*===========================================* | Minutes of the IOMASA Progress Meeting 1, | | 24-25 April, 2003 | | held at DMI, Copenhagen, Denmark | *===========================================* Start of meeting: 24 April, 9:30 End of meeting: 25 April, 14:00 Participants: ------------- IUP: Georg Heygster GH Christian Melsheimer CM Jungang Miao JM DTU-DCRS: Leif Toudal Pedersen LT Roberto Saldo RS DMI: Søren Andersen SA Rasmus Tonboe RT met.no: Harald Schyberg HS Frank Thomas Tveter FT SMHI: Per Dahlgren PD Tomas Landelius TL UAG: Stephen English (UK Met. Office) SE Carl Fortelius (FMI) CF Keld Q. Hansen (DMI) KH Markku Rummukainen (SMHI) MR Helge Tangen (met.no) HT 1. Introductory items: ======================= Welcome address. Introductory round. History of IOMASA initiative. 2. User Advisory Group (UAG): ============================= Rationale for the formation of the UAG - purpose: dialogue with end users in the below-mentioned fields - introduction of UAG members and their specific fields 3. Project Overview for the UAG members: ======================================== - Summary and Innovation - Project logic graphic - Project Structure: 5 Parts and 4 Phases - Work breakdown and Deliverables, see table, 1/WP 4. Presentation of the five Project Parts for the UAG : ======================================================== 4.1 Part 1, IUP: Remote sensing of atmospheric parameters ---------------------------------------------------------------- (G. Heygster): - Total water vapour (TWV) from SSM/T2, AMSU-B: * J. Miao' algorithm (using ratios of brightness temperature differences of 3 frequencies). * Example image: comparison of TWV over Antarctica; ECMWF data vs. data retrieved by the algorithms above. C C M. Rummukainen: What determines the TWV distribution seen over Antarctica C G. Heygster/J. Miao: Mainly the orography, but details and changes are C caused by local weather patterns that might be missed by the sparse C observing stations. C M. Rummukainen: It would be interesting to see a time series of these C TWV images (retrieved from SSM/T2 data). C * Transfer the algorithm to AMSU-B data and arctic conditions (needs redetermining parameters of the algorithm using radiosonde data) * Problems of the algorithm: The emissivity of the ground (water, ice), this is the subject of N. Selbach's PhD thesis (SEPOR/POLEX data). - Cloud liquid water (CLW) over ice from SSM/I * Ratio of polarisation differences of 2 frequencies. - Temperature sounding, reduce error caused by ground (sea, ice) emissivity: - include info on ice concentration/type - feed this info into the assimilation scheme, since the AMSU-A radiances rather than the temperature retrieved from them are fed into the assimilation scheme. C C M. Rummukainen: How sensitive is the CLW algorithm to the lowest clouds C (important in the arctic) C G. Heygster/J. Miao: Not known very well yet, would need validation with C data (i.e. CLW profile measurements) C M. Rummukainen: Consider SHEBA data or modeling based on SHEBA experiment C as to vertical structure. C C. Fortelius: Effect of ice clouds? C J. Miao: In general, the scattering by ice clouds is not very C important at those SSM/I frequencies (for CLW) and those AMSU-B C frequencies (for TWV) that are used here; except for strong C convective clouds. C But occasionally, retrieval did not work within arctic because of ice C clouds. C L.T. Pedersen: Can you detect where the retrieval is bad C (because C of ice clouds)? C J. Miao: Not yet, for this, we would need more validation C 4.2 Part 2, met.no/SMHI: Improving Numerical Weather Prediction ----------------------------------------------------------------- (H. Schyberg): Goal: Improve use of sounding data over Arctic. SMHI: AMSU-B, humidity met.no: AMSU-A, temperature - better surface emissivity modeling using ice info (now: just assuming constant emissivity, i.e. constant ice cover) - setup HIRLAM 3D-VAR for this; conduct impact studies Approach: Start with simple approach, later: use better emissivity models (from other Project Part) - quality control: cloud contamination * temperature and humidity channels can be contaminated by CLW (CLW is not a HIRLAM variable) * develop strategy to handle that. C C S. English: Implemented variable ice coverage at Met Office, but C effective quality control (CLW,cloud ice) is also required. C (T. Landelius): Humidity Assimilation (AMSU-B): - Sensitivity of T_b to TWV depends on humidity and temperature => need colocated temperature information - Arctic is dry => peak of weighting functions at low altitude ("surface shines through") => need surface emissivity Surface heat flux modeling: - Ice cover reduced heat flux by 2 orders of magnitude - HIRLAM uses 100% ice cover within ice edge - new surface scheme with different tiles (i.e., different surface types within one grid square, e.g. ice and open water) - ice information from within IOMASA 4.3 Part 3, DTU-DCRS: Empir.models for emissivity and backscatter of sea ice: ----------------------------------------------------------------------------- (L.T. Pedersen) SSM/I ice concentration algorithm (Comiso bootstrap algorithm) for emissivity information for other IOMASA Parts Problem in ice concentration retrieval: - Time series of area with 100% ice cover year-round: Algorithm yields > 100% in summer - Time series of area with ice cover of 0% (summer) to 100% (winter): Algorithm yields >100% in late winter (wet snow cover) - Time series permanently ice-free area: Some variation, between -20% and +20%, more variation in summer (!) Reason for that problem: - poorly known ice emissivity, ocean emissivity (wind influence) and atmospheric humidity Data from several campaigns (e.g. DTU/Convection campaign, 17 & 20 Mar 2001: AWI VIS scanner, TUDRAD 16/34 GHz, [...??]): Areas covered by 100% young ice (frazil ice) yield 50% ice concentration using both Comiso Bootstrap and NASA Team algorithms Prelim. Study AMSU-B 150 GHz and 89 GHz shows distinct signature caused by ice type (surface properties) AMSR: Like SSM/I, but some more low-frequency channels: explore that Microwave models: - Radiometry: Empirical model (from time series analysis); MWMOD - Scatterometry: Empirical model (from time series analysis); R. Tonboe's backscatter model 4.4 Part 4, DMI/met.no: Sea ice concentration retrieval ---------------------------------------------------------------------- (S. Andersen): - SAF HL (high latitudes) center (DMI, met.no) - Tight connection to Part 3 - State-of-the-art: * operationally: (1) NASA Team, Bootstrap (weather-filtered, static tie points); (2) SAF products (weather-corrected, monthly tie points) * Advances at 85 GHz (more experimental state) - ASI (L. Kaleschke) - SEALION (S. Kern) - Poorly known ice/snow contributions affects ice concentration retrieval, - SSM/I ice type not reliable (=> use scatterometer data ...) - Little reference data over sea ice => difficult to assess performance Goals: Overall: Improve description of leads/polynias in daily/hemispheric analyses Further: - Better accounting for sea ice/snow properties - Improve use of multiple sensors for concentration retrieval - Improve use of atmospheric fields - Improve use of high resolution channels - extensive assessment/validation Expertise: - DMI/met.no ice charting service: about 500 SAR images/year (classification, texture) - Thin ice detection scheme: (SSM/I-PR)/(Seawinds-PR) (R. Tonboe) - Backscatter model (snow cover/thickness, snow LWC (wet snow), snow grain size etc.) (R. Tonboe) - Tie points * weather correction * sensitivity of the various algorithms to CLW * Bayesian multi-sensor (scatterometer & radiometer, e.g., QuikSCAT & SSM/I) C C G. Heygster: As to ice edge plot (1%, 10%, 20%, 50%) just shown: NASA Team C algorithm smears out even ideally sharp ice edge to about 70 km (because C of the low resolution). Are the isolines in the plot just the C smeared-out edge? But then, how can it be that there are the 1% to 50% C isolines within only a few km? C S. Andersen/R. Tonboe: This is an artefact caused by the land mask. C 4.5 Part 5, DTU/DMI: Demonstration of real time processing and user interface ------------------------------------------------------------------------------ (L.T. Pedersen): IWICOS - demonstrate near real-time service running for an extended period of time - takes into account limited bandwidth of many users - for standard web browsers, JAVA - free - there is an offline archive, accessible to, e.g., all IOMASA partners IWICOS has regular users, about 100-200 hits/day, about 100-200 users/month, from about the following domains: research institutions, ice services, fishery, shipping companies, oil companies - suitable for data distribution to end-users in CONVECTION, IOMASA, GreenICE - also suitable for storing historic data for within IOMASA C C M. Rummukainen: Atmospheric Model developers would need quality C information in addition to data, and not just image data C G. Heygster/S. Andersen: We should discuss distribution for IOMASA C end users (like M. Rummukainen:); IWICOS as is (image data) might C not be suitable. C H. Tangen: How much maintenance does IWICOS need? C R. Saldo: Very little, mainly archiving to DVD; about one day per month C 5. Review of Phase 1: Results of Phase 1 of each Partner: ========================================================== 5.1 Part 1: WP 1.1: Sensor data and day 0 algorithms: ------------------------------------------------------ (C. Melsheimer): Data inventory: - Radiosonde: Global radiosonde station data available at IUP and DMI (preliminary quality check: only few profiles from Siberia o.K.); in addition, radiosonde data taken during research cruises of RV Polarstern are available from Alfred-Wegener Institute - AMSU-A,B, level 1B: Years 2000-2002 are available at IUP (about 200 GB of data); scripts exist to convert to Level 1C - SSM/I, F-14: Available at DMI, years 2001-2002 - QuikSCAT Data: Available at DMI - Further details in IOMASA web site member area C C M. Rummukainen: The polar ozone campaigns also have radiosonde data, C and also water vapour measurements; DMI people know details C S. English: Include information on the exact data source and format C (e.g., for radiosonde, TEMP, GTS, whatever); also document changes C in calibration (e.g. there was a change in calibration of F-14) C IOMASA member web site - Password-protected (i.e. non-public web site accessible to IOMASA partners) - Makes available information relevant for all project partners, e.g. * E-mail addresses * Meeting schedule, agenda, minutes * Reports * Documents: Proposal, DoW * List (and possibly links to PDF files) of relevant Literature * Publications and Presentations from IOMASA (with links to files) * Algorithms developed and used within IOMASA * Data inventory and information how to access/get the data - URL: http://www.iup.physik.uni-bremen.de/iuppage/psa/2001/iomasa-member or click on Member Area on the public IOMASA page on IUP web site http://www.iup.physik.uni-bremen.de/iuppage/psa/2001/14iomasa.html - New virtual web server at University of Bremen operational soon: easier URL C C L.T. Pedersen: Copyright problem with putting links to PDF files C of published papers (even own ones, depending on the journal; C even more so papers by other authors). C Maybe don't link them, since most of use can get online version C from our libraries. C For own papers, maybe put a preliminary version (internal report) C on the web site. C (G. Heygster): - Sea ice emissivity Literature: not much on emissivity at AMSU-A temperature-sounding frequencies except Miao's algorithm => we take that (Miao's) algorithm - Further relevant literature: * Liu & Curry 2003: "Arctic 'Hot Spots' at 37 and 85 GHz" - Anomalies up to 30K (15K) at 85 GHz (37 GHz) - Most important factors contributing * CLW ~ 10K * T_s ~ 10K * Open Leads ~ 2K (?) * Voss et al. 2003: "Ice types from SSM/I & QuikSCAT" - NTA: MY ice concentration increases in summer - one contribution: transitional melting / wet precipitation * Jossberger/Mognard, 1999: "Snow depth over land" - in spite of title, relevant for us (over ice) - snow metamorphism (driven by temperature gradient in snow) - Temperature gradient index (TGI): temporal average of (T_ground - T_air)/snowdepth - TGI estimate from T_b(19H)-T_b(37H) => estimate snow depth C C H. Schyberg: Use Miao's algorithm together with NWP profiles instead of C radiosonde C S. English: Work by Weng (will send information later) C (J. Miao): Validation of TWV retrieval from AMSU-B with TWV from GPS - Cooperation of IUP, GKSS, TU Dresden - Use GPS from Antarctica coastal stations (giving ground-based humidity measurements), and some over the continent using CHAMP satellite (occultation measurements) - Result: agreement, but error increases toward high TWV (algorithm saturation) - Exception: for 1 station (the northernmost and relatively warmest one, near the tip of the Antarctic peninsula) there is not much correlation. 5.2 Part 2 (met.no,SMHI): WP 2.1: Prepare NWP activities ---------------------------------------------------------- (H. Schyberg): Delivery of NWP fields for other project parts - Code/scripts to extract 2003-2004 HIRLAM20 data is implemented - 4 fields: Total CLW, TWV, 2m temperature, 10 m wind, 4 times daily (00, 06, 12, 18 UTC) - Area: Europe and Arctic except areas north of Bering Strait (East Siberian Sea, Chukchi Sea) - 1 daily GRIB file containing the 4 fields (1 vector, 3 scalar), about 6 MB - GRIB: free and open documentation and software available on the WWW - grid is rotated spherical with mesh-width 0.2 degree. - How to disseminate to IOMASA partners? - Probably generated monthly. C C M. Rummukainen: Only these 4 fields? How about, e.g. precipitation? C Generally, one recognises more parameters to be of importance later. C L.T. Pedersen: Precipitation, (accumulated over 6 hours) C H. Schyberg: We'll see if it can be added too, but this would need C 6-hour forecast data instead of analysis data. C C. Fortelius: But there is a spin-up effect. C C. Fortelius: Beware of boundary effect in HIRLAM (depending on C the forecast period, this may affect a zone up to about 20 grid C points) C Set-up of operational data stream for assimilation (H. Schyberg) ----------------------------------------------------------------- - Will set up experimental OSI SAF chain for production of additional ice products for IOMASA (MY ice fraction is not operational): Input -> Output Operational: SSM/I Total ice concentration (and more) (HDF, GRIB + quicklooks) Experimental: QuikSCAT, MY ice fraction (new for IOMASA), AVHRR later: Better emissivity information - Develop interface subroutine to import SAF ice data at AMSU footprints - Implement processing chain for near-real-time production of collocation files with AMSU-A level 1c, HIRLAM data and SAF sea ice data (for statistics, experimentation and assimilation): * AMSU-A 1c data from both own (met.no) receiving station and EUMETSAT ATOVS Retransmission Service (EARS) * EARS: 3 receiving stations (Tromsø, Greenland, Canary Isl.), availability within 30 min (?) * Reception facilities installed and AMSU-A level 1c files received at met.no * Work with interfacing ongoing (in parallel with change of operating system at met.no): cross-platform ASCII transfer, BUFR coding,production of colocation files etc. C C S. English: EARS 1c data might have calibration changes, why not C start with EARS AMSU-A 1a and produce 1c yourself. C G. Heygster: AMSU-B? C H. Schyberg: Done by SMHI. C Data stream (P. Dahlgren): ------------------------- - EARS will start running in about 1 week. WORKING: -local AMSU-A 1c -> BUFR_hirlam -> HIRLAM TO DO: - AMSU-B 1c -> BUFR_hirlam -> HIRLAM OR ? - EARS AMSU-A/B 1c -> BUFR_ears -> HIRLAM When EARS is running, the local antenna will no longer be used. (T. Landelius): Setup of operational data stream (ice concentration) for experimental data stream - Test data from OSI SAF - Use 10 km OSI-SAF data to create super-observations on the HIRLAM scale (ca 10-40 km depending on configuration). The super-observations of ice concentration are then fed into the HIRLAM surface analysis scheme and assimilated using successive corrections. - The SST and the fraction of water are modified based on the outcome of the ice analysis in order to get consistent information. 5.3 Part 3 (DTU): WP 3.1: Prepare sea ice modeling activities -------------------------------------------------------------- (L.T. Pedersen): Day 0 model of emissivity: empirical - analysis of time series (20 years of SSM/I); now also MWMOD - DTU has SSM/I global data since August 2002, stored on DVD - Data from own airborne field campaign CONVECTION (Greenland Sea Convection Mechanisms And Their Implications) - Requested snow data from NSIDC C C S. English: Alaska Ice Center might also have relevant data, S. English C will give L.T. Pedersen the contact information C April or May 2004: DTU campaign GreenICE (together with Scottish Marine Institute, AWI -- C. Haas's induction measurement of ice thickness from helicopter-- and others): 10 day drifting camp on sea ice north of Greenland; won't carry a radiometer, but other measurements will be done) 5.4 Part 4 (DMI): WP 4.1: Prepare sea ice concentration retrieval ------------------------------------------------------------------- (S. Andersen): Preparation and day 0 tools ---------------------------- - Radiative transfer models: MWMOD, ARTS - Ku band wind model function NSCAT2 - Sea ice concentration algorithms: * NASA, NASA2, Bootstrap, Norsex, Calval, Bristol, * Svendsen near 90 GHz, ASI, SEALION - Scatterometer ice type detection schemes * SAF revised ice type scheme * thin ice detection scheme - Synergistic ice concentration algorithm * Grandell 1999 * Day 0.5 (presentation by R. Tonboe later) - SSM/I / QuikSCAT Data: F-14 swath data 2001-2002 (43 GB) (?rather use F-13 because of calibration changes in F-14? see earlier comment by S. English, section 5.1) Seawinds (QuikSCAT) global 2001-2002 (63 GB) - Validation strategy (5 main areas: once/month; 1 additional area: once/(2 months)), relying on SAR scenes from the following sources: * AO (Announcement of Opportunity) data (free) ASAR, - 170 scenes (Assist) - 311 - 1270 (Cryosat) * 29 kEUR in IOMASA to purchase scenes (contingency, dual pol.) * ice service scenes Example images for ice information from Radarsat scene: - neural network approach - fuzzy logic classification (H. Schyberg): Example image: Bayesian ice type classification from QuikSCAT data: - supervised, with OSI SAF ice type classes - closed MY ice clearly over-estimated (extending far south especially along coasts, Anomalous winter sea ice SAR and scatterometer sigma_0 and radiometer T_b ------------------------------------------------------------------------- (R. Tonboe): - caused by temporary melts =>increase of: snow grain size, surface roughness; formation of ice crusts and layers in snow - after melt: FY ice mimics MY ice - ice concentration retrieval affected during and after melt. - melt occurs frequently near the ice edge in winter - sigma_0 and emissivity changes are long-lived (> 3 months) - Example: temporary melt at Baffin Bay, Dec. 2001: * FY ice: sigma_0 increases to MY ice value and stays * sigma_0 pol. ratio drops during melt * SSM/I PR goes up * SSM/I GR first goes up slightly, then goes down * NASA Team ice concentration drops from 100% to 75% and stays there - microwave signature modeling: * sigma_0 at C, Ku band: RT model (Tonboe) * emissivity at SSM/I frequencies: MWMOD - Conclusion: * temporary melt changes snow grain size, roughness, forms ice layers in snow * Bootstrap and 85 GHz algorithms least sensitive to emissivity changes * detection of temporary melts with GR, sigma_0, HIRLAM * IOMASA report before 2004 * paper to Remote Sensing of Environment before 2004 5.5 Part 5 (DTU-DCRS): WP 5.1: Preparation of Real time processing ------------------------------------------------------------------- and user interface ------------------- (L.T. Pedersen): - tool is there and running (see earlier presentation) - exact definition of IOMASA products needed 6. Review of Phase 1: Are we ready to start Phase 2? ==================================================== Yes. Exception: SMHI as to AMSU-B: slightly behind schedule, not ready yet, but within a few weeks. 7. Feedback from UAG ===================== K. Hansen: IOMASA is a very ambitious scientific project; overall objective: new products and improved input to weather models - in this case the end users are operational weather centers. Problems ­ challenges: - Need for UAG, role of UAG not clear; specific end user(s) not clear. - How to bring science to operation? - Quantify need for IOMASA, - Quantify expected output and long term goals for IOMASA - Quantify near real time operational products - Relations to SAF? GMES? Other? Focusing at the DMI operational users needs, a number of secondary products may be a result of IOMASA, i.e.: - sea ice edge (user defined, 0%, 1%, 10%...) - realistic representation of ice concentration near ice edge - ice types / thickness - ice surface roughness - surface winds - precipitation over water (visibility) S. English: Operational continuity is important for NWP centres, and therefore a fully supported operational product will be preferred to a product with limited lifetime. S. Andersen: Note: IOMASA = Development project, i.e. no guarantee to deliver things that are better H. Schyberg/G. Heygster: Of course, we can not oblige SAF to take our product H. Tangen: SAF is asked to seek partners to help improve their work, but be careful about quality. What are the end products (not clear)? There are other sea ice activities running in parallel (e.g. ICEMON and Northern View, both in GMES) -- need to communicate/coordinate S. Andersen: It is quite well defined and straightforward how an improved sea ice product will be implemented in OSI SAF H. Tangen: e.g., ICEMON is the opposite strategy compared to IOMASA: 1st phase (20 months): no research & development, but state-of-the-art products. Only then: R&D M. Rummukainen: Climate is not a focus of IOMASA (so why should I participate) -- -- but: Availability of new data is relevant for regional climate modeling. - for example melt episodes: how often do they occur? - for IOMASA data to be used for climate modeling: need quality flags, numerical data (not images) - Swedish ice breaker campaigns might be interesting for IOMASA - GLIMPSE project: talk to Klaus Detloff (AWI) G. Heygster: How to get SHEBA data? M. Rummukainen: Contact Colin Jones at Rossby-Centre (SMHI). L.T. Pedersen: Does that mean that, e.g., CLW fields, together with error bars and quality flags, are of interest for you? M. Rummukainen: Yes C. Fortelius: There is no CLW in HIRLAM, just TCW condensate (ice + water). - Boundary problems: TCW near boundary unreliable - As to assimilation experiments: validation with case studies and forecast verification statistics - suggest also to look at analysis increments (i.e., the impact of putting new information like humidity fields into model) and look at the physics that causes the analysis increment. - work with reference HIRLAM, not your national HIRLAM version. This would enhance your impact: If changes ad improvements are done and tested with reference HIRLAM, then chances are good that this is taken over by the national HIRLAM versions (but not vice versa!). S. English: As to your 3 experiments (1. ice data -> better heat flux; 2. assimilation of AMSU-A; 3. Assimilation of AMSU-B): how are you going to do that? Do you have the resources? It might be difficult to achieve the -- to me, rather ambitious -- goals if IOMASA. You need to run the model for a period of several weeks for each experiment Changes should be tested individually and together where interactions are considered likely (e.g. improved sea ice and improved sea ice emissivity model for ATOVS assimilation). H. Schyberg: Maybe we won't succeed to check all combinations S. English: Assimilate AMSU-B radiances instead of TWV product. Preferable since you want to assimilate the variable that has as Gaussian an error statistics as possible (which is rather radiances than quantities retrieved from them - TWV is related non-linearly to radiances). Gaussian error statistics is an assumption behind the assimilation. Is your main goal science (you mentioned publications as a major outcome) or operational application (as stated in DoW). ------- - UAG members are asked to submit their comments given here in some written form. 8. Action Items for PM1: ======================== 8.1 List of Deliverables: ----------------------- - Going through list of deliverables from Description of Work (DoW) and discussing which are the main deliverables that are greater interest outside IOMASA, either for the general public (P), or just the weather/ice services (S); this is also relevant for the IOMASA brochure (see 10.1) No. Deliverable title Date Nature for: Part 1: Remote sensing of atmospheric parameters (Partner 1) 1.1 Baseline data and algorithms for atmo- 6 Re,Da,Me P spheric remote sensing 1.2.1 Retrieval algorithm for TWV from 20 Re,Da,Me P AMSU-B data 1.2.2 Retrieval algorithm for cloud signature 29 Re,Da,Me P from SSM/I-B data 1.2.3 Retrieval algorithm for surface emissivity 29 Re,Da,Me P at AMSU-A frequencies (1.3.1 Fields of TWV of investigation period 32 Da ) (1.3.2 Fields of cloud signature of investigation 32 Da ) period 1.3.3 Operational processing chain for cloud 32 Re,Da P signature (1.4 Validation report for TWV and cloud sig- 36 Re ) nature Part 2: Improving numerical weather prediction models (Partners 4,5) (2.1.1 Report on setup of operational data 6 Re ) stream 2.1.2 2 data years of NWP fields: Wind, 6 Da S TWV,liquid water path, and surface tem- perature 2.2.1 Report and programme code on humid- 20 Re,Me P ity assimilation into NWP 2.2.2 Report and programme code on temper- 29 Re,Me P ature assimilation into NWP 2.2.3 Report and programme code on interface 29 Re,Me S+P implementation (sea ice) 2.3 Report on real time assimilation system 32 Re,Da P for TWV and improved temperature as- similation 2.4 Validation report on assimilation impact 36 Re P Part 3: Empirical model for emissivity and backscatter of sea ice (Partner 2) 3.1.1 Offline data and day 0 algorithms for 6 Re,Da,Me P Part 3 (3.1.2 HIRLAM data of project year 1 12 Da ) 3.2.1 Report and programme code for emissiv- 22 Re,Me P ity and backscatter model of sea ice 3.2.2 Report on improvement potential with 22 Re P sensors AMSR(-E) 3.3 Report and programme code for influ- 32 Re,Da P ence of snow (3.4 Validation report for sea ice model 36 Re ) Part 4: Sea ice concentration retrieval (Partner 3) 4.1 Data sets and day 0 algorithms for sea ice 6 Re,Da P retrieval 4.2 Report and programme for retrieval of 29 Re,Me P sea ice concentration 4.3 Sea ice concentration data set of investi- 32 Da P gation period 4.4 Validation report for sea ice retrieval 36 Re P Part 5: Real time processing and user interface (Partner 2) (5.1 Data formats and software interfaces for 12 Re ) real time production (5.3 Real time production and distribution 32 Re,Da) system 5.4 Demonstration report and data 36 Re,Da P Nature: Re = Report, Da = Data, Me = Method C C L.T. Pedersen: Copyright issues with, e.g., Radarsat images C 8.2 Results in the sense of TIP: ---------------------------- - Draft of TIP must be ready latest after one year, i.e. November 2003 - main task: identify results, such as: algorithms, models, data sets (in all about 5 to 10 results) 1) TWV, CLW retrieval algorithms 2) Assimilation scheme, heat flux - Each IOMASA partner to submit about 2 results from his project part to coordinator within about 1 month - C. Melsheimer)to send example TIP (project on detection of hydrocarbon in water) to all partners. 8.3 Input for Management Progress Report (due end of May): ----------------------------------------------------------- - State that you are on schedule, and if not, why so - State person-months used - submit by mid-May to coordinator - L.T. Pedersen sends an example to all 9. Project Management: ====================== 9.1 Reports that come with the Phase 1 deliverables: ------------------------------------------------------ - please send ASAP to coordinator who will produce a common cover page and forward everything to EU. 9.2 Next meetings: ------------------ (UAG only expected to attend MTR in Aug 2004 and FP in November 2005) PM2: Oct 30/31 at SMHI, Norrköping (not finalised yet, finalise date by end of May, get feedback on date when circulating minutes). MTR: Aug 26/27, 2004, met.no, Oslo (tentatively) 10. Additional topics ===================== 10.1 IOMASA brochure (C. Melsheimer) ------------------------------------- - brochure to present/introduce IOMASA to scientists and forecast people - request for comments. => first comments: - it is not clear if things refer to * end user products * end user software * better scientific knowledge - more emphasis on deliverables (see 8.1, discussion about list of deliv.) - which contact person for what deliverable/domain - brochure should be printed once only (not updates; too expensive) 10.2 AMSR (G. Heygster) --------------------------- - G. Heygster is PI for AMSR sea ice - 3 problems with AMSR: * hot load has uneven temperature distribution; * cold load: mirror too small for antenna beam * radio frequency interferences (at 6 GHz H and V) over industrialized areas - CLW from AMSR: first results 11. Any other business: ======================== - link to IOMASA IWICOS from IOMASA page - add e-mail addresses of UAG to IOMASA member page (how safe is a password-protected web page with respect to address-collectors for spam?), or make a distribution list. - presentation files of this meeting to C. Melsheimer --------- GLOSSARY/ACRONYMS: ------------------ ASI ARTIST sea ice algorithm CHAMP CHAllenging Minisatellite Payload (GFZ Potsdam/DLR) CLW cloud liquid water DMI Danish Meteorological Institute DTU-DCRS Technical Univ. of Denmark, Danish Center for Remote Sensing GKSS GKSS Research Center (near Hamburg, Germany) GLIMPSE Global implications of Arctic climate processes and feedbacks (http://www.awi-bremerhaven.de/www-pot/atmo/glimpse/) GMES Global Monitoring for Environment and Security GreenICE Greenland Arctic Shelf Ice and Climate Experiment (http://www.dmi.dk/f+u/ocean/GreenICE.htm) Icemon operational ice monitoring for marine operations and climate change (http://www.nersc.no/ICEMON/) IUP Institut für Umweltphysik (Environm. Physics), Univ. Bremen IWICOS Integrated Weather, Sea Ice and Ocean Service System LWC liquid water content met.no Norwegian Meteorological Institute NSIDC National Snow and Ice Data Center OSI SAF Satellite Application Facility (SAF) on Ocean and Sea Ice SEPOR/POLEX Surface Emissivities in POLar Regions - POLar EXperiment SHEBA Surface HEat Budget of the Arctic ocean (http://sheba.apl.washington.edu/) SMHI Swedish Meteorological and Hydrological Institute TCW total cloud water TWV total (column) water vapour Minutes prepared by Christian Melsheimer