Abstract
Global observations are important for detecting climate change, understanding the present climate and predicting climate variability. Such observations, integrated into models provide immediate benefits to society in the form of improved forecasts of weather and climate. Clouds are a high priority problem for the planned Global Climate Observing System and for atmospheric models (GCM’s and weather forecast).
CLIWA-NET focuses on observations of cloud liquid water and vertical structures, and evaluation/improvement of parameterisations. A prototype of a European cloud observing system will be established. CLIWA-NET co-ordinates the use of existing, mostly operational, ground-based microwave radiometers and profiling instruments. The network data will be integrated with satellite estimates of cloud water. Based on these observations cloud parameterisations will be evaluated/improved.
The project is carried out in co-ordination with BALTEX.
Objectives
Contribute to the development and implementation of the Global Observing System with a focus on cloud observations.
Description of the work
The CLIWA-NET project establishes a prototype of a European cloud observing system by co-ordinating the use of existing, ground-based passive microwave radiometers and profiling instruments. In total 12 stations within the BALTEX modelling area will contribute to this network. An unprecedented microwave radiometer calibration campaign will be organised in combination with a regional network (100x100 km 2 ). The data from the ground-based remote sensing instruments will feed high quality cloud information, with high temporal but poorly spatial resolution, into the calibration of satellite-based estimates of cloud water content with high spatial resolution. New procedures will be developed to fully exploit the synergy.
The combination of vertical profiles of cloud water and temperature information will enable an accurate detection of super cooled water layers. These layers are responsible for in-flight icing, which is considered to be one of the major risks in today’s aviation..
The retrieved CLIWA-NET data-sets are used for an objective evaluation of the performance of state-of-the-art cloud parameterisation schemes. The focus will be on liquid water path (LWP) and vertical structure of cloud amount and cloud water. Three lines of research are pursued:
The cost and complexity of the available microwave radiometers presently hamper the implementation of an operational network. For this reason, the design of a low-cost operational microwave radiometer by a commercial company is included in this project.
The end users organised in the "CLIWA-NET user’s advisory group" will provide suggestions on, and judge the social-economic aspects of the project.
Milestones and expected results
The 5 CLIWA-NET workshops and the observational periods are important milestones.
2.1 Objectives
The overall objectives of this project are:
Observations
The prospect of global climate change resulting from increasing concentrations of greenhouse gases has become a major concern and has moved climate issues to the forefront of the international political agenda. Systematic and comprehensive global observations will lay the foundation for improving our capabilities for detecting climate change, understanding the present climate and predicting climate variability. Such observations, integrated into models of the climate system would provide immediate benefits to society in the form of improved understanding and forecast of climate.
The Global Climate Observing System (GCOS) organisation is the international body which co-ordinates all activities which are related to an international observational program for climate research. One of the gaps in the Global Climate Observing System is the lack of reliable/systematic observations of clouds in general and more specifically, parameters like vertically integrated cloud liquid water (LWP). Cloud water/ice and its vertical structure are very important parameters, which link dynamics, water vapour, precipitation and radiation.
In this project it will be demonstrated that a cloud observing system based on the intelligent integration of existing ground-based observations and (operational) satellite remote sensing data will be able to fill this gap. A network of already existing microwave radiometers, which are the only instruments measuring the LWP with sufficient accuracy, will be employed on a continental and a regional scale. At many of these stations additional information on the cloud vertical structure will be available from radar/lidar/IR-radiometer data which also will be used in the integrated approach. The data-set is complemented by the observations from the precipitation radars over the area and the standard meteorological observations.
The spatial distribution of vertically integrated liquid water will be reconstructed from a synergetic combination of measurements from the satellite based AMSU and AVHRR instruments and measurements from a network of microwave radiometers. In this way we will obtain a consistent, validated, high quality, high-resolution data set on integrated cloud liquid water fields and vertical structure of cloud water on a continental and regional scale.
In-flight icing is considered to be one of the major risks in today’s aviation. Icing is the deposition of super cooled liquid water on the aircraft frame. Icing affects the aircraft’s flight characteristics. The identification of super cooled water layers is of utmost importance for aviation. The combination of vertical profiles of cloud water and temperature information will enable an accurate detection/prediction of these conditions. The results for the detection of super cooled water layers will be evaluated with representatives from the aviation authorities.
Model evaluation/improvement
The importance of cloud water observations and the information on the vertical structure of clouds will be demonstrated by the evaluation/ improvement of state of the art cloud parameterisations. Because of the known poor representation of clouds in the NWP models, the quality of cloud forecasts from models is still limited. This has large socio-economic impacts for business areas like: airports, solar energy, road construction, tourism, breweries, ice industry, etc..
Mainly the lack of adequate data hampers the development of better cloud parameterisations in NWP’s and climate models. At present, differences between the predictions from various atmospheric models can, to a large extent, be traced back to the differences in the treatment/representation of cloud processes.
As stated by the IPCC’95: "the most urgent scientific problems requiring attention to determine the rate and magnitude of climate change and sea level rise are the factors controlling the distribution of clouds and their radiative characteristics....". The same was concluded in the AMIP project (e.g. Gates et al., 1999) in which the outputs of 30 atmospheric models were compared for a ten-year run. The differences in the output and the effect of cloud parameters were enormous.
Frequency bands for satellite systems operating in the millimetre part of the radio spectrum have been allocated for both broadcast and communication systems. The push to use higher carrier frequencies stems from the already large demand for use of the Ka-Band spectrum allocations, and also from a desire to use ever wider bandwidths for very high data rate transmissions of several hundred mega bits per second - this will be necessary for the transmission of internet and video information.
Experience gained has shown that there is a need to account for attenuation caused by cloud and gases in the atmosphere when transmission frequencies above 20 GHz are employed and, so called, ‘low margin’ system availabilities below 99% are considered. The characteristics of fade depth and duration associated with cloud cover needs to be quantified in order to develop, and assess the accuracy of, path attenuation prediction methods at these frequencies.
Systematic observations of the spatial distribution of cloud liquid water path on a regional and continental scale will provide a highly valuable input to prediction models of the slant path fading at EHF that is to be expected over such areas.
Implementation
The implementation of an operational network is presently hampered by the enormous cost and complexity of the available microwave radiometers. For this reason, the design of a low-cost operational micro-wave radiometer by a commercial company is included in this project.
The most prominent experts in ground-based microwave radiometry are among the project partners. This together with the network of about 10 different microwave radiometers and the intensive, unprecedented radiometer comparison will provide the necessary information required for the design of the low cost radiometer (number of channels, frequencies, stability and absolute accuracy). The design will focus on a low maintenance and simple calibration techniques to enable the use in operational networks.
Because of the known importance of cloud water current and future satellites (e.g. MSG, METOP, ENVISAT, CLOUDS) put in large efforts to measure this parameter. Over land, however, this parameter can only be retrieved in a rather indirect way, which opens a gap in our current validation possibilities. The network envisioned in the proposal will close this gap.
Contribution to BALTEX/BRIDGE
BALTEX has clear applied objectives with important socio-economical deliverables. The exchange of saline water between the Baltic Sea and the world ocean, improved prediction of river flooding and extreme weather events are examples of such deliverables. A particular important question is to better understand how such events may change in a future climate.
The BALTEX Science Committee has outlined a series of fundamental tasks which each will be needed to successfully carry through a large scale field experiment, BRIDGE, to be undertaken during the period from autumn 1999 until spring 2002. It will include close co-operation with Poland, the Baltic States, Belorussia and Russia. Institutes within the participating countries are prepared to enhance the observing system during this period as well as making other institutionally resources such as computing and other technical facilities available for BRIDGE.
The key aim of the BALTEX applications to the EU 5th Framework program is to explore and integrate these data by developing new methods and new techniques. The main result of the joint exercise will be the development of new predictive capabilities and thereby the necessary tools to realistically predict future changes of the hydrological cycle. The CLIWA-NET project is the major activity in BRIDGE in the area of cloud observations and process/modelling studies.
2.2 INNOVATION
Integration of cloud observations from ground-based and satellite RS on a routine basis
The retrieval of the spatial distribution of vertical integrated liquid water over land from a combination of a ground-based network of microwave radiometers/cloud radars/lidar ceilometers/infrared radiometers and data from different satellite instruments is truly innovative.
Satellite retrievals have a good spatial coverage but are relatively inaccurate and the time sampling is often poor. Ground-based microwave radiometers measure the column integrated amount of liquid water with high accuracy. Information on the vertical distribution of the liquid water is limited or not available at all. Cloud radars measure the vertical structure of clouds with high accuracy. However, the retrieval of the absolute liquid water amounts from the observed radar reflectivity is hampered by the presence of single larger droplets in many water clouds. From the lidar ceilometer and infrared radiometer data information on the cloud base height and cloud base temperature can be retrieved. Both instruments have the disadvantage that they cannot penetrate clouds with optical depths of 2 or larger. Precipitation radar data monitor the onset of precipitation and a measure of precipitation intensity, but provides no information on cloud liquid water.
From the above it is clear that each instrument contributes in a unique way to the characterisation of a cloud field. This project proposes to combine these already existing sources of information into synergetic products of the highest quality and best possible temporal and spatial resolutions.
Design of a low-cost microwave radiometer
Cloud liquid water is an important parameter for characterising clouds, which can only be measured with sufficient accuracy over land by ground-based passive microwave radiometry. Up to now the cost of purchasing a microwave radiometer suitable for low maintenance operation are high, typically 150 - 600 KEURO. As a result these instruments are mainly used and even built in research environments. We estimate that the cost of a newly designed microwave radiometer will be less than half of existing commercial systems. Rough estimates indicate that a price less than 50 KEURO may be achievable. The active participation of national meteorological services in this project demonstrates the importance of these observations. This will help to stimulate the integration of the low-cost microwave radiometers into the existing meteorological observational networks.
BALTEX/BRIDGE: Innovative way to bridge the gap between observations and models.
The BALTEX programme is an international program under the World Climate Research Programme with the objective to understand and predict the hydrological cycle in the Baltic Sea region. This work also includes the prediction of extreme events such as river flooding. It constitutes a broad interdisciplinary effort of some 50 research institutes in a wide range of countries involved in meteorological, hydrological and oceanographic research.
Within BALTEX a large effort is made to integrate multi-disciplinary research areas with the joint goal of observing, analysing and understanding the hydrological cycle of the Baltic Sea catchment area. The basic BALTEX programme elements include experimental and numerical process studies, numerical modelling, data-assimilation, re-analysis of existing data-sets, climate studies, and application of remote sensing data.
BALTEX will conduct its main modelling and observational experiment BRIDGE from October 1999 - February 2002. The central aim of BRIDGE is to generate comprehensive data sets for process understanding and budget studies of the entire Baltic Sea (modelling) area. The CLIWA-NET project is the leading project within BRIDGE for clouds and radiation studies.
Clouds affect our daily life in many ways. Much more than air temperature clouds dominate our perception of weather and thus have an enormous influence on our daily activities and our health. This fact is completely at variance with our knowledge about clouds, their representation in climate and weather forecast models and our ability to predict clouds. It is their high variability in time and space, which makes clouds both hard to monitor and to model.
It is well-known, that clouds are directly linked to the dynamics of the atmosphere. The most important parameter linking dynamics to clouds, in both the real world and in forecast models is the water content of clouds. Passive microwave remote sensing is by far the most direct and accurate method to estimate cloud water content. Over the oceans microwave remote sensing from satellites has been proven to be the most accurate method to determine this parameter. Unfortunately, over land areas this technique fails. Here remote sensing methods must rely on very indirect information mainly taken from cloud reflection of solar radiation. Thus, important cloud information does not get into our models in the area where people live.
LWP alone is an important parameter for the validation and assimilation of numerical weather forecast models (as proved in the EU project CLOREVAL, ENV4/CT96-0302). However, for many applications it is also crucial to know at which altitudes the water is located. Several instruments can be used to obtain this information: cloud radars, cloud lidar ceilometers and IR-radiometers.
We propose to close the gap between models and observations by establishing a prototype of a European cloud water observing system. This will be achieved by co-ordinating the use of existing, mostly operational, ground-based passive microwave radiometers and profiling instruments. This network will feed high quality cloud information, with high temporal but poor spatial resolution, into the calibration of satellite-based estimates of cloud water content with high spatial resolution.
The observing strategy is strongly related to work done within the Dutch Cloud Detection System (CDS) which was operational in 1995 and 1996. In the ground-based network ten cloud and radiation stations were distributed over a 120x120km2 area. Every station consisted of a lidar ceilometer, an IR-radiometer and a pyranometer. The ground-based observations were combined with the observations from the operational meteorological satellites Meteosat and NOAA/AVHRR. It was demonstrated that the integration of the observations from the ground-based and satellite instruments result in a considerable improvement of the retrieved cloud products. The CDS provided a data set on cloud cover and cloud base heights for the evaluation of atmospheric model output on clouds .
To demonstrate the potential and necessity of liquid water and cloud structure observations, cloud parameterisations from the major operational European weather forecast models are evaluated with this data. Possible improvements will be investigated.
The proposal includes also the definition of an optimal but low-cost instrument to enable the extension and/or export of the network to less developed areas of the globe.
A wide range of benefits will accrue from the network. Firstly, scientific and technical knowledge, so far unused, will be transferred into an operational environment. The expected improvement of cloud forecast will impact many areas. These range from the prediction of precipitation as the most important process influencing the hydrological cycle, water availability and quality, to solar radiation for solar energy use to UV-radiation influencing people's health, to mention but a few. Apart from this, the set-up of the network links together technical and scientific people in an area still rather poorly developed in Europe concentrating on a most important, but unique topic. Last but by no means least, the network will serve to monitor for the first time quantitatively the parameter most important for the human perception of global change, namely clouds.
Similar observing strategies can be used to identify super cooled water layers. The identification of super cooled water layers is of uttermost importance for aviation. Super cooled water layers are responsible for in-flight icing which is considered to be one of the major risks in today’s aviation. Icing is the deposition of super cooled liquid water on the aircraft frame and affects the aircraft’s flight characteristics seriously. The combination of vertical profiles of cloud water and temperature information will enable an accurate detection/prediction of these conditions. The results for the detection of super cooled water layers will be evaluated with representatives from the aviation authorities.
Representatives of the end-users will be organised in a "User Advisory Group" which will be installed to advise on the optimisation of the applicability of the project output.
The project will be carried out under the umbrella of BALTEX. The co-ordination of the experimental periods within the BALTEX/BRIDGE program assures the availability of a large variety of supporting data and infra-structure. The major aim of BALTEX/BRIDGE is to provide an extensive data base of observations to be used by the modelling groups of the BALTEX community. By incorporating CLIWA-NET in the BALTEX/BRIDGE program, the exploitation of the data in a broader context is guaranteed. Furthermore, access to the standard meteorological observations for the Baltic area is obtained through the BALTEX Meteorological Data Centre.
In the remainder of this section we will discuss the relation between CLIWA-NET and earlier EU-projects. Furthermore, the project management and co-ordination, the observations (ground-based network, Satellite products, Integration), the model evaluation/improvement, and the implementation will be discussed in more detail.
The project will be managed by the project Management Team (MT). The MT is chaired by the co-ordinator (KNMI). The three workpackage managers are the other three members of the MT (WP2000: MIUB, WP3000: IFM and WP4000: KNMI). The MT has the following tasks:
The User Requirement Document will be part of the kick-off workshop report. Decisions affecting the objectives/outcome of CLIWA-NET have to be approved by the MT. Each workpackage manager is responsible for the organisation of the workpackage. They organise workpackage meetings and monitor the progress made. They report to the MT.
This workpackage has the major goal to come up with highly accurate column integrated cloud liquid water values (LWP) and gather additional information on the vertical structure of clouds at 12 stations in the BALTEX modelling area. To achieve this goal a network of microwave radiometers, IR-radiometers and possibly lidar ceilometers (see Figure 1) will be operational in the BALTEX area (during two campaigns of approx. two months) and on a local scale (100x100 km2, one campaign of approx. 2 months). At max. 4 stations cloud radars will be employed to measure vertical cross-sections of clouds. All the instrumentation is already existing. However, to optimise the coverage, some of the instruments will be moved to a different location. Also data from the BALTRAD precipitation radar network will be included in the data sets. The following groups will participate by providing the data for the campaigns: MIUB, KNMI, CNRS, HUT, DWD, Chalmers, UNIBE, GKSS and CCLRC. The archiving and co-ordination of this part will be organised by MIUB (BALTIC Area Cloud Network) and KNMI (BBC-campaign). Research and analysis of the data for the study of cloud processes is done by MIUB, GKSS and KNMI.
A major contribution to the CLIWA-NET observational plan will come from this workpackage. MIUB will participate in the CLIWA-NET MT.
Instrumentation
Several types of passive microwave radiometers (see list of stations), which are otherwise used for different purposes (meteorology, geodesy or telecommunication) will be involved. Algorithms for the LWP retrieval from the measured brightness temperatures have to be adapted for each site due to the different frequencies, number of channels and the available additional instrumentation (see below). These measurements will not only be included in the LWP retrieval but also give important information on the vertical cloud structure. The LWP error estimates calculated for each site will give the optimal frequency and accuracy requirements for the low cost radiometer. A by-product of the measurements is the vertical integrated water vapour (IWV), which will be provided to the GPS related efforts in BALTEX.
The vertical structure of clouds can be measured best by state-of-the-art cloud radars. These radars measure the radar reflectivity with a vertical resolution of typically 50 m. The radar systems operate at different frequencies: 3, 35, and 94 GHz. The observed reflectivity profiles can be converted to liquid water profiles. However, the occasional presence of single large drops in the cloud will cause a serious degradation of the accuracy of the retrieved liquid water profiles. For this reason it is important to have a collocated microwave radiometer next to a cloud radar, which give a constraint for the integrated liquid water content (LWC). Measurements of the doppler velocity and polarimetric quantities (ZDR, LDR) will help to distinguish between water and ice clouds.
BALTRAD is the network of 29 weather radar systems, mostly C-band Doppler systems, in and proximate to the BALTEX region. Products consist of radar reflectivity factor (dBZ) images, composites based on these, wind profiles, and three and twelve-hour gauge-adjusted, radar based accumulated precipitation analyses. Due to clutter, anomalous propagation and the variation of the radar reflectivity factor as a function of range and altitude, these products only have a basic level of quality, if such factors are not treated adequately. For CLIWA-NET, enhanced quality control of these products will be achieved by combining several methods and integrating METEOSAT IR and NWP temperature, humidity and pressure fields information. Time series of radar reflectivity and precipitation will be extracted at the locations of the stations to derive the environmental parameters for the onset of precipitation.
Cloud base heights can be measured by lidar ceilometers. Laser pulses are transmitted and the instrument measures the power, which is reflected by clouds and/or aerosols. Internal algorithms are applied to derive the cloud base heights from these profiles. Up to three cloud layers can be identified. The algorithms are based on the detection of pre-defined features in the backscatter profile. Several types of lidar ceilometers will be used within the project. However all systems are comparable in wavelength (911nm), measurement range (7 or 12 km) and resolution (15 m). The instruments are fully operational and will be operated for 24 hours a day.
Infrared radiometers will be used to measure the cloud base temperatures. The wavelength range of the vertically pointing narrow beam infrared radiometer is 9.6-11.5 m m. The measurement range of the sky temperature is between +20 and -50oC. The instruments will be operated continuously. The observed cloud base temperatures have to be corrected for the atmospheric contribution. For this atmospheric correction the Modtran radiative transfer code will be used.
| Resp. Inst. | Station Nr | Instrumentation | ||
| UNIBE | Bern | 47.9 |
7.4 |
21.3/31.5 GHz microwave radiometer, infrared radiometer, sun photometer, GPS receiver, standard meteorological instrumentation |
| KNMI | Cabauw* | 52.2 |
5.2 |
22 channel microwave radiometer, 35 GHz cloud radar, lidar ceilometer, infrared radiometer |
| Chalmers | Chalmers | 57.7 |
12.0 |
21/31.4 GHz microwave radiometer, micro rain radar, standard meteorological instrumentation |
| CCLRC | Chilbolton | 51.4 |
-0. 5 |
22.2/28.8/93 GHz microwave radiometer, 3 and 94 GHz cloud radar, lidar ceilometer and standard meteorological instrumentation |
| CNRS | Gotland | 57.7 |
18.3 |
23.8/36.5 GHz microwave radiometer, lidar ceilometer |
| HUT | Helsinki | 60.3 |
25.0 |
18.7/23.8/36.5 GHz microwave radiometer |
| GKSS | Geesthacht | 53.4 |
10.3 |
21/35 GHz microwave radiometer, 95 GHz cloud radar, lidar ceilometer |
| Chalmers | Kiruna | 68.5 |
20.5 |
21/31.4 GHz microwave radiometer |
| DWD | Lindenberg | 52.2 |
14.2 |
microwave spectrometer, lidar ceilometer, standard meteorological instrumentation |
| CNRS | Paris | 48.8 |
2.3 |
21.3/ 23.8/31.7 GHz microwave radiometer |
| DWD | Potsdam | 52.4 |
13.1 |
23.8/31.4 GHz microwave radiometer, lidar ceilometer, infrared Fourier spectrometer |
| KNMI | St. Petersburg** | 59.8 |
30.5 |
13.5/22/37/89 GHz microwave radiometer, 3 and 9.4 GHz MRL-5 radar |
Table 1: List of stations particpating in the BALTIC Area Cloud Network
*Cabauw station will operate during CLIWA-NET campaign in April/May 2001. For the first campaign the 22 channel micro-wave radiometer will be stationed in Geesthacht.
**The Institute of Radioengineering and Electronics, Moscow make the data from the experimental site in St. Petersburg available to KNMI on a voluntary basis. The validation, processing, archiving and analysis is performed by KNMI.
Observational periods
It is proposed to co-ordinate the observations from 12 stations in the Baltic area. The selected sites are given in Table 1 together with an overview of the available instrumentation. It has to be stressed that all the instruments are already available, only some of the instruments will be located at a different station to get better coverage of the area of interest (see Figure 1). The network will be operated during 2 Enhanced Observation Periods of the BALTEX BRIDGE field experiment (EOP 1 in August/September 2000 and EOP 2 in April/May 2001). Both periods guarantee enough sunlight for the satellite retrieval and help to focus on the role of water clouds. These type of clouds (mid-latitude summer clouds) have the strongest impact on radiative [Li and Navon, 1998] and climate processes. The observed and derived parameters are summarised in Table 2.
Figure 1: Location of the ground-based stations participating in the BALTIC Area Cloud Network (see Table 1). The area for which the modelling studies within CLIWA-NET are performed incorporates the shown BALTEX modelling area. The purple line indicates the drainage basin area of the Baltic Sea.
Different microwave radiometers are taking part in the network. The instruments differ in viewing geometry, frequencies and calibration methods. To establish the accuracy of each microwave
radiometer type and thus the reliability of the network, it is important to have an intercomparison of these instruments. This intercomparison will take place at Cabauw, a location in the centre of the Netherlands. Cabauw will be equipped with a 35 GHz cloud radar, lidar ceilometer, wind profiler RASS, 200m high meteorological tower and the standard meteorological observations. Radio soundings are performed twice a day close to Cabauw (De Bilt). Cabauw is selected as an anchor station of the BALTEX/BRIDGE experiment. Within the Netherlands the Cloud Detection System (CDS) system will be operational during BALTEX BRIDGE. This system consists of a ground-based network and a processing environment for Meteosat and NOAA/AVHRR (MSG when available). At twelve stations, lidar ceilometers (range 14 km), pyranometers and (most likely) infrared radiometers will be installed. After the microwave radiometer intercomparison the microwave radiometers will be distributed over the CDS stations. So, apart from the network on the continental scale in the Baltic area, the network will also be operated on a regional scale (typically 100x100 km2). The higher density of the network stations gives detailed insight on the variability of clouds at these scales. Information on the sub-grid scale variability of GCM model grid boxes will be obtained. This also makes it possible to evaluate the output of high-resolution atmospheric models used for now-casting of clouds and precipitation in more detail. The BALTEX BRIDGE Cloud campaign will take place in August/September 2001. Please note that although the Netherlands is not in the Baltic area, it is well within the BALTEX modelling area (see Figure1).
Analysis of observations (MIUB, GKSS and KNMI)
For a fast availability of the retrieved LWP/LWC times series and precipitation fields a product definition phase before the first EOP will collect sample data from all stations/radars. Algorithms will be adapted for the different stations depending on their instrumentation. The accuracy of the retrieved products will be assessed (see list of measured and retrieved quantities for estimated values) and a data centre will be set up. This data centre will be accessible through internet. The analysis of this unique set of time series data yields important information on cloud processes. The radar derived precipitation and LWP are investigated regarding the LWP threshold when clouds start to rain, which is typically assumed to be 0.4 kgm-2. Additional information about the water budget comes from the integrated water vapour measured by the microwave radiometers. While ice clouds are transparent for the microwave radiometers the multi-parameter cloud radar measurements show significant signatures especially in the case of melting layers. The data on the vertical structure of the clouds will help to identify multi layered clouds. This is very important information for the analysis of the satellite data. Furthermore, the observed vertical profiles will contribute strongly to studies on cloud overlap, an important ingredient of NWP model cloud parameterisations.
Super cooled water layers can be detected in several ways. In the CLARA and CLARE’98 cloud campaigns, layers of super cooled water were detected by analysing the ratio of the lidar and radar backscatter signals. Because of the large wavelength differences of these instruments, the ratio of the signals is very sensitive to the particle size (which is much smaller for water droplets then for ice). A second method is based on combining LWP information from the microwave radiometers, altitude information from the radar/lidar and temperature information from infrared radiometer data or Numerical Weather Prediction (NWP) model output. Both methods will be applied.
Development of low-cost microwave radiometer (RPG, MIUB)
Based on the experience gained in this project and already available knowledge, RPG will design a low cost microwave radiometer. This design will be optimised for measuring LWP.
Measurements from the AVHRR (Advanced Very High Resolution Radiometer) and AMSU (Advanced Microwave Sounding UNIT) instruments on board the NOAA polar satellites will be used to obtain spatial distributions of cloud properties over the BALTEX area. The cloud parameters under consideration are: cloud presence, cloud cover fraction and cloud top temperature for both water and ice clouds. For water clouds more advanced products like optical thickness and infrared emissivity will be derived. The cloud property retrievals are well validated and described in literature.
The main focus of CLIWA-NET is on the more demanding retrieval of LWP. Over sea this parameter can be obtained from AMSU however over land measurements of this instrument can not be used for LWP retrieval because of the high surface emissivity and its variability. A number of approaches can be found in literature, but up to now results are of varying quality and the number of validation experiments is limited. The second part of WP3000 develops a retrieval method from a combination of AVHRR measurements and collocated ground-based microwave radiometer measurements. This method will be used for retrievals over land and, together with the AMSU results over the sea will yield a complete spatial distribution of vertically integrated liquid water.
The AMSU and microwave radiometer are dedicated to the retrieval of liquid water columns, but have limited applicability. The AMSU can only measure over sea, while the microwave radiometers measure at individual locations, continuous in time. In order to obtain a spatial distribution of vertical integrated liquid water over land the AVHRR measurements will be used. The basic approach to the retrieval of liquid water from AVHRR was published at the end of the 1980’s by Kriebel et al. . Nevertheless, the retrieval method is not well validated and the results have been of varying quality, due to the numerous assumptions made in the retrieval.
In order to obtain a reliable spatial distribution of vertically integrated liquid water over land, the microwave radiometer network data will be used. From each microwave radiometer the time series will be collocated with an area in the AVHRR derived LWP field. Spatial pattern recognition and time series analyses techniques will be used to link the observations from the different sources. Transfer functions will be derived, which integrate the surface based measurement and the satellite derived products.
Techniques to optimise the AVHRR retrieval will be further developed on the basis of results from the BBC campaign and the CLARA measurement campaign, performed in 1996. During the intensive observational periods additional information on cloud properties will be available for analysis. Please note that the AVHRR retrievals of LWP are only possible at daytime, because the parameters are derived from reflected sunlight. The observed and derived parameters are summarised in Table 2.
The methods developed during this project may be applied to other satellites then AVHRR, e.g. Severi on board of the Meteosat Second Generation platform. The Severi instrument contains all AVHRR channels. This enables the retrieval of liquid water column fields every 15 minutes for a very large area., The development of this product would take a considerable research effort and the timely availability of the MSG data is uncertain. However, a feasibility study will be developed to transfer the schemes developed for the integration of AMSU, AVHRR, and ground-based measurements for LWP to the MSG, METOP and CLOUDS missions.
Th AMSU analysis and archiving is the responisbility of IFM. The AVHRR archiving and processing takes place at SMHI. The integration of the satellite observations and data from the ground stations is a shared research topic of IFM and KNMI.
Accuracy |
Temporal resolution |
Vertical resolution |
Horizontal resolution |
|
| Measured Quantities | ||||
| Microwave brightness temperature | 1-2 K |
1 s - 1 min |
- |
- |
| Cloud base height | 30 m |
30 s |
30 m |
- |
| Cloud base temperature | 1-2 K |
1 s – 1 min |
- |
- |
| Cloud radar reflectivity | 2 dBZ |
~ 30 s |
~ 50 m |
- |
| Cloud radar depolarisation ratio | 1 dB |
~ 30 s |
~ 50 m |
- |
| Cloud radar doppler velocity | 0.2 m/s |
~ 30 s |
~ 50 m |
- |
| Precipitation radar reflectivity | 0.5 dBZ |
15 min |
- |
2 km |
| Temperature at ground level | 1 K |
1 min |
- |
- |
| Pressure at ground level | 1 hPa |
1 min |
- |
- |
| Relative humidity at the ground | 5 % |
1 min |
- |
- |
| Derived Quantities (single points) | ||||
| Integrated liquid water (LWP) | 10–30 gm-2 |
1 s – 1 min |
- |
- |
| Integrated water vapour | 1 kgm-2 |
1 s – 1 min |
- |
- |
| Liquid water profiles | ~ 50 % |
1 min |
~ 50 m |
- |
| Ice water profiles | 30 –70 % |
1 min |
~ 50 m |
- |
| Temperature profile | 2 K |
1 min |
0.2–1 km |
- |
| Humidity profile | 1 gm-3 |
1 min |
0.2–1 km |
- |
| Derived Quantities (fields) | ||||
| Precipitation fields | - |
3 h sum |
- |
2 km |
| LWP fields | 30% |
4 /day |
- |
2 km (max) |
| Cloud cover fraction | 0.15 |
8/day |
2 km (max) |
|
| Cloud top temperature | 2 K |
8/day |
2 km (max) |
Table 2: List of measured and derived quantities
One of the aims of the current proposal is to use the retrieved data-sets from both the continental and the regional observational campaigns in an objective evaluation of the performance of present-day state-of-the-art cloud parameterisation schemes with a focus on liquid water path (LWP) and vertical structure of cloud amount and cloud water. The use of high-quality data-sets with sufficient temporal and spatial resolution is an absolute requirement in order to put constraints in the formulation of cloud processes and their interaction with other model components. The model component of CLIWA-NET pursues three lines of research involving atmospheric models:
To achieve the goal of item 1) the following strategies are considered:
Item 2) will be investigated with a non-hydrostatic atmospheric model including data-assimilation. Focus will be on the sensitivity of model cloud parameters to the employed horizontal grid spacing from 10 km down to 1 km. Over this range of horizontal resolutions, contributions from resolved processes are expected to become overlapping, at least partly, with contributions from processes which are normally considered "sub-grid scale".
With item 3) the following parameterisation issues will be considered, which are expected to be verifiable with the data-sets inferred from the observational periods within the CLIWA-NET project:
Evidently, the impact of any change in parameterisation can straightforwardly be evaluated with the remaining observations of the CLIWA network, which part can be viewed as an independent data-set.
4.1 Overview of the
Consortium
The consortium consists of 12 (assistant-) contractors and two sub-contractors. This large number of groups can be divided in different groups:
A) Five institutes/universities (KNMI, MIUB, SMHI, IFM and GKSS) play a leading role in the organisation of the project, the archiving and analysis of the data and/or modelling activities. One commercial company (RPG) will perform a design study of an optimal but low-cost micro wave radiometer instrument.
B) Six institutes/universities (Chalmers, HUT, DWD, CNRS, UNIBE, CCLRC) contribute mainly through providing the observations. They all own advanced remote sensing instruments (microwave radiometer, cloud radar, lidar and/or infrared radiometers). In general their participation in the project is limited and the majority of them participates as an assistant contractor.
C) Two sub-contractors (ECMWF and Moscow) will participate within the project. Moscow is a sub-contractor of MIUB. The Russian group performs the necessary measurements in the most Eastern part of the Baltic Sea.
ECMWF is a sub-contractor of KNMI and will actively participate in the evaluation/improvement of cloud parameterisations in atmospheric models (WP4000). The modelling division of the DWD will participate in a similar role in this workpackage.
KNMI will act as co-ordinator of the project.
4.2 Description of the
Participants and third-party assistance
(includes email
addresses of contact persons)
1). Royal Netherlands Meteorological Institute, KNMI
European Centre for Medium-range Weather Forecasts, ECMWF (sub-contractor)
2). Meteorological Institute University of Bonn, MIUB
3). Centre des Environnements Terrestre et Planetaires, CNRS
4). Helsinki University of Technology, HUT
5). Radiometer Physics GmbH, RPG
6). Deutsche Wetter Dienst, DWD
7). Chalmers University of Technology, Chalmers
8). Institute of Applied Physics, University of Bern, UNIBE
9). Rossby Center and Swedish Hydrological and Meteorological Institute, SMHI
10). Institut fur Meersekunde Kiel, IFM
11). GKSS-Research Center Geesthacht, GKSS
12). Rutherford Appleton Laboratory, CCLRC
Important dates on the CLIWA-NET calendar:
Date |
Event |
| 1 March 2000 | Start of the CLIWA-NET project |
| 5 - 7 April 2000 | Kick-off workshop |
| Aug./Sept. 2000 | EOP1 |
| Jan. 2001 | EOP1 workshop |
| April/May 2001 | EOP2 |
| Oct. 2001 | EOP2 workshop |
| Aug./Sept. 2001 | BBC campaign |
| April 2002 | BBC workshop |
| March 2003 | Final workshop |
| 31 March 2003 | End of the CLIWA-NET project |
The data policy of the CLIWA-NET project can be found here (as a PDF
document).
7.
User's requirement document
The user's requirement document of the CLIWA-NET project can be found here (as a PDF document).
| (c) Copyright 2000, Last Updated: 10 April 2001 | Back to homepage |