24-26 November 2003
Zuckerman Institute / Climatic Research Unit, East Anglia University, Norwich, UK
1. Circulate to members and inform CLIVAR SSG of the ET interpretation of the terms of reference that resulted from discussion during the team meeting and is summarized in the meeting report. Zwiers/Yan
2. Prepare a template letter to WMO Permanent Representatives in relevant countries, which states objectives and benefits of workshop and type of participant desired. Mokssit/Zwiers/Alexander/Haylock/Hewitson(Llanso/Detemmerman to resurrect previously used letter by early January 2004)
3. Develop a template screening questionnaire for choosing workshop participants, based on previously used. Mokssit/Zwiers/Hewitson
4. Check all aspects of the ClimDex software that will be used in workshops. Klein-Tank/Alexander
5. Prepare a Resource CD containing ClimDex documentation and other supporting materials for workshops. Alexander/New/Stephenson
6. Arrange for translation of the software document into other languages. Stephenson for Portuguese /Llanso for French and Spanish (Carriers of Actions 4&5 to make sure that the document is ready by 1 March 2004)
7. Identify targeted regions not covered by pervious workshops. Mokssit/Alexander/Haylock
8. Organize and deliver workshops in coordination with supporting organizations and funders (see Table 1). Local organizers / ET contacts
9. Prepare a short document on preferred practices for index construction and analysis, including guidance for 'reanalysis' of previously collected indices. Include a cross-referenced list of indices computed by the various pieces of software that are currently available so as to identify differences in index definitions. Also include advice on analysis methods appropriate to each index. Publish as a CCl/CLIVAR report and post on the web. Stephenson/Klein-Tank/Haylock/Zwiers/Zhang
10. Make sure that workshop information (including outputs and follow-up activities) is available at the ET website and linked to the CLIVAR website. Zhang/Zwiers/Yan
11. Prepare a short document on suggested large-scale atmospheric, oceanic and marine climate indices for monitoring and detection that should be considered for the IPCC 4th Assessment Report (AR4). Bindoff/Folland/Swail/Karoly
12. Write to JCOMM expressing need for marine climate change detection, monitoring and indices. Co-chairs/Swail
13. Develop an ET website focused on local indices. If feasible, include with group email/discussion capability and data-visualizing tools. Zhang/Zwiers
14. Prepare a timetable for ETCCDMI activities, taking the IPCC AR4 and CLIVAR and relevant WMO activities into account. Co-chairs/Yan
Table 1. Summary of ET regional workshop planning
Asia Minor /Caucasus
Africa with priority for regions not covered by Casablanca Workshop
Most of S American
Central and northern S America
W Asian and some SE European countries
31 May - 4 June 2004
2 weeks in May 2005
University of Cape Town and South Africa Weather Service, Cape Town, South Africa
University Federal de Alagoas, Maceio-Alagoas, Brazil (offers also from CPTEC and IAI)
WMO Beijing Climate Center, Beijing, China
Meteorological Guest House Alanya, Antalya, Turkey
/Instructors (more to be considered)
New Stephenson Collins
9th Int. Mtg. Stat. Climat. 24-28 May 2004
To be considered
To be considered
Length: a week with/plus two days for lectures
Language: English desirable for participants
Technical needs: computer facilities with Windows95 up and better one person capable of Visual Basic
Data needs: GSN sites and reference stations with metadata (less than 10% missing records) from 1950 (or at least 1961) onwards, in format suitable for input into ClimDex software
Variables: including daily Tmean, Tmax, Tmin and precipitation
· Inventory of daily data
· Improved access to daily data
· Greater understanding of data-sparse regions problems
· Contribution of results to a central (ET) website within 2 months
· Feedback on software and documentation
· Feedback on indices
· Peer reviewed publication
· CD of outputs
· Inventory of daily data
· Quality-controlled daily data
· Skills in climate data analysis
· Appreciation of wider picture (global change issues in the region) and of value of sharing data
· National report on climate change as indicated by indices (in regional context)
· Increased capacity to institute regular climate change reporting
· Ongoing regional network
· Peer reviewed publication for input to IPCC
· CD of outputs
The Expert Team for Climate Change Detection Monitoring and Indices (ETCCDMI) is jointly sponsored by the Commission of Climatology (CCl) of the World Meteorological Organization (WMO) and the Climate Variability and Predictability (CLIVAR) project of the World Climate Research Programme (WCRP). It plays an important role in developing and implementing CLIVAR's research programme for climate change detection and attribution and in coordinating and providing guidance for CCl's global climate monitoring activities. The ETCCDMI is the successor of the CCl/CLIVAR Working Group on Climate Change Detection (with a history archived at http://www.clivar.org/publications/wg_reports/wgccd/wgccd_report_3.pdf). The terms of reference for the new team (http://www.clivar.org/organization/etccd/index.htm) had been discussed by the team by email and its interpretation is summarized later in this report (section 2.3).
The current team members are:
Dr. A. Mokssit (Co-chair), Meteo Moroc, Casablanca, Morocco
Dr. F. Zwiers (Co-chair), Canadian Centre for Climate, Victoria, Canada
Dr. N. Bindoff, University of Tasmania, Hobart, Australia
Prof. C. Folland, Met Office, Exeter, UK
Prof. P. Jones, University of East Anglia, Norwich, UK
Prof. D. Karoly, University of Oklahoma, Norman, USA
Mr. L. Molion, Universidade Federal de Alagoas, Alagoas, Brazil
Mr. S. Sensoy, Turkish State Met Service, Kalaba / Ankara, Turkey
Dr. D. Stephenson, University of Reading, Reading, UK, is a Special Advisor to the team. Dr. X. Zhang, Climate Monitoring and Data Interpretation Division, Meteorological Service of Canada, coordinates the development of the ET/CCDMI website. The International CLIVAR Project Office contact for the team is Dr. Z. Yan.
The objectives of the first team meeting were
The first CCl/CLIVAR ETCCDMI meeting was held at the new Zuckerman Institute, University of East Anglia, Norwich, UK. Dr. F. Zwiers (Co-chair) opened the 2.5-day meeting on the morning of 24 November 2003. He thanked Prof. P. Jones, Ms. J Burgess and Dr. Z. Yan for their efforts in organizing the meeting and CCl and CLIVAR for their sponsorship. Prof. Jones, the director of the host institute (Climatic Research Unit), welcomed the team and experts (Appendix 1) to Norwich and expressed appreciation of the team meeting at East Anglia as a first international venture in the new institute.
The Co-Chairs introduced the meeting agenda (Appendix 2) and objectives. They expected to make a 2-3-year work plan for the team, as annual team meetings are unlikely. Dr. Zwiers expected that the team would contribute to CLIVAR and IPCC by improving global coverage for climate change indices and analysis tools and providing global assessment to indices. The team could also contribute to CCl's implementation programme by developing an official toolkit (software, documentation and supporting materials) to be promoted for application in national operational agencies, reviewing the Annual WMO Climate Assessment and by the capacity building in the detection and monitoring area through regional workshops.
Dr. Mokssit stressed that the team should work on both scientific (CLIVAR) and operational (CCl) aspects. It was desirable that the teamwork would lead to more national operational agencies being involved in the process of capacity building in the CCDMI area. For workshops, Dr. Mokssit stressed the importance of workshop follow-up activities in order to assure sustainability of the teamwork.
The participants discussed the meeting agenda and agreed with the deliverables and expectations introduced by the co-chairs. Dr. M. Haylock reminded the team of possible links to the START / Asian-Pacific Network (APN), which also would organize similar regional workshops. It was noted that CLIVAR SSG (12th session, Victoria, May 2003) encouraged the team to work with START and GCOS when planning regional workshops. The team agreed to keep up information and communication with relevant organizations through both the ICPO and the team members.
2.1. The CCl perspective
Mr. P. Llanso, Chief of the World Climate Data and Monitoring Programme (WCDMP) under the World Climate Programme (WCP), outlined relevant aspects of the WCP, WCDMP and CCl. He pointed out some WCP areas of interest, including
As Mr. Llanso outlined, the WCDMP was aimed at
The team is well-positioned to contribute to these objectives.
With support from WCDMP, CCl has established 3 Open Programme Area Groups (OPAGs), each with an Implementation/Coordination Team (ICT), and an Expert Team (ET) and Rapporteurs. The 3 OPAGs are (1) Climate Data and Data Management, (2) Monitoring and Analysis of Climate Variability and Change, and (3) Climate Applications, Information and Prediction Services. As a part of the effort to fulfill the CCl's tasks, a CCl/CLIVAR Working Group for Climate Change Detection (WGCCD) was established in 1999, which organized two regional workshops, one in Kingston (January 2001) for the Caribbean area and another in Casablanca (February 2001) for Africa. To progress further, CCl established the ETCCDMI under OPAG2, in coordination with CLIVAR during late 2002.
Mr. Llanso summarized the CCl expectation for the ETCCDMI. These include activities to
During discussion, Dr. N. Bindoff questioned the relationship between the Expert Team (ET) and the Implementation/Coordination Team (ICT). Logically, it was anticipated that the ET provide scientific guidance and develop software while the ICT would carry out applications and capacity building in WMO member countries. Mr. V. Swail commented that CCl might need to pay more attention to marine and oceanic aspects of climate.
2.2. The CLIVAR perspective
Dr. Yan briefed the team on relevant aspects of WCRP and CLIVAR. He forwarded an apology from Dr. H Cattle, the director of the International CLIVAR Project Office, who co-authored the presentation but was unable to attend the meeting. As the presentation showed, the Joint Scientific Committee (JSC) reaffirmed during its recent annual meeting that, after discussions initiated by the WCRP Banner Project, the aims of WCRP remained as originally specified, i.e.,
To help achieve this overall objective, CLIVAR has been set up with 4 specific objectives:
Dr. Yan anticipated that the ETCCDMI's activities would contribute directly to the 2nd and 4th specific objectives and help in some aspects for achieving the other two. He suggested the team consider possible links and coordination with other CLIVAR activities, especially those of the CLIVAR Data and Information System, the CLIVAR/PAGES Working Group, and the Working Group for Coupled Modeling. Links to developing CLIVAR basin panels are also encouraged, as part of the team's effort to promote the development of marine and oceanic aspects of CCDMI.
The team was also briefed on some outcomes from the 12th CLIVAR SSG session (Victoria, May 2003). The SSG agreed to suggested changes in TOR resulting from discussions among team members early this year and encouraged the team to work with START and GCOS when planning regional workshops. The ET Terms of Reference are discussed further below.
The team was encouraged to use the CLIVAR Exchanges newsletter for publishing outcomes of the regional workshops as appropriate and to take part in the first International CLIVAR Science Conference, 21-25 June 2004, Baltimore, USA.
During discussion, it was stressed that the anthropogenic climate change (ACC) is a cross-cutting theme of the CLIVAR programme. However, as Prof. D. Karoly commented, CLIVAR might not have made sufficient effort in that direction. The ETCCDMI should be able to make substantial contributions in this area. Dr. Bindoff supported links between ETCCDMI and CLIVAR basin panels, in order to promote marine and oceanic aspects of CCDMI across the CLIVAR community.
2.3. The ET interpretation of the terms of reference
With respect to the CCl and CLIVAR perspectives, Drs. Zwiers and Mokssit guided the discussion on ET's terms of reference (TOR). A 10-item TOR was reported by Dr. Mokssit at CLIVAR SSG-11 (Xian, May 2002). The SSG recommended there could be fewer and simpler items, while keeping details in working plan. This led to discussions among team members and a shorter version resulted and presented to the CLIVAR SSG-12 earlier this year. The team reviewed each of the items. Table 2 summarizes the team's interpretation.
Table 2. The ET interpretation of the TOR
Interpretation (with item number in the shorter version)
1. To further develop and publicize indices and indicators of climate change and variability, with particular emphasis on the creation of indices of daily to seasonal extremes covering the global land surface using standardized software packages
2. To further develop other indices of value to the IPCC, related to changes in mean climate and its variability from the sub-surface of the oceans to the stratosphere
3. To provide input on indices to WMO publications such as the Annual Statement on the Status of the Global Climate
4. To compare modeled and observed indices, and report on the comparisons, with some emphasis on changing extremes
5. To assist in the specification and implementation of observing system experiments with models used for global and regional climate change detection, with emphasis on the GUAN and GSN networks
The team will work in this area as appropriate under item 8
6. To arrange for or make assessments that identify and quantify the magnitude of biases introduced by automated means of measurements and their consequences for detection and attribution
The homogeneity issues are recognized as being important and within the ET purview
7. To consider other issues of homogeneity as deemed appropriate
Within the ET purview
8. To collaborate with and provide inputs to other groups, especially those set up under IPCC auspices, regarding the adequacy of the global observing system for the purposes of supplying advice to Conferences of the Parties to the Rio Greenhouse Gas Convention, and regarding the development of indices
To collaborate with and provide inputs to other groups, especially those set up under IPCC auspices, regarding the adequacy of the global observing system and the development of indices (5)
9. To maintain plans for capacity building in developing countries in the above activities, particularly through Workshops. In particular, to work closely with START on capacity building through its Monitoring Extreme Climate Events (START-MECE) group
The team will support capacity building in the above activities, particularly through workshops and collaboration with START
10. To submit reports in accordance with timetables established by the COPAG and/or Management Board, and agreed with CLIVAR participants
Within the ET purview
During discussion, Prof. C. Folland and Mr. L Molion noted that it could be worthwhile explicitly stating the homogeneity issues in the working plan for ETCCDMI activities, especially for the regional workshops, where homogeneity could be overlooked. The team agreed to circulate its interpretation of the TOR to relevant parties.
Action 1. Circulate to members and inform CLIVAR SSG of the ET interpretation of the terms of reference that resulted from discussion during the team meeting and is summarized in the meeting report. Zwiers/Yan
A number of international programmes/organizations are engaged in the field of CCDMI. The team was briefed on some of these programs so that it would have the information necessary to ensur that its activities well coordinated with the existing activities.
GCOS - In his presentation of the Global Climate Observation System, Prof. Folland pointed out that indices could be a key issue for climate change detection and monitoring. He showed a few aspects to be considered in order to guide GCOS/ETCCDMI's work. The first is to identify questions indices should answer. Example questions are:
Based on data availability, Prof. Folland discussed examples of indices related to these questions, including the global mean temperature, annual surface temperature anomalies and percentiles, upper-300m ocean temperature anomalies, seasonal anomalies of global temperature for the lower troposphere, global tropopause level, OLR, sea ice extent, land snow cover, monthly precipitation in typical regions such as the Asian monsoon region, and circulation indices for the North Atlantic Oscillation, Southern Oscillation and other modes of variability.
Prof. Folland called on the team to consider who its audiences are, what space and time scales are important, who else has developed climate indices, and how the GCOS/ETCCDMI group should choose from the vast range of possible indices. He wished for current index states to be put in historical context and include uncertainties where possible. He concluded with a recommendation to set up a dedicated web site hosting a new GCOS/ETCCD diagnostics/index monitoring system, which should complement and be linked to other sites. He noted this would need considerable expertise and time. He mentioned possible use of the ERA-40 reanalysis data set, which has recently been made available (free online).
JCOMM - Mr. V. Swail thanked Peter Dexter (JCOMM Secretariat, Geneva) for his input to his presentation. The WMO/IOC Joint Technical Commission for Oceanography and Marine Meteorology is an intergovernmental body of experts, which provides the international, intergovernmental coordination, regulation and management mechanism for an operational oceanographic and marine meteorological observing, data management and services system. As Mr. V. Swail showed, JCOMM makes effort to further develop the observing network, partly under the guidance of GCOS. Its data management activities are coordinated with other bodies including GOOS and WMO/CCl. JCOMM has archived various types of data for meteorology, surface oceanography, sea level and sea ice, including those from vo nteer vessels, drifters, buoys, and satellites, which are useful for studying seasonal to interannual and longer-term climate variability.
As Mr. Swail discussed, a notable JCOMM activity is the MCSS - Marine Climatological Summaries Scheme. Established in 1964, the MCSS has as its primary objective the international exchange, quality control and archival of delayed mode marine climatological data, in support of global climate studies and the provision of a range of marine climatological services. Two Global Data Collecting Centres (GCC) were established in 1993 in Germany and the United Kingdom. All data are eventually archived in the appropriate World Data Centres, such as the National Climatic Data Center (NCDC). Another notable advance took place with the CLIMAR workshops. As Mr. Swail outlined, the 1st CLIMAR workshop (Vancouver, 1999) brought together COADS and JCOMM, and was followed by workshop in Boulder (2002) and CLIMAR II (Brussels, 2003), which further developed CLIMAR's scope. CLIMAR set 3 streams of work on:
A complete description of JCOMM's data activities can be found on the JCOMM website at: http://www.jcommops.org.
START - The IGBP/IHDP/WCRP co-sponsored SysTem for Analysis Research and Training plays an important role in capacity building for global change studies. START was involved in earlier CCl/CLIVAR regional workshops and is continuing its involvement by supporting the up-coming South African workshop. Although a START representative could not attend the meeting, the team highly appreciated START's contribution in past activities and would keep close collaboration with START in the future.
IPCC - Working Group I of the Inter-governmental Panel for Climate Change undertakes a periodic scientific assessment of climate change research and publishes IPCC Assessment Report (AR) every 4-5-6 years. Dr. Zwiers outlined plans for IPCC AR4 (co-chaired by Dahe Qin and Susan Solomon), to which the team will make considerable contribution. Specifically, the outputs of the regional workshops (details in section 5) will fill geographical gaps in observations of climate change for the AR4's Chapter 3 (Atmospheric and Surface Observations). The team may also be able to contribute to Chapters 4 and 5 (Cryosphere and Ocean). Dr. Zwiers anticipated potential impacts on other chapters, given the role of indices in:
Dr. Zwiers showed the AR4 time line and called on the team to plan its activities accordingly. A key time point is December 2005, when all material referenced should be in press or published. This should be considered when planning regional workshops. Dr. Mokssit added that the previous Caribbean and African workshops could also contribute to help fill the gaps in the map of climate change assessment based on CCI.
The team agreed in principle to adjust the working plan to the IPCC time line. However, as Dr. Stephenson argued and suggested, the team should also develop a wider scope. Nonetheless, it was recognized that in practical terms, the IPCC activities could serve as a driving force for the team activities. Thus the IPCC Third Assessment Report became a clear focus for the 1997-2001 CLIVAR/CCl Working Group on Climate Change Detection that preceded the current Expert Team and considerably helped the delivery of results.
There have been a number of European projects, which lead in the field of climate change, extremes and indices. The team was briefed on some of the scientific advances, in order to identify priorities for future works.
MICE - Modeling the Impact of Climate Extremes. Prof. J. Palutikof introduced this project, as a part of a cluster of 3 projects dealing with extremes within European Union Framework 5, running from February 2002 until July 2005. The MICE project aims to
As Prof. Palutikof explained, MICE does not carry out downscaling analysis, as many model-observation comparative studies do. It considers more weather types than single-station or -grid statistics. Prof. Palutikof showed some examples linking extremes and impacts, including energy consumption and crop production. For impact studies, the MICE experience suggests that down-scaling is not necessarily a better way. More details are online at http://www.cru.uea.ac.uk/cru/projects/mice/.
STARDEX - STatistical and Regional dynamical Downscaling of Extremes for European regions, is another of the three EU projects noted above. STARDEX is coordinated by the University of East Anglia and includes 12 European partners. As Dr. Haylock introduced, the objectives of STARDEX are to improve downscaling techniques of extremes and apply more robust techniques to providing projections of changes to climate extremes under climate change. STARDEX has developed a software tool that calculates 57 climate indices, mostly dealing with extremes. A core set of 10 indices is being used throughout the project to develop scenarios for Europe. There are differences in the definitions of some indices as compared with the methods of calculation used in other European studies. The question of what constitutes a climate 'extreme' has been an important point of discussion in STARDEX, but the need for statistical confidence in downscaling extremes means that only more moderate extremes (return periods of less than a year) can be currently considered. More details are available at http://www.cru.uea.ac.uk/cru/projects/stardex.
Questions were raised about the 57 indices and how to choose indices. Prof. Jones and Dr. Haylock explained that the project's top 10 indices were based on considerations of both practical and statistical significance and of modeler's interests. The team recognized the need for identifying different types of extreme indices for different climate regimes. In later discussions on workshops, Dr. Haylock agreed to collaborate with Dr. A. Klein-Tank and some others to make a table comparing all indices available and commonly used in international research, in order to assess the various indices.
ECA - European Climate Assessment. Dr. A. Klein-Tank introduced the project and discussed recent results obtained from studying trends in indices for extremes in Europe, as well as future plans. Currently ECA joins 41 participants from 39 countries in Europe and the Mediterranean. Under a contract of the organization of meteorological services in Western Europe (EUMETNET), the ECA project will continue until at least 2008. The public website http://www.knmi.nl/samenw/eca gives access to all results (report, papers, presentations) and data (over 500 daily station series). Definitions for a core set of internationally agreed indices are also provided at the website. Among the indices described are those in the ClimDex software tool that has been developed by NCDC for the Caribbean and African workshops on extremes. Current ECA work includes improving data accessibility (netCDF, DODS, etc.), further developing techniques for homogeneity assessment of daily time series, providing monitoring of indices products (like the prototype shown for the 2003 summer heat wave in Europe) and investigating trends in extremes using more advanced statistical methods.
Dr. Klein-Tank noted that IPCC WGI (Beijing, 2002) recommended using more advanced methods than simple description of extremes. However, the question ³which methods best supplement the 'simple' descriptive indices" remains open, although it is recognized that the indices that have been used so far describe only part of the characteristics of extremes (amplitude, rate, duration, persistence, etc.). Later discussions at the meeting made clear that the infrastructure developed at KNMI in the past 5 years and the coming years for ECA will be available for use in other regions and for the ETCCDMI.
Comparing observed and modeled changes - Prof. Folland introduced a recent study by Kiktev et al., who compared observed and modeled trends in extremes of surface temperature and precipitation during the second half of the 20th century. The study derived from station daily data a few annual climate indices (e.g., Frost Days and Consecutive Dry Days) and interpolated them to model grids. The maps show that data are available in North America, the Eurasian continent and Australia, leaving gaps elsewhere. The model outputs are from 3 ensembles of HadAM3 runs with observed SST and 3 types of external forcing: solar variability and volcanic forcing; greenhouse gases, direct aerosol effects and ozone factors; and indirect aerosol effects. The main conclusions are:
There is obviously a need for comparison with coupled model data to determine if changes in extremes are unusual in the context of natural climate variability, and to estimate total anthropogenic effects on changing climate extremes. Prof. Folland supposed this could be done with HadCM3 and CCM3 first. He suggested that this should be a recommendation of ETCCDMI to the CLIVAR WGCM.
Extremes simulated in models - Dr. Zwiers started his talk by showing a few local daily precipitation series, where the observed series can hardly be distinguished from the modeled series. Undoubtedly, models can produce local variability similar to the observed. However, as he further showed, there are still big gaps and difficulties. The study he described involved daily temperature from 12 models participating in AMIP2 and precipitation from 16 such models, compared with station data and NCEP and ERA-15 reanalysis data. Classical extreme value theory was applied to fit annual extremes and to estimate 20-year return values. For temperature, models tend to underestimate warm and overestimate cold extremes. For precipitation, there are large differences between different data sets. Several models exhibit insufficient variability in the tropics and an erroneous split ITCZ. Some scenarios for 2050 simulated by CGCM2 were also presented.
Dr. Zwiers concluded that (1) while data for assessment of extremes is a serious problem, models are different from 'reality'; (2) it is difficult to identify causes of model problems, though there are indications that land surface parameterizations and parameterizations of convection are involved; and (3) there is confidence in some qualitative aspects of changes in future extremes (e.g., it seems likely that future precipitation extremes will be more intense than those experienced in today's climate).
Methodology in CCDMI - Dr. Stephenson focused his talk on statistical methods for estimating trends in climate change indices. He discussed indices used for monitoring and detecting climate change and how best to assess long-term trends in such indices. The basic definition of ETCCDMI "extremes" indices (as used in the CLIMDEX, ECA, STARDEX, and PRUDENCE projects) was first described and an interpretation in terms of "marked point process" of daily exceedences was presented. This was followed by a brief discussion of what is meant by an "extreme". The question of what exactly is a "trend" was then raised and various definitions were given. Trend analysis was illustrated on a particularly difficult example of a Russian Heat Wave Duration Index (HWDI) provided by Lisa Alexander at the Met Office. The probability model approach to describing a trend in the mean was explained. Residuals from a linear fit to the HWDI were shown to have very strange non-normal (non-Gaussian) behavior caused by the non-normal distribution of the HWDI variable. The advantage of using non-local in time robust fits such as LOWESS was demonstrated along with the idea of assessing trends on only the non-zero values of the HWDI. A list of problems with the usual linear least-squares trend approach was discussed and some alternative methods such as those based on Generalized Linear Models (GLM) was suggested. There is no unique definition of trend and so care should be taken when doing trend analysis of extremes indices. This will form ongoing work for the ETCCDMI team.
It was argued, however, that linear trend remains a most straightforward expression for climate change in many cases (e.g., global warming). Nevertheless, non-linear 'trends' will be more useful in analyzing series of climate extreme-related indices. During discussion, Prof. Folland emphasized that statistical uncertainty should be estimated for all climate change analyses. It was noted that well-designed GLMs could serve as a useful tool in estimating climate extremes and trends, with statistical uncertainty assessments explicitly included.
Due to technical problems, the planned phone session with Dr. T Peterson was replaced by a later telephone-communication between Drs. Zwiers and Peterson. Dr. Yan briefed the team on a presentation by Dr. Peterson sent before the meeting, which provides context for the discussion of regional workshops. After a brief historical review, the presentation listed the general goals of regional workshops:
Specific goals for each workshop include producing a peer-reviewed journal article on analysis of climate change for the giving region, and making available the data and indices used in the analyses. Careful post-workshop data analysis and coordination will be necessary to accomplish these goals.
Funding remains a critical factor for organizing workshops. The current information shows there could be limited funds for three regional workshops. Top three priorities were identified to be Southern Africa, Central America and South America. It was recommended that the Asian plateau and Southwest Asia be the next two targets. Possible actions for the ETCCDMI include
Having considered Dr. Peterson's report and other relevant workshops (e.g., the SCAR READER project for Antarctic daily/sub-daily data in September 2004, the APN SE Asia workshop organized by Neville Nichols in March 2004, and the APN Oceania Workshop by Jim Salinger in December 2003), the team agreed a prioritized list of 5 workshops for S Africa, S America, C America, the Asian plateau and SW Asia during 2004-2005.
Taking the Casablanca workshop (summarized by Easterling et al., Bulletin of American Meteorological Society October 2003) as an example, Dr. Mokssit guided the team to consider more details of the objectives, deliverables and contents of such workshops. It was suggested that the workshops should include two context-setting seminars that would be given by invited experts. It would be very useful to have one lecture provide a general analysis of observed climate change with a focus on the region over the last 100 years or so. This would be based on instrumental datasets, gridded data sets, and where possible, could also be set in the context of a longer period of time through the examination of proxy datasets. It would also be useful to have a second lecture about climate change modeling with an IPCC-type background and scenarios with regional focus. Some significant analysis work would be required to create such regionally specific lectures. The discussion also led to a number of action-related agreements, including:
The team was briefed on the preparation for the S African workshop by Dr. B. Hewitson over the telephone. The workshop will be hosted by the University of Cape Town and the South Africa Weather Service, Cape Town, during the 1st week of June 2004, immediately after the 9th International Meeting for Statistical Climatology. START has agreed to provide paritial support for this workshop and it is hoped that CLIVAR will be able to provide some additional support. The next steps in the development of this workshop will be to identify participants with the required data, to start training on software, and to plan seminars with invited experts. At least 15 computers plus appropriate technical support will be available for participants. As the team suggested, all countries in Africa are eligible to participate, with priority for those countries that were not covered by the Casablanca workshop. Some detailed requirements resulting from the team discussion will also be followed, as summarized in Table 1. The meeting and outputs will be reported over the coming weeks through a central ETCCDMI web. Dr. M. New agreed to draft a paper of the output of workshop as input to IPCC AR4.
The team recognized the necessity to prepare a Resource CD for the coming workshop. A list of contents of the CD-ROM were suggested, including:
The CD will have 2 directories, one including software and documents, and another with reference papers and reports.
L. Molion and S. Sensoy briefed the team on their preliminary proposals for the S America and SW Asia workshops. As Mr. Molion described, the working environment for climate study in Brazil is getting better. He was optimistic for organizing the S America workshop in August 2004. The team appreciated Mr. Sensoy's proposal of a 2-week workshop in Turkey in May 2005. All possible aspects were discussed for each workshop, including dates, venue, targeted area, associated international meetings, experts and ET coordinators. The number of participants was estimated at 15-25 for each workshop, with 1-2 person(s) per country. It was recommended that SW Asia workshop include some SE European countries. Some members hoped the Asian plateau workshop would include India, while acknowledging the difficulty due to the large difference in climate between India and the plateau area. Table 1 summarizes the workshop planning, general requirements and expected deliverables. Administratively, a letter to the PRs of involved countries should be sent out as early as possible from WMO. A screening questionnaire for choosing candidates should be sent out through the PR from WMO. Additional participants might also be identified through the use of a selection committee.
Action 2. Prepare a template letter to WMO Permanent Representatives in relevant countries, which states objectives and benefits of workshop and type of participant desired. Mokssit/Zwiers/Alexander/Haylock/Hewitson (Llanso/Detemmerman to resurrect previously used letter by early January 2004)
Action 3. Develop a template screening questionnaire for choosing workshop participants, based on that previously used. Mokssit/Zwiers/Hewitson
Action 4. Check all aspects of the ClimDex software that will be used in workshops. Klein-Tank/Alexander
Action 5. Prepare a Resource CD containing ClimDex documentation and other supporting materials for workshops. Alexander/New/Stephenson
Action 6. Arrange for translation of the software document into other languages. Stephenson for Portuguese /Llanso for French and Spanish (Carriers of Actions 4 and 5 to make sure that the document is ready by 1 March 2004)
Action 7. Identify targeted regions not covered by pervious workshops. Mokssit/Alexander/Haylock
Action 8. Organize and deliver workshops in coordination with supporting organizations and funders (see Table 1). Local organizers / ET contacts
Action 9. Prepare a short document on preferred practices for index construction and analysis, including guidance for 'reanalysis' of previously collected indices. Include a cross-referenced list of indices computed by the various pieces of software that are currently available so as to identify differences in index definitions. Also include advice on analysis methods appropriate to each index. Publish as a CCl/CLIVAR report and post on the web. Stephenson/Klein-Tank/Haylock/Zwiers/Zhang
Action 10. Make sure that workshop information (including outputs and follow-up activities) is available at the ET web and linked to the CLIVAR web site. Zhang/Zwiers/Yan
There is great potential for developing marine and oceanic aspects in the area of CCDMI. A few talks were invited at ETCCDMI-1 to address relevant issues.
Status of marine data and monitoring - Prof. Folland provided a briefing on recent advances in developing global marine and ocean surface data sets, including:
Having illustrated some improvements and problems in data, Prof. Folland concluded:
Monitoring and assessing marine climate change - Focusing on winds and waves, Mr. Swail outlined current status, plans and targets for marine climate monitoring. A 42-year wind and wave hindcast developed by the Climate Monitoring and Data Interpretation Division of the Meteorological Service of Canada showed that in the period 1958-1997, the 90-percentiles of significant wave height for winter increased in the northeast North Atlantic by 2-6 cm/year and decreased by 1-3 cm/year in the subtropical North Atlantic. Analysis showed possible links between the wave height increases in the northeast Atlantic and the NAO variability. Study of marine climatology has obviously benefited our understanding of climate system. Mr. Swail noted that a first Global Wave Climatology Atlas derived from 45-year ECMWF reanalysis data was recently published (http://www.knmi.nl/onderzk/oceano/waves/era40/index.html). Monthly and seasonal statistics, anomalies and other wave/wind data and information are also available online, e.g., the I-COADS web http://www.cdc.noaa.gov/coads-las/servlets/dataset.
Mr. Swail briefed the ET on some of the recommendations from CLIMAR-II 2003, including
The team discussed how to coordinate with JCOMM and agreed to write to JCOMM to express the need for marine and oceanic climate monitoring, change detection and indices.
Status of deep ocean data and monitoring - Dr. Bindoff reported on sub-surface ocean temperature and salinity observations (deeper than 10m) that are part of the CLIVAR program. He briefed the ET on the historical data that exist prior to the 1990's, during the WOCE experiment, and those that are planned for the CLIVAR repeat sections and for the carbon cycle work over the next 10-15 years. In addition, Dr. Bindoff reported on the Argo program, and showed new results from the November 2003 Argo Workshop in Japan. Although there are only ~1000 floats currently deployed in the oceans these data are already useful for estimating heat and freshwater changes on a zonal basis in the Northern and Southern subtropical gyres over the last decade. The importance of the Argo data set for climate change detection in the coming decades is going to increase. He appealed to the CLIVAR International Program for additional help in promoting and supporting Argo, and the CLIVAR repeat hydrography and carbon cycle sections.
Monitoring and assessing ocean climate change - Dr. Bindoff went on to report on ongoing efforts internationally to detect climate change signals in the major ocean basins. He showed examples from all ocean basins that support the notion that key oceanographic water masses in both the Southern and Northern Hemispheres are changing, in particular, the mode waters and intermediate waters in both hemispheres. The mode waters show warming, and the intermediate waters show a freshening, suggesting that both the surface temperatures and surface salinities are changing. The changes are broadly consistent with results from climate changes simulations.
The team discussed indices for detection of marine and ocean climate change. As Dr. Bindoff pointed out, one of the key issues in detecting climate change is the selection of suitable variables. Signal and noise are important considerations. It is clear from coupled model simulations of natural variations and climate change that variables such as the transport of heat (at 24N in the Atlantic) are not significantly different from the control in the climate change simulations. On the other hand, sea-surface temperature, and water mass properties are significantly different. Similarly it is important to look at the water masses that are most likely to change, i.e., those that are relatively close to their source regions, and in areas where the natural variations are likely to be smaller. Coupled models suggest that water masses in the southern Ocean are a more sensitive indicator of change, because the signal-to-noise ratio is higher there. Thus, for the detection and attribution of climate change, the best ocean variables to use for indices would emphasize storage terms (such as heat, salt, carbon) variations in the water-mass properties and volumes, sea-level and perhaps large scale changes in ocean stratification. It is also important that these indices have value for establishing relationships between the oceans and terrestrial variations. Together, these requirements suggest that there will be a need for both global scale and regional scale indices.
The team recognized that for climate change detection and attribution, it is important to create proper indices. Example indices were proposed, including:
However, it was noted that the number of indices should remain small and should include those that have been commonly used, such as the well-known ENSO indices. The team charged Dr. Bindoff, Prof. Folland, Mr. Swail and Prof. Karoly with the task of summarizing and circulating a list of ET-recommended marine and oceanic indices for the IPCC AR4, together with the comparative list of indices made by Dr. Haylock.
Action 11. Prepare a short document on suggested large-scale atmospheric, oceanic and marine climate indices for monitoring and detection that should be considered for the IPCC 4th Assessment Report (AR4). Bindoff/Folland/Swail/Karoly
Action 12. Write to JCOMM expressing need for marine climate change detection, monitoring and indices. Co-chairs/Swail
ETCCDMI indices website - Dr. Zwiers proposed that an ET website be developed. Such a site could be focused on indices, and could contain information on software for calculating indices (or links, e.g., those of NCDC, ClimDex, APN and STARTDEX), post-processed peer-reviewed indices (downloadable), workshop reports and follow-up activities, homogenized daily data (or links) where available, post-workshop indices and an updated list of references. The site could also establish a mechanism for recognizing users through password-controlled download. It was hoped that the site would have a clickable map for choosing regional subsets of indices. Dr. Mokssit suggested that some modeled scenarios vs observations (e.g., from the EU project PRUDENCE and Meteo France) could be included. Dr. Bindoff noted that the web should be viable for marine indices, which are presently not available. Prof. Jones enquired whether frequencies of extremes could be included. It was also recommended that the web incorporate bulletin board service to facilitate discussion among the team members and users.
GCOS website - Prof. Folland introduced the planning of the possible website. Challenging questions the web site will face include:
It is possible that BMRC and Met Office will lead the initiative in 2004. One problem deals with some official limits for data release from Met Office/Hadley Center.
Additional web resources - Existing web resources dealing with climate extremes and indices include those web pages in association with the relevant EU projects at the Climatic Research Unit (http://www.cru.uea.ac.uk) and that of the ECA (http://www.knmi.nl/samenw/eca/index.html). As Dr. Klein-Tank reported, the ECA web will remain operational until at least 2008. It is updated every 6-months, Statistics show good visiting rates, about 10 questions and 30k page views per week. A list of FAQs is therefore very helpful. Dr. Yan assured the team that CLIVAR website will include all these links.
National activities - Almost all countries are involved in studies of climate change detection and monitoring, due to the widespread impact of climate change on many economical and social aspects. In Australia, as Dr. Bindoff briefed, relevant research activities are mainly organized through BMRC, CRC, CSIRO and several universities. A main driving force for Australia is associated with the role of land cover change. For regional cooperation, Australia plays a role in organizing APN workshops on climate extremes and indices. In Brazil, as Mr. Molion briefed, it is difficult to obtain daily data from met offices for research. However, it is easier to get daily hydrological data associated with hydropower applications, including rainfall and river runoff. He suspected the current indices (e.g., the 57 indices of STARDEX) might not be suitable for tropical climate regimes and urged the development of indices that are more appropriate to the tropics. These indices might include those that reflect regional surface pressure variations that result from ENSO. He hoped the ET workshop would promote studies of climate in the country. A detailed report for Turkey by Mr. Sensoy, and one for the US special ad hoc Working Group on Climate Change Detection (IDAG, 2004) by Dr. Zwiers were also presented at the meeting. The reports are available at CLIVAR's web site (http://www.clivar.org/organization/etccd) through the ETCCDMI-1 meeting agenda. The team appreciated all the reported national activities, which are valuable for the development and implementation of CCl/CLIVAR's science plan.
During the last half hour of the meeting, the co-chairs reviewed the action list and deliverables. It was suggested that a timetable for ETCCDMI activities be established, taking into account the IPCC AR4 time lines and WMO/CLIVAR's long-term working plans. The co-chairs highly appreciated contributions from all participants and the help from the host, Prof. Jones and his team, which led to a very fruitful meeting. The 1st ETCCDMI team meeting was closed at noon on 26 November 2003.
Action 13. Develop an ET website focused on local indices. If feasible, include with group email/discussion capability and data-visualizing tools. Zhang/Zwiers
Action 14. Prepare a timetable for ETCCDMI activities, taking the IPCC AR4 and CLIVAR and relevant WMO activities into account. Co-chairs/Yan
The International ad hoc Detection and Attribution Group (IDAG) 2004: Detecting and Attributing External Influences on the Climate System: A Review of Recent Advances. Submitted to J Clim
Kiktev D, Sexton DMH, Alexander L, Folland CK 2003: Comparison of Modeled and Observed Trends in Indices of Daily Climate Extremes. J Clim 16: 3560-3571
Klein-Tank AMG, Konnen GP 2003: Trends in Indices of Daily Temperature and Precipitation Extremes in Europe, 1946-99. J Clim 16: 3665-3680
Easterling DR, Alexander LV, Mokssit A, Detemmerman V 2003: CCl/CLIVAR workshop to develop priority climate indices. Bull Amer Meteor Soc (October): 1403-1407
Frich P, Alexander LV, Della-Marta P, Gleason B, Haylock M, Klein-Tank AMG, Peterson T 2002: Observed coherent changes in climatic extremes during 2nd half of the 20th century. Clim Res 19: 193-212
Houghton JT, Ding Y, Griggs DJ, Noguer M, van der Linden PJ, Dai X (Eds) 2001: Climate Change 2001: The Scientific Basis. Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change (TAR IPCC). Cambridge University Press, 944 pp
Peterson TC and coauthors 2002: Recent Changes in Climate Extremes in the Caribbean Region. J Geophys Res 107 (D21): 4601-
A number of relevant reports and recent research papers are listed online at http://www.clivar.org/organization/etccd/index.htm.
Ms. Lisa Alexander
Met Office, Hadley Centre for Climate Prediction and Research
Fitzroy Road Exeter EX1 3PB United Kingdom
Tel: +44 (0)1392 884088, Fax: +44 (0)1392 885681
Dr. Nathan Bindoff
Director, Tasmanian Partnership for Advanced Computing
Antarctic CRC, University of Tasmania, GPO Box 252-80,Hobart, Tasmania 7001, Australia
Tel: +61-3-62-262986, Fax: +61-3-62-262973
Prof. Chris Folland
Met Office, Hadley Centre, Fitzroy Rd, Exeter, Devon EX1 3PB United Kingdom
Tel: +44 (0)1392 886646, Fax: 0870 900 5050 (UK), 44 (0)113 336 1072 (International)
Dr. Malcolm Haylock
Climatic Research Unit, School of Environmental Sciences, University of East Anglia, Norwich NR4 7TJ UK
Tel: +44 (0) 1603 593119, Fax: +44 (0) 1603 507784
Dr. Bruce Hewitson
Department of Environmental & Geographical Science, University of Cape Town
Private Bag, Rondebosch, South Africa 7701
Tel: +27 21 650 2878, Fax: +27 21 650 5773
Prof. Phil Jones
Climatic Research Unit, School of Environmental Sciences, University of East Anglia, Norwich NR4 7TJ UK
Tel: +44 (0) 1603 592090, Fax: +44 (0) 1603 507784
Dr David Karoly
Williams Chair and Professor of Meteorology, School of Meteorology, University of Oklahoma
100 E. Boyd St, Norman, OK 73019 USA
Tel: +1-405-325-6446, Fax: +1-405-325-7689
Dr. Albert Klein-Tank
Royal Netherlands Meteorological Institute, P. O. Box 201, 3730 AE De Bilt, The Netherlands
Tel: + 31 30 2206872, Fax: + 31 30 2204614
Mr. Paul Llanso
Chief, World Climate Data and Monitoring Programme, WMO
7 bis Ave. de la Paix, C. P. 2300, CH-1211, Geneva 2, Switzerland
Tel: (+4122) 730-8268, (+4122) 730-8214, Fax: (+4122) 730-8042
Dr. Abdalah Mokssit, co-chair
Direction de la Metéorologie Nationale, BP 8106Casa-Oasis Casablanca, Morocco
Tel: (212-22) 913435, Fax: (212-22) 913699
Mr. Luiz Molion
Departamento de Meteorologia, Universidade Federal de Alagoas
Cidade Universitaria, 57.072.970 Maceio-Alagoas, Brazil
Tel: (55-82) 2141365, Fax: (55-82) 3551025/2141615
Dr. Mark New
Climatology Research Group, School of Geography and the Environment, Oxford University
Mansfield Road, Oxford OX1 3TB, UK
Tel: +44 (0)1865-271917
Prof. Jean Palutikof
Climatic Research Unit, School of Environmental Sciences, University of East Anglia, Norwich NR4 7TJ UK
Tel: +44 (0) 1603 593647, Fax: +44 (0) 1603 507784
Dr. Thomas Peterson
National Climatic Data Center, Scientific Services Division
151 Patton Avenue; Asheville NC 28801-5001 USA
Tel/Fax: +1-828-271-4328, Tel (work): +1-828-271-4287
Mr. Serhat Sensoy
Turkish State Meteorological Service, Department of Agricultural Meteorology
Climatology Section06120 Kalaba/Ankara, Turkey
Tel: (90-312) 3022456, Fax: (90-312) 3612371
Dr. David B. Stephenson
Head of Climate Analysis Group, Department of Meteorology, University of Reading
Earley Gate, PO Box 243, Reading RG6 6BB, UK
Tel: +44 (0) 118 378 6296, Fax: +44 (0) 118 378 8905
Mr. Val Swail
Chief Climate Monitoring and Data Interpretation Division, Climate Research Branch
Meteorological Service of Canada, 4905 Dufferin Street, Downsview, Ontario M3H 5T4, Canada
Tel: 416-739-4347, Fax: 416-739-5700
Dr. Zhongwei Yan
International CLIVAR Project Office, Southampton Oceanography Centre
European Way, Southampton SO14 3ZH UK
Tel: +44 (0) 23 8059 6431, Fax: +44 (0) 23 8059 6204
Dr. Francis Zwiers, co-chair
Chief, Canadian Ctr for Climate Modelling and Analysis, Meteorological Service of Canada
c/o University of Victoria, PO Box 1700, STN CSC, Victoria, BC V8W 2Y2 Canada
Tel: (250)363-8229, Fax: (250)363-8247
Most of the acronyms used in this report are listed here. More can be found at http://www.clivar.org/publications/other_pubs/iplan/iip/appendix_6_acro.htm.
ACC Anthropogenic Climate Change
AMIP Atmospheric Model Intercomparison Project
AR4 4th Assessment Report (IPCC)
BMRC Bureau of Meteorology Research Centre (Australia)
CCDMI Climate Change Detection, Monitoring and Indices
CCl Commission on Climatology (of WMO)
CCM NCAR Community Climate Model
CGCM Coupled General Circulation Model
CGCM2 The Second Generation Coupled General Circulation Model (Canada)
CLIMAR Workshop on Advances in Marine Climatology
CLIVAR Climate Variability and Predictability (WCRP component)
COADS Comprehensive Ocean-Atmosphere Data Set
CRU Climatic Research Unit (University of East Anglia, UK)
CSIRO Commonwealth Scientific and Industrial Research Organization
ECA European Climate Assessment
ECMWF European Centre for Medium Range Weather Forecasts
ENSO El Niρo Southern Oscillation
ET Expert Team
ETCCDMI Expert Team for Climate Change Detection Monitoring and Indices
EU European Union
GCOS Global Climate Observing System (IOC/WMO/ICSU/UNEP)
GCM General Circulation Model
GLM Generalized Linear Model
GODAE Global Ocean Data Assimilation Experiment
GOOS Global Ocean Observing System (IOC)
GSN GCOS Surface Network
GTS Global Telecommunication System
GUAN GCOS Upper-Air Network
HadCM Hadley Centre Coupled Model (UK)
ICPO International CLIVAR Project Office
ICSU International Council of Scientific Unions
ICT Implementation/Coordination Team (CCl OPAG)
IDAG International ad hoc climate change Detection and Attribution Group (IDAG)
IGBP International Geosphere Biosphere Programme
IHDP International Human Dimensions of global change Programme
IOC Intergovernmental Oceanographic Commission
IPCC Intergovernmental Panel on Climate Change
ITCZ Inter-Tropical Convergence Zone
JCOMM Joint Commission for Oceanography and Marine Meteorology (GCOS/GOOS)
JSC Joint Scientific Committee for the World Climate Research Programme
KNMI Koninklijk Nederlands Meteorologisch Instituut (The Netherlands)
MCSS Marine Climatological Summaries Scheme
MICE Modeling the Impact of Climate Extremes
MSLP Mean sea level pressure
NAO North Atlantic Oscillation
NCAR National Center of Atmospheric Research (US)
NCDC National Climate Data Center (US)
NCEP National Center for Environmental Prediction (US)
OLR Outgoing Longwave Radiation
OPAG Open Programme Area Group (CCl)
PAGES Past Global Changes (IGBP component)
SLP Sea level pressure
SSG Scientific Steering Group
STARDEX STatistical and Regional dynamical Downscaling of Extremes
START Global Change SysTem for Analysis, Research & Training
UNEP United Nations Environment Programme
WCDMP World Climate Data and Monitoring Programme
WCP World Climate Programme
WCRP World Climate Research Programme
WGCM Working Group on Coupled Modelling (JSC/CLIVAR)
WMO World Meteorological Organization
WOCE World Ocean Circulation Experiment (WCRP component)