Objectives for the cloud climate project
SMHI is a heavy user of data from meteorological satellites, in particular data from the polar orbiting NOAA satellites as provided by the AVHRR imaging sensor. The cloud classification algorithm SCANDIA (Karlsson, 1996_1) was developed with the purpose of achieving an automatic extraction of information on cloud occurrence and cloud types by utilisation of the complete AVHRR dataset (radiances in five spectral channels) at its maximum horizontal resolution. SCANDIA is a multi-spectral thresholding scheme where thresholds are dynamically varied according to the present weather and illumination conditions. Thresholds have been determined by supervised training procedures during several years. Cloud classification results are used both qualitatively in the forecasting service and quantitatively in the objective analysis scheme MESAN (Häggmark et al., 1997).
The SNSB project has the following five objectives:
1. Investigate if SCANDIA results can be used to compile high-resolution cloud climatologies over the Scandinavian area covering a long period (here 1991-2000).
2. Investigate if SCANDIA cloud climatologies can be used to validate results from presently available satellite-derived global cloud climatologies (in particular, from the ISCCP project) and from regional climate simulations by climate models (e.g., as derived in the SWECLIM programme).
3. Compile cloud climatologies over the Baltic Sea area which could be of interest for model validation purposes and for physical process studies in the BALTEX project.
4. Develop an improved SCANDIA model capable of providing high-resolution cloud analyses over the entire Northern Europe including the North Atlantic area.
5. Assess the impact of long-term sensor degradation on cloud climatology results.
In addition, an obvious long-term goal of the project is to facilitate the establishment of an operational cloud climate monitoring at SMHI.
Methods
The first objective will be studied by use of archived SCANDIA classifications for the period 1991-2000. Cloud classifications from four daily NOAA overpasses over the Scandinavian area will be used to compile monthly cloud climatologies. In principle, the same method as described in the early SMHI pilot project focused on Nordic cloud conditions in 1993 (see Karlsson, 1994_1 and Karlsson, 1997) will be used. Results will be compared to corresponding information as defined by surface observations (SYNOP observations).
For the second objective, comparisons will be carried out with the ISCCP D2 and DX datasets (see Rossow et al., 1996) covering the Scandinavian region. If possible, also other cloud datasets will be studied (e.g., METEOSAT cloud climate datasets, the ECMWF ERA dataset and cloud climatologies from SWECLIM simulations).
The third objective means that the achieved cloud climate results from the period 1991-2000 must be adapted to the needs expressed by the BALTEX research community. This would concern both the included parameters and the database structure (e.g., access format and packaging). A good description of the information content and quality is also necessary.
The fourth objective reflects the obvious ambition to improve or replace the SCANDIA cloud classification scheme. The long-term cloud climate dataset should, when compared to corresponding surface observations and other cloud datasets, reveal the main deficiencies and strengths of the scheme. Remedies to deficiencies should be presented and tested. An important aspect here is to get assurrance that the strengths of the former scheme is retained in order not to introduce unwanted inhomogeneities in cloud climate datasets when changing to the use of a new SCANDIA scheme.
Finally, the fifth objective is an important quality control aspect of satellite-derived climate datasets. The cloud climatology is based on real-time results from SCANDIA cloud processing and is therefore subject to unwanted but existing errors due to degrading sensors over time. An attempt to assess the impact of these degradations must be made. Results from international efforts to monitor sensor degradation will be utilised and applied.
The international context
The development work in this project is closely linked to the SMHI participation in EUMETSAT SAFs involving cooperation with scientists at several European weather services. SMHI is a member of four SAF (Satellite Application Facilities) consortia established by EUMETSAT (European organisation for exploitation of weather satellite data) for increased and efficient use of weather satellite data and for the preparation for new satellites. The SAFs are denoted SAF for support of Nowcasting and Very Short-range Forecasting (SAFNWC), Ocean and Sea Ice SAF (SAFOSI), Climate monitoring SAF (CM-SAF) and Land surface applications SAF (Land-SAF). The SMHI responsibility is to contribute to the development of algorithms to derive cloud mask, cloud type, cloud top height and precipitating clouds products. For the Climate SAF, the SMHI focus is on development of methods to monitor the cloud climate over the entire Europe and adjacent ocean areas. Also the Arctic region will be covered to some extent. The SMHI SAF activities are described at at http://www.smhi.se/saf.
Contacts have also been taken with the ISCCP project and SWECLIM. There are also plans for an active participation in the main BALTEX BRIDGE project starting in 1999 (see SNSB project "Validation of cloud parameters in regional climate model simulations using NOAA AVHRR data")..
Acronyms
AMSU Advanced
Microwave Sounding Unit (NOAA-K and onwards)
ATOVS Advanced Tiros Operational
Sounder (NOAA-K and onwards)
AVHRR Advanced Very High Resolution
Radiometer (NOAA satellite)
BALTEX BALTic sea Experiment
BRIDGE The Main BALTEX Experiment, planned
for 1999-2001
ECMWF European Centre for Medium Range
Weather Forecasts
(Reading, UK)
ERA
ECMWF Re-Analysis project
EUMETSAT EUropean organization for the exploitation of METeorological
SATellites
HDF
Hierarchical Data Format
HIRLAM HIgh Resolution Limited Area Model - NWP
model developed by the Meteorological Institutes
in the Nordic countries plus Ireland, the Netherlands and Spain
ISCCP International
Satellite Cloud Climatology Project - part of the WCRP GEWEX project
MESAN MESoscale ANalysis scheme (SMHI)
METEOSAT Geostationary METEOrological SATellite (EUMETSAT)
MODTRAN RTM model
NOAA National
Oceanographic and Atmospheric Administration (USA)
NWP
Numerical Weather Prediction
RTTOVS Radiative Transfer model for ATOVS
calculations
RTM
Radiative Transfer Models
SAF
Satellite Application Facility - distributed element of the EUMETSAT ground segment
SAFCLIM SAF for Climate Monitoring (formed at the end of
1998)
SAFNWC SAF for Support to Nowcasting and Very Short-Range
Forecasting (formed in 1997)
SCANDIA SMHI Cloud Analysis model using Digital AVHRR data
6S
RTM model covering visible and near infrared region
SMHI The Swedish
Meteorological and Hydrologica Institute
SWECLIM SWEdish CLImate Modelling programme
SYNOP SYNOPtical weather
observations at surface stations
WCRP World Climate
Research Programme
References
Publications produced in the SNSB project
Karlsson, K.-G., 1996_2: Validation of modelled cloudiness using
satellite-estimated cloud climatologies, Tellus, 48A, pp 767-785.
Karlsson, K.-G., 1996_3: Monitoring the regional cloud climate in northern
Europe from NOAA AVHRR imagery, Proc. European Conference on
Applied Climatology, Norrköping,
Sweden, 7-10 May, 1996, pp 15-16.
Karlsson, K.-G., 1997_1: Cloud climate investigations in the Nordic region
using NOAA AVHRR data, Theor. Appl. Clim.,57, pp 181-195.
Karlsson, K.-G., 1997_3: AVHRR cloud processing for Nowcasting
applications and for cloud climate studies, Proc. 1997 Meteorological
Satellite Data Users’ Conference,
Brussels, Belgium, 29 Sep - 3 Oct 1997, EUMETSAT, EUM P 21, pp 69-76.
Karlsson, K.-G., 1998: Satellite sensing techniques and applications for
the purposes of BALTEX, Proc. Second Study Conference on BALTEX,
Juliusruh, Island of Rügen, Germany, 25-29
May 1998, International BALTEX Secretariat, Publ. No. 11, pp 98- 103.
Karlsson, K.-G., 2000: Satellite sensing techniques and applications
for the purposes of BALTEX, Contr. Atm. Physics, in press.
Other publications relevant for the SNSB project
Fernández, P., Fernández, J.M., Martín, F., LeGléau, Zwatz-Meise,
V. and Karlsson, K.-G., 1997: The SAF of EUMETSAT to support Nowcasting:
An introduction, Proc. 1997
Meteorological Satellite Data Users’ Conference, Brussels, Belgium, 29 Sep - 3
Oct 1997, EUMETSAT, EUM
P 21, pp 591-594.
Häggmark, L., Ivarsson, K.-I. and Olofsson, P.-O., 1997: MESAN -
Mesoscale Analysis (in Swedish), SMHI Reports and Meteorology,
No. 75, 77 pp.
Karlsson, K.-G., 1994_1: Satellite-estimated cloudiness from NOAA AVHRR
data in the Nordic area during 1993, SMHI Reports Meteorology
and Climatology, No. 66, 51 pp.
Karlsson, K.-G., 1994_2: Validation of HIRLAM cloud forecasts using
satellite-derived cloud cover statitistics, Proc. ECMWF Workshop on
Modelling, Validation and Assimilation
of clouds, Reading, UK, 31 Oct - 4 Nov. 1994, ECMWF, pp 263-276.
Karlsson, K.-G., 1995: Estimation of cloudiness at high latitudes from
multispectral satellite measurements, AMBIO, 24, pp. 33-40.
Karlsson, K.-G., 1996_1: Cloud classifications with the SCANDIA model, SMHI
Reports Meteorology and Climatology, No. 67, 36 pp.
Karlsson, K.-G., 1997_2: An Introduction to Remote Sensing in Meteorology,
SMHI, ISBN 91-87996-08-1, 315 pp.
Rossow, W.B., Walker, A.-W., Beuschel, D.E. and Roiter, M.D., 1996:
International Satellite Cloud Climatology Project (ISCCP), Documentation
of new cloud datasets, WMO/TD, 737.