Thirty years after satellite data were first used for assessing rainfall it is appropriate to critically review what has been achieved in this field to date, and what has not. Good progress has been made in developing and understanding the capabilities of satellite sensing systems, the physics of the atmosphere and land surfaces of the Earth, and in the development of computers capable of multiple, high-speed operations for near real-time data processing. However, all types of satellite monitoring of falling precipitation are subject to limitations, and progress is still relatively poor in several regards. To make matters worse, there has been a global trend towards the collection of less, not more, in situ data on precipitation over at least half of the last century. Foci for further research and operations are proposed. Some problems will require concentrated effort and significant expenditure if the recent rate of progress in this field is even to be maintained.

The use of Special Sensor Microwave Imager (SSM/I) data for snow cover detection has been well documented since the launch of the first Defence Meteorological Satellite Program (DMSP) platform in 1987. One of the major problems yet to be resolved is the successful discrimination and subsequent removal of precipitation areas from snow cover estimates at high latitudes: this can have a significant impact on the ability to detect snow cover. Both snow cover and precipitation can exhibit similar responses at SSM/I microwave frequencies. The majority of snow cover algorithms eliminate precipitation using a single brightness temperature threshold at either 19 or 22 GHz and perform adequately under most conditions. It has been observed that this threshold must be varied under given surface and atmospheric conditions by as much as 5-10 K. This can result in large errors in both snow cover and precipitation estimates when climatically aggregated, but is also evident in some individual case study events. By using additional thermal infrared (IR) data from the DMSP Operational Line Scanner (OLS) a synergistic approach has been applied. In theory a snow cover will have a much warmer OLS IR temperature than the type of precipitating clouds that give a similar response at microwave frequencies to the snow cover. The IR data can be used to identify more accurately the snow cover. The OLS has the advantage also of an improved spatial resolution over the SSM/I, and a synergistic approach will not deteriorate the spatial resolution of the SSM/I estimate. A synergistic algorithm has been developed and tested over three case study areas and demonstrates a qualitative improvement in the detection of snow cover under difficult conditions.

Conventional measurements of rainfall over the ocean are fraught with physical and practical problems and instrument difficulties, leading to spatial and temporal inconsistencies in climate datasets. Estimates of rainfall based on the scattering and emission characteristics of passive microwave radiation, measured by instruments on board orbiting satellites, substantially improve the knowledge of rainfall patterns and processes over open water. A new passive microwave satellite-based Atlas of Rainfall and Wind Speed over the Eastern North Atlantic and North Sea has been prepared, including rainfall totals (mm), rates (mm h−1), and percent occurrences (%) for the period 1979–1996 inclusive. Annual, seasonal and monthly maps have been generated, as well as histogram and time series products. In this paper, the main rainfall patterns over this region are summarised and interpreted in comparison with several other existing datasets. The interpretation of passive microwave-derived wind speed over the eastern North Atlantic and North Sea shall be reported in a future paper. Upon the analysis of above and below normal rainfall periods, significant relationships are revealed with respect to the mean sea level pressure, air and sea surface temperature, and atmospheric precipitable water. It is shown that the location and strength of the belt of prevailing Westerlies, and their interaction with major land masses, dictate the overall rainfall distribution. Results show that rainfall over this area is well-related to various indices of the general circulation. Also, patterns of rainfall over the North Sea are significantly related to the direction and strength of the wind. Finally, it is shown that there is a significant linear relationship between climate station rainfall and the estimate over the nearest open water pixel, although the percent of explained variance seldom exceeds 50%. Copyright © 1999 Royal Meteorological Society

A continuously calibrated infrared (IR) geostationary satellite rainfall estimation technique (CCB4) is introduced, in the context of the Nile River Forecast System, an operational system for hydrological modelling and forecasting. The CCB4 incorporates near-real-time rain gauge data to continuously calibrate optimum IR rain/no-rain thresholds and daily, rain rates on a daily time step. The ability of the CCB4 and two comparative techniques to estimate daily rainfall at the regional and pixel scales is assessed using Meteosat IR imagery, and gauge data from six wet season months covering three years. The CCB4 shows improved skill in identifying rain days and estimating daily rain amounts at a range of spatial scales, from regional to pixel scales. At the pixel scale, however, improved root mean square errors remain relatively high, ranging between 66% and 84% of the mean unconditional rain rate.

The second WetNet Precipitation Intercomparison Project (PIP-2) evaluates the performance of 20 satellite precipitation retrieval algorithms, implemented for application with Special Sensor Microwave/Imager (SSM/I) passive microwave (PMW) measurements and run for a set of rainfall case studies at full resolution-instantaneous space-timescales, The cases are drawn from over the globe during all seasons, for a period of 7 yr, over a 60 degrees N-17 degrees S latitude range. Ground-based data were used for the intercomparisons, principally based on radar measurements but also including rain gauge measurements. The goals of PIP-2 are I) to improve performance and accuracy of different SSM/I algorithms at full resolution-instantaneous scales by seeking a better understanding of the relationship between microphysical signatures in the PMW measurements and physical laws employed in the algorithms; 2) to evaluate the pros and cons of individual algorithms and their subsystems in order to seek optimal "front-end" combined algorithms; and 3) to demonstrate that PMW algorithms generate acceptable instantaneous rain estimates.It is found that the bias uncertainty of many current PMW algorithms is on the order of +/-30%. This level is below that of the radar and rain gauge data specially collected for the study, so that it is nor possible to objectively select a best algorithm based on the ground data validation approach. By decomposing the intercomparisons into effects due to rain detection (screening) and effects due to brightness temperature-rain rate conversion, differences among the algorithms are partitioned by rain area and rain intensity: For ocean, the screening differences mainly affect the light rain rates, which do not contribute significantly to area-averaged rain rates. The major sources of differences in mean rain rates between individual algorithms stem from differences in how intense rain rates are calculated and the maximum rain rate allowed by a given algorithm. The general method of solution is not necessarily the determining factor in creating systematic rain-rate differences among groups of algorithms, as we find that the severity of the screen is the dominant factor in producing systematic group differences among land algorithms, while the input channel selection is the dominant factor in producing systematic group differences among ocean algorithms. The significance of these issues are examined through what is called "fan map" analysis.The paper concludes with a discussion on the role of intercomparison projects in seeking improvements to algorithms, and a suggestion on why moving beyond the "ground truth" validation approach by use of a calibration-quality forward model would be a step forward in seeking objective evaluation of individual algorithm performance and optimal algorithm design.

Historically, rainfall has been measured by in situ devices, principally by raingauges, and since World War II, by rainfall radar. In the 1970s and 1980s growing interest was paid to the potential of satellite data, particularly infrared and passive microwave imagery, for such purposes. Today, informed attention is focusing increasingly upon the advantages to be gained from combinations of data from as many different sources as possible. After a general introduction, this paper describes and explains a rainfall monitoring method development strategy initiated by the Centre for Remote Sensing, Bristol, UK in 1991, involving climate data sets, synoptic weather reports, satellite infrared and passive microwave data, and other GIS database information. Results have been obtained from several major areas of the humid tropics over periods of up to five years, leading to much improved understanding of the cloud-rainfall relationships and the estimation models themselves. The Bristol methodology has been designed to be operable in real time, flexible in its response to end-user product requirements, and widely applicable through its use of data sets available on the Internet. The paper concludes with a summary of further methodological and related research and development planned for this on-going project, elements of which are already in operational use in support of some humid tropical regions.

A new empirical algorithm for retrieving rainfall rates from passive microwave (particularly Special Sensor Microwave/Imager) data is presented. Errors caused by spatial and temporal variation of surface temperature, emissivity, and atmospheric effects are minimized by modeling the nonraining brightness temperatures within 0.5 degrees latitude by 0.5 degrees longitude regions based on the statistics of the satellite data from the whole of the month prior to the date of interest. Displacement of data away from the modeled relationship by more than a threshold value, calculated from the standard deviation of the nonraining data, is detected as rainfall. The algorithm is able to detect rainfall over land, sea, and coasts. An initial calibration and validation is performed using data from WetNet Precipitation Intercomparison Project 2 over the British Isles. To overcome collocation errors during calibration and validation, techniques are presented for matching radar rain pixels with appropriate satellite data. The algorithm detects light rainfall less than 0.5 mm h(-1), which is common at midlatitudes, and exhibits a large dynamic range suitable for measuring heavy rainfall. The critical success index ranges from 50% over land to 61% over sea. The algorithm is not limited geographically, although it is likely that the rain rate relationship would benefit from recalibration for other regions or from the inclusion of a physical inversion technique designed to retrieve the vertical structure of the precipitation.

This paper reviews the basis of passive microwave algorithms that derive rainfall rates directly from relationships between brightness temperatures and rainfall rates established by statistical relationships and empirical calibration. The performance of these algorithms and their present and future roles are assessed in comparison with the increasing number of modeling techniques used for passive microwave rainfall retrievals.

Some issues in the field of remote sensing application in hydrology are discussed in this editorial paper based on several contributions presented at Session HS11 ‘Remote Sensing in Hydrology’ during the XXI General Assembly of EGS in The Hague. These include downscaling problems, reliability estimates, and the assessment of the uncertainty associated with the retrieval of hydrological variables at various scales of concern. Calibration and validation issues are among the most pressing problems, for a major difficulty confronting many hydrological applications is still that of providing the available sophisticated models with very detailed initial and boundary conditions, from extensive high resolution monitoring campaigns.

Studies have established that the common causes of disasters in the Mediterranean region are high intensity rains, and associated flash floods, landslides and mud flows. A project code named ''STORM' (Satellite Tracking and Observation of Rainfall Run-off Monitoring), funded by the European Commission under the 3rd framework of its Environment and Climate Programme was undertaken from 1992 through 1994 by eight laboratories in the UK, Italy and Spain. The aim of this Project was to improve the identification, monitoring and local forecast of high-intensity rainfall events in the northern coastal zones of the Mediterranean Sea, based on data from in situ and remote sensing sources, plus synoptic data and analyses. The present paper describes one of the models developed for the analysis, key data types and their co-ordination within an overall monitoring scheme. Once moderate to heavy precipitation areas have been identified and evaluated, potentially hazardous rain areas can be analysed, tracked and forecast using data from synoptic weather analyses, weather prediction models and radar (if available), taking account of the terrain characteristics over which the rain areas are moving and/or likely to move in the near-term future. The paper concludes with suggestions for ways in which such methodology could be further developed to assess high intensity rainfall-related risks, and so help alleviate associated disasters, more broadly across the Mediterranean region, or neighbouring areas such as western Europe.

AbstractGround‐based multi‐parameter radar is used to validate and evaluate satellite passive microwave parameters related to rainfall over land. Light precipitation over the southern part of England is considered. The issues of diverse resolutions of radar and satellite data in space and time are investigated, as well as the satellite beam‐filling problem, in relation to present passive microwave satellite algorithms related to rainfall over land, and their perceived behaviour. Several statistical analyses are applied and their sensitivities to data integration problems are quantified. Passive microwave parameters are evaluated and compared with each other. It is concluded that for radar passive microwave satellite data intercomparisons, calibrations and validations the time resolution seems to be more important than the spatial resolution, and high‐resolution radar can be used operationally to tackle the satellite beam‐filling problem. It is found that several passive microwave algorithms have performances very similar to each other. Copyright © 1997 Royal Meteorological Society