling trends in the Mediterranean basin. However, from the regional time-series for vapour pressure, rising trends in recent decades are common to both areas resulting from higher temperatures in recent decades. Cloud anomalies show falling trends in both regions during recent decades, with the exception of a period during the late 1990s and just beyond the turn of the millennium when higher cloud amounts were present. Of note is the appearance of the Mediterranean basin cloud anomalies before ca The flat appearance of the regional anomaly time-series is due to a general lack of cloud-amount data in this geographical region before The interpolation of all variables relaxes to zero anomaly values (i.e average values) when station data are insufficient. This phenomenon, due to a lack of station data, shows the potential benefits from (data rescue efforts) in other parts of EURO4M WP1. The rising trends in potential evapotranspiration rates, seen in both regions, reflects the generally higher growing season temperatures. The comparisons with alternative products, at both regional and global/hemispheric scales and for both temperature and precipitation show generally very close agreement. Brohan, P., Kennedy, J., Harris, I., Tett, S.F.B. and Jones, P.D., 2006: Uncertainty estimates in regional and global observed temperature changes: a new dataset from J. Geophys. Res. 111, D12106, doi: /2005jd

18 Legates, D. R. and C. J. Willmott, 1990: Mean Seasonal and Spatial Variability Global Surface Air Temperature. Theoretical and Applied Climatology, 41, Mitchell, T.D. and Jones, P.D., 2005: An improved method of constructing a database of monthly climate observations and associated high-resolution grids. Int. J. Climatol. 25, Schneider, U., A. Becker, A. Meyer-Christoffer, M. Ziese and B. Rudolf, 2010: Global Precipitation Analysis Products of the GPCC. Global Precipitation Climatology Centre (GPCC), DWD, Internet Publikation, (pdf 1414 KB ). MF has no activities in this WP for this period. 18

19 WP2: REGIONAL REANALYSIS This work package has further developed the model-based European regional reanalysis capabilities and the 2D downscaling capabilities. MO has worked with ECMWF to build on their extensive infrastructure and experience resulting from ECMWF's earlier reanalysis activities. SMHI and MF have coordinated closely their comparison and evaluation studies in which they use the existing ERAMESAN and SAFRAN methods for downscaling the HIRLAM-based regional reanalysis results to the local scale. There have been no deviations in this work package from the description in the DoW (Annex I). The actual use of resources (about 80 pm) is close to what has been planned for these activities. MO has made significant progress in year 1 to develop a model-based European regional reanalysis capability for the EURO4M project: Setting up the detailed configuration of the new EURO4M domain. Technical developments to port required components of the Met Office model to ECMWF in order to interface with the global ERA model and its observation datasets; Scientific developments to further improve the use of observations for reanalysis applications. The EURO4M project is the first application of Met Office data assimilation capabilities at ECMWF, and so there is a significant technical aspect to the effort in the first two years of the project. An internal milestone is to implement an initial ERA/EURO4M global/regional reanalysis configuration by March 2012, to permit further testing/tuning prior to formal EURO4M deliverables in years 3 and 4. The planned system is shown in Figure 16. At end of year 1, most of the Met Office components have been ported. Testing of the interfaces has begun, including a significant collaboration with ECMWF to make further use of ECMWF s Observation Database (ODB) software within the Met Office system. Figure 16: Nested ERA/EURO4M global/regional reanalysis system. ERA components are shown in blue, Met Office EURO4M components in green, and the required interfaces in grey. The Recon and MakeBC algorithms convert ECMWF gridded data for use as initial/lateral boundary conditions for the Unified Model (UM). The Observation Preprocessing System (OPS) performs quality control, thinning etc of ECMWF BUFR observations in preparation for assimilation in the unified 3/4D-Var variational data assimilation (VAR). 19

20 In parallel with the technical developments for EURO4M, scientific development and initial testing of EURO4M capabilities has been performed on the Met Office supercomputer. Work completed in year 1 includes: producing ancillary files for new domain, comparing Met Office regional NWP and climate configurations (decision to use NWP configuration for now, as differences as small and diminishing as we move to a seamless configuration), and tuning the QC of observations in the new domain. In addition, a preliminary cycling test of the EURO4M configuration has been performed for the period 00Z 16 Feb to 00Z 8 March 2011 using Met Office observations and lateral boundary conditions. Results indicate the system is performed satisfactorily, but has also highlighted a need to further tune the system for reanalysis use. As an example, Figure 17 shows the mean surface temperature analysis increment (analysis minus previous six-hour forecast) for this test period. Investigations show the large positive increment at 12Z over North-East Europe correspond with model snow cover over land, indicating soil temperatures are too cold under the snow. Soil temperatures are usually updated with increments from the screen-level temperatures in the data assimilation. This is deliberately not done in areas of snow, where it is assumed soil temperature and screen temperature are decoupled. In the case of thin snow, cold soil temperatures will prevent screen temperature rising sufficiently during the day, leading to a large positive mean increments at 12Z. Further work is required to eliminate this model limitation. Figure 17: Mean screen level temperature for the 4 synoptic times within the six-hourly cycling EURO4M test period of 00Z 16 Feb to 00Z 8 March

21 MS activities in this WP are formally starting in month 24. Early technical work has been done to setup an archiving structure for evaluation datasets and a corresponding retrieval interface in R. DWD has no activities in this WP for this period. SMHI has been involved in three workpackages of WP2: Dynamical downscaling of ERA (3D-VAR downscaling) (WP2.2) The HIRLAM 3D-VAR data assimilation and modelling system has been set up over a large area encompassing the EEA countries (Europe, Mediterranean and East and North Atlantic) (see Figure 18). Several resolutions and options have been tried, including HIRLAM 4D-VAR and HARMONIE (ALADIN) 3D-VAR. The resolution was chosen to be 22 km and 60 levels up to 10 hpa. The lateral boundaries are provided by the ECMWF ERA-Interim re-analysis. Furthermore, since only conventional observations (and not satellite data) are used, the coupling with ERA- Interim is important. This is achieved by activating the newly developed Jk cost function describing the analysis differences between HIRLAM and ECMWF. It provides large scale information at analysis time from ERA-Interim. Figure 18: HIRLAM 3D-VAR model domain showing the adjusted albedo field for January Numerous test assimilations have been made for parts of 1989 and 2009 and compared with ERA- Interim and other data sources. Assimilation and forecast statistics have been computed and evaluated. Some deficiencies regarding soil type over Africa and snow density over Greenland as well as the climatological sea ice limit were found and they have been addressed. 21

22 Already during the test months run for 2009, one can see that the HIRLAM re-analysis performs well and adds detail that are not present in the coarser resolution ERA-Interim (Figure 19). Figure 19: SMHI manual accumulated precipitation analysis for October 2009 (left) and HIRLAM 22 km reanalysis (middle) compared with ERA-Interim 78 km reanalysis (right). The most recent HIRLAM Reference version (7.3) contains a new snow/soil parameterisation and gravity wave/surface roughness scheme and it has been validated extensively. In spite of a remaining pressure bias in the ensuing forecasts, there are many advantages of using this version and the full year of 2009 has started to be assimilated. The output field requirements have been checked and 3D fields of radiation and precipitation fluxes have been added. 2D-mesoscale (MESAN and SAFRAN) downscaling (WP2.3) SMHI together with MF have set up MESAN at 5.5 km over France and climate information (fixed fields needed interpolations, downscaling and the analysis) has been derived for the entire EURO4M area. Experiments have been made with MESAN over France and surrounding areas with a comprehensive set of climatological observations from Météo-France for the purpose of the SAFRAN inter comparison. Different downscaling strategies from the first guess 22 km HIRLAM model have been tried and further developed. It is particularly for T2m where the orography differences are important. For 24hour accumulated precipitation (RR24) there is a regression based on physiographic fields that determines the standard deviation of background error in the analysis, as illustrated in the left panel of Figure 20. A further step is to then also analyse the actual standard deviations of the observations and this gives a very good representation of the local conditions (right panel in Figure 20). 22

23 Figure 20: Regression of RR24 based on physiographic fields (left) and analysis of STD(RR24) using the regression as first guess (right). Comparison of MESAN and SAFRAN (Evaluation) (WP2.4). MESAN has been employed together with SAFRAN and CANARI to re-analyse 3 months using the comprehensive observation set over France. There have been extensive validations and investigations resulting in some adjustments and corrections. The variables T2m, RH2m and RR24 have been inter compared and these are reported in EURO4M Project Report D2.10: Comparison of existing ERAMESAN with SAFRAN downscaling (Deliverable No ). The three different 2D-analysis systems have been compared both against the observations used and also against a somewhat smaller independent set of observations, The first guesses that come from HIRLAM and MF ARPEGE models, respectively, and their downscaling to the analysis grid, have been verified and compared. Then final analyses have been verified against the observation sets. Even if the analysis systems are quite different, they perform well in terms of drawing to the observations. The main differences and the area which has required most work is the analysis of T2m and its interpretation. Vertical downscaling and particularly vertical corrections of the analysis when verifying against observations show differences between MESAN and the two MF systems. MESAN draws closer to the observations than SAFRAN and CANARI do but after vertical correction, the MESAN analysis is somewhat deteriorated (for the winter months) (see Figure 21). It seems to be mainly for high elevations and may be a result of drawing to close to observations, which in turn is dependent on the specifications of the assumed first guess errors. For RH2m and RR24 all the three systems performed very well and the differences were small. A lot of insight has been gained from the extensive validation and inter comparison. The performance (observation fit) varies episodically as well as with time of the day. The vertical correction in the verification is important and the application of a standard temperature gradient is unrealistic most of the time. 23

24 Figure 21: Time evolution of the daily mean rmse of analyses of T2m for December 2009, not corrected (a), and (b) corrected with the standard vertical temperature gradient, respectively. Verification with respect to about 1290 observations.the curves show MESAN with just horizontal downscaling of first guess (1a) and also vertical (1c) and SAFRAN and CANARI. All at 5.5 km resolution. During Year 1 SMHI has used somewhat more resources than expected (32 % of the total for the Project at SMHI). This is due to rather more work than expected to rectify deficiencies found in the HIRLAM re-analysis system initially and because the comparison between MESAN and SAFRAN was extended beyond what was initially planned. It is important to find weaknesses before the full re-analysis is started. Originally it was planned to compare SAFRAN with existing data from an older ERA-MESAN reanalysis but it was decided to instead generate new data with a MESAN setup that is close to what is indented for the reanalysis in EURO4M. Important findings came out from the MESAN SAFRAN intercomparison that will improve the final MESAN results. UEA has no activities in this WP for this period MF has worked closely with SMHI on 2D-mesoscale (MESAN and SAFRAN) downscaling (WP2.3) and comparison of MESAN and SAFRAN (Evaluation) (WP2.4). See their joint report provided above under SMHI. 24

25 WP3: USER-ORIENTED INFORMATION AND CLIMATE CHANGE PRODUCTS This work package has started the work towards developing the main user-oriented information products. In a joint effort of DWD, NMA-RO and KNMI the so-called Climate Liaison Team (CLT) has been set up. This team will implement a user feedback loop by compiling user requirements for all GEO areas and providing scientific guidance on the best ways to disseminate and use the EURO4M products and services. The CLT will further evolve when user feedback is provided to shape the first Climate Indicator Bulletin (CIB). This bulletin will contain user-oriented information and multi-purpose climate change products related to European temperature. This bulletin is due for Month 24 and will serve the full range of climate users and applications sectors in Europe within the wider global community. There have been no deviations in this work package from the description in the DoW (Annex I). The actual use of resources (about 50 pm) is close to what has been planned for these activities. KNMI has written a User Requirements Document (URD) for the Climate Indicator Bulletins (CIB) environment. This document contains use cases which describe how users interact with the CIB environment. MediaWiki has been investigated to see whether it can be used to create and host the Climate Indicator Bulletins. MediaWiki seems to be a good candidate. A password protected MediaWiki has been installed on the EURO4M website (password protection will be removed when operational). A MediaWiki extension for the KNMI Web Mapping Service (ADAGUC tools) has been made in order to make Open Geospatial Concortium (OGC) WMS available within the MediaWiki content. Python visualization scripts have been created to make graphs of CIB data. These Python visualization scripts are configurable in the MediaWiki editor. The results of this work have been presented at the EURO4M General Assembly Romania and at the EGU in Vienna. Also, KNMI has coordinated the development of the content for the first CIB, which will be issued in Month 24. The first bulletin will focus on surface temperature products, because this is the highest priority ECV (as defined by GCOS) and there is a high user demand for information about the temperature evolution in Europe (in the context of global warming). MO has no activities in this WP for this period. NMA-RO has worked in preparation for future Climate Indicator Bulletins (CIBs). More userdriven indices need to be identified using a regionally-oriented approach. First attempts have been made in this regard with indices such as potential evapotranspiration in the Danube Delta and precipitation averaged over the Danube basin. Feedbacks on user requests and needs could be received from the Climate Liaison Team (CLT). Precipitation over the Danube basin The Danube basin is important in Europe from the standpoint of historical, economical, political, cultural and environmental criteria, and its climate-related issues are of interest for a wide community of users. The climatic relevance of the Danube basin to European climate is due mainly to its Mediterranean links. The Danube runoff gives more than twice the Nile s contribution as freshwater flux into the Mediterranean Sea via the Black Sea (Lucarini et al., 2007). On the other hand, Mediterranean cyclones provide a large part of precipitation over the Danube basin. Also, the storms of Atlantic origin reinforced over Mediterranean water in the presence of specific orographic forcing give a larger European context to the floods occurring in the regions from or nearby the Danube basin (Lucarini et al., 2007). Potentially relevant climate indices for the Danube basin include catchment s averages of precipitation. There are homogeneity problems with this quantity 25