Search

Satellite-based precipitation estimates (SPEs) are generally validated using ground-based rain gauge or radar observations. However, in poorly instrumented regions, uncertainty in these references can lead to biased assessments of SPE accuracy. As a result, at regional or continental scales, an objective basis to evaluate SPEs is currently lacking. Here, we evaluate the potential for large-scale, spatially continuous evaluation of SPEs over land via the application of collocation-based techniques [i.e., triple collocation (TC) and quadruple collocation (QC) analyses]. Our collocation approach leverages the Soil Moisture to Rain (SM2RAIN) rainfall product, derived from the time series analysis of satellite-based soil moisture retrievals, in combination with independent rainfall datasets acquired from ground observations and climate reanalysis to validate four years of the European Organisation for the Exploitation of Meteorological Satellites (EUMETSAT) Satellite Application Facility on Support to Operational Hydrology and Water Management (H-SAF) H23 daily rainfall product. Large-scale maps of the H23 correlation metric are generated using both TC and QC analyses. Results demonstrate that the SM2RAIN product is a uniquely valuable independent product for collocation analyses, because other available large-scale rainfall datasets are often based on overlapping data sources and algorithms. In particular, the availability of SM2RAIN facilitates the large-scale evaluation of SPE products like H23—even in areas that lack adequate ground-based observations to apply traditional validation approaches.

Heavy precipitation systems that are typical in the Mediterranean area and often devastate the coastal regions, are described and analyzed here by exploiting active and passive microwave measurements and state-of-the-art precipitation products available in the Global Precipitation Measurement (GPM) mission era. The GPM is boosting its key role in integrating the established observational ground-based and satellite-borne tools not only for precipitation monitoring, but also for understanding and characterizing severe weather in the Mediterranean area. In this Chapter, we present three events that have recently challenged observational and forecasting capabilities, and caused damages at the ground. Making use of ground based and satellite-borne instruments, we address the problem of estimating precipitation of a small-scale and short-living intense thunderstorm, the capability to render the 3-D structure of a mesoscale organized convective system, and the key role of spaceborne microwave sensors in the characterization and monitoring of a tropical-like cyclone. To this end, we exploited satellite measurements probably beyond the role they have been designed for, showing few strategies to blend satellite data and products with conventional meteorological data, with the aim to increase the knowledge of severe systems in the Mediterranean area and to support operational forecasting activities in a climate change perspective.

Difficulties in the representation of rainfall variability using ground-based sensors over mountainous areas necessitate the use of high-resolution satellite precipitation data sets from combined passive microwave (PMW) and geostationary infrared observations. The accuracy of these data sets depends greatly on the uncertainty characteristics of the PMW sensor retrievals and sampling frequency. In this paper, we evaluate the retrieval error characteristics from different PMW sensors over mountainous terrain using high temporal and spatial resolution ground-based radar (GR) reference rainfall data sets from two field experiments: one in the mid-Atlantic East Coast of the United States and the second in the Northeastern Italian Alps. We extracted matchups of PMW/GR rainfall based on a matching methodology that identifies GR volume scans coincident with PMW overpasses, and scales GR parameters to the satellite products’ nominal spatial resolution. Different PMW precipitation retrieval algorithms are evaluated for four PMW sensors. The error analysis shows varying error characteristics across different PMW retrievals. Moreover, the magnitude-dependent systematic error, going from overestimation or weak underestimation of light precipitation to mainly underestimation of heavier precipitation, shows weak covariation with the ground reference. Detection capabilities for all sensors increase markedly with an increasing reference rate and convective occurrence because of the more pronounced ice-scattering signal associated with deeper convection and high rain rates.

A new precipitation retrieval algorithm for the AMSR2 is described. The algorithm is based on the cloud dynamics and radiation database (CDRD) Bayesian approach and represents an evolution of the previous version applied to SSMIS observations, and used operationally within the EUMETSAT H-SAF program. This new product presents as main innovation the use of a very large database entirely empirical, derived from coincident radar and radiometer observations from the NASA/JAXA GPM-CO launched on February 28, 2014. The other new aspects are: 1) a new rain-/no-rain screening approach; 2) use of EOF and CCA for dimensionality reduction; 3) use of new ancillary variables to categorize the database and mitigate the problem of non-uniqueness of the retrieval solution; and 4) development and implementations of modules for computation time minimization. A verification study for case studies over Italy and for coincident AMSR2/GPM-CO observations over the MSG full disk area has been carried out. Results show remarkable AMSR2 capabilities for RR retrieval over ocean (for RR > 0.1 mm/h), good capabilities over vegetated land (for RR > 1 mm/h), while for coastal areas the results are less certain. Comparisons with NASA GPM products, and with ground-based radar data, show that the new CDRD for AMSR2 is able to depict very well the areas of high precipitation over all surface types. The algorithm is also able to handle an extremely large observational database available from GPM-CO and to provide rainfall estimate with minimum latency, making it suitable for NRT hydrological and operational applications.

The accurate estimation of rainfall from remote sensing is of paramount importance for many applications as, for instance, the mitigation of natural hazards like floods, droughts, and landslides. Traditionally, microwave observations in the frequency between 10 and 183 GHz are used for estimating rainfall based on the direct interaction of radiation with the hydrometeors within precipitating clouds in a so-called top-down approach. Recently, a bottom-up approach was proposed that uses satellite soil moisture products derived from microwave observations (

In Mediterranean area (MA), floods are, after droughts, the most dangerous meteorological hazard causing severe damages of manmade infrastructures, produce large loss of lives, and adversely affect the opportunities for socio-economic development of MA countries. MA is characterized by a relatively high number of flash floods (FFs) and floods occurring in small catchments normally triggered by deep convective systems associated with intense precipitation with a limited areal extent and high initial wetness conditions. For these reasons, the understating of the role of rainfall and soil moisture as well as the techniques available for their monitoring is particularly important. In this chapter, we review the most important studies carried out on floods in MA, which focuses on (1) the importance of the rainfall and soil moisture in the triggering and evolution of these events; (2) the monitoring techniques and models available to estimate these two variables along with their operational use for their prediction.

The project aimed at quantifying the impact of Precipitation Measurement Biases (PMBs) on the modelling of hydrological processes at different catchment scales. Both local and areal methods for estimating precipitation rates were addressed, as well as instrumental and environmental sources of inaccuracy in measurements.

This work proposes a multi-parameter method for the detection of cloud-to-ground stroke rate (SRCG) associated to convective cells, based on the measurements of a low-cost single-polarization X-band weather radar. To train and test our procedure, we built up a multi-year dataset, collecting 1575 radar reflectivity volumes that were acquired in the pilot study area of Naples metropolitan environment matched with the LIghtning NETwork (LINET) strokes and meteorological in-situ data. Three radar-based variables are extracted simultaneously for each rain cell and properly merged together, using “ad hoc„ classification methods, to produce an estimation of the expected lightning activity for each rain cell. These variables, proxies of mixed-phase particles and ice amount into a convective cell, are combined into a single label to cluster the SRCG into two categories: SRCG = 0 (no production of strokes) or SRCG > 0 (stroke production), respectively. Overall, the main results are comparable with those that were obtained from more advanced radar systems, showing a Critical Success Index of 0.53, an Equitable Threat Score of 0.34, a Frequency Bias Index of 1.00, a Heidke Skill Score of 0.42, a Hanssen-Kuiper Skill Score of 0.42, and an area under the curve of probability of detection as a function of false alarm rate (usually referred as ROC curve) equal to 0.78. The developed technique, although with some limitations, outperforms those based on the use of single stroke proxy parameters.

The assessment of the actual observational capabilities of snowfall by spaceborne microwave radiometers is crucial to develop and improve precipitation retrieval algorithms. Exploiting coincident spaceborne active and passive microwave sensor datasets can effectively enhance our understanding of high-frequency microwave channels sensitivity to snowfall. This study illustrates the results of the analysis of matched Global Precipitation Measurement (GPM) Microwave Imager (GMI) and CloudSat Cloud Profiling Radar (CPR) snowfall observations (mainly found at latitudes between 55° and 65°) providing insights on GMI multi-frequency signals associated with different snowfall types. These findings are used to develop a new algorithm to retrieve snow water path associated to surface snowfall from GMI multichannel measurements.

The uncertainties associated with rainfall estimates comprise various measurement scales: from rain gauges and ground-based radars to the satellite rainfall retrievals. The quality of satellite rainfall products has improved significantly in recent decades, however, such algorithms require validation studies using observational rainfall data. For this reason, this study aims to apply the H-SAF consolidated radar data processing to the X-band radar used in the CHUVA campaigns and apply the well established H-SAF validation procedure to these data and verify the quality of EUMETSAT H-SAF operational passive microwave precipitation products in two regions of Brazil (Vale do Para&iacute, ba and Manaus). These products are based on two rainfall retrieval algorithms: the physically based Bayesian Cloud Dynamics and Radiation Database (CDRD algorithm) for SSMI/S sensors and the Passive microwave Neural network Precipitation Retrieval algorithm (PNPR) for cross-track scanning radiometers (AMSU-A/AMSU-B/MHS sensors) and for the ATMS sensor. These algorithms, optimized for Europe, Africa and the Southern Atlantic region, provide estimates for the MSG full disk area. Firstly, the radar data was treated with an overall quality index which includes corrections for different error sources like ground clutter, range distance, rain-induced attenuation, among others. Different polarimetric and non-polarimetric QPE algorithms have been tested and the Vulpiani algorithm (hereafter, R q 2 V u 15 ) presents the best precipitation retrievals when compared with independent rain gauges. Regarding the results from satellite-based algorithms, generally, all rainfall retrievals tend to detect a larger precipitation area than the ground-based radar and overestimate intense rain rates for the Manaus region. Such behavior is related to the fact that the environmental and meteorological conditions of the Amazon region are not well represented in the algorithms. Differently, for the Vale do Para&iacute, ba region, the precipitation patterns were well detected and the estimates are in accordance with the reference as indicated by the low mean bias values.

The NASA/JAXA Global Precipitation Measurement (GPM) Core Observatory (CO) carries, for the first time, a dual-frequency Precipitation Radar (DPR). The DPR provides insights into the three-dimensional structure of precipitating clouds and intensity and is currently considered the main calibration instrument for radiometer derived global precipitation estimates. The DPR employs three scanning modes: Matched Scan (MS), Normal Scan (NS) and High Sensitivity scan (HS). The Ka- and Ku-band radar (MS) footprints of the inner swath consist of 25 angle bins with a vertical resolution of 125 m while the Ku-band radar (NS) covers the full swath with 49 angle bins with the same vertical resolution. The HS Ka-band radar footprints are interlaced with matched Ku/Ka-band footprints and consist of 24 angle bins with range sampling at 250 m. The swath widths of Ka- and Ku-band radars are 120 km and 245 km, respectively, while the diameter of both Ka- and Ku-band footprints at nadir is 5 km about. While the precipitation retrieval algorithms of the Ka- and Ku-band radar Matched (MS) footprints provide a double-frequency based output, the Ku-band Normal Scan (NS) and Ka-band High Sensitivity scan (HS) precipitation estimates for same footprints, are single-frequency based. The goal of the present works is to carry out an intercomparison of the single- and double-frequency DPR algorithms by considering the respective outputs over the Mediterranean area during rain events. A number of DPR related variables have been considered (i.e. measured and corrected reflectivity, surface rainfall rate, mean mass diameter, normalized intercept parameter, liquid and ice water content vertical profiles) to perform the analysis. The data have been categorized for surface type (land and sea) and precipitation type (stratiform and convective). The results show the differences in the performance of the single- and double-frequency algorithms as function of the surface and precipitation type. The vertical profiles of the single- and double-frequency based retrieved parameters are compared in order to assess their variability along the precipitating column.

Passive microwave sensors with high-frequency channel diversity offer a unique opportunity for significant ad- vances in surface snowfall retrieval. However, surface snowfall rate quantification using spaceborne microwave radiometers remains a challenging task due to the complex microphysics of snow clouds, and to the many sources of signal contamination (e.g., frozen background, presence of supercooled droplets). Recently, Panegrossi et al. (2017) highlighted the impact of falling snow on the high frequencies channels of GPM Microwave Imager (GMI). In particular: 1) they identified the environmental conditions under which the snow detection could be optimally achieved (e.g., atmospheric moisture, sea ice coverage); 2) they analyzed the sensitivity of GMI high-frequency channels and of the polarization signal at 166 GHz to falling snow as identified by CloudSat Cloud Profiling Radar (CPR); 3) they analyzed the impact of supercooled droplets as identified by CALIPSO/CPR observations on the GMI signal. Building on these results we developed a brand new algorithm to retrieve snowfall, based on the exploitation of the global GMI/CloudSat snowfall coincidence dataset, mainly occurring at latitudes between 55 and 65oN/S. This algorithm makes use of all GMI channels brightness temperature measurements and of ancillary environmental variables (atmospheric water vapor content and 2-meter temperature, ground classification). The retrieval is achieved in three steps and uses Cloudsat 2C-SNOW-PROFILE product as a reference. A first module detects snowfall events in the field of view of GMI high-frequency channels. Then, a second module is devoted to the detection of supercooled droplets, since its presence can strongly affect the GMI channels response to snow- fall. Finally, a third module provides the snowfall rate quantification. The new GMI snowfall retrieval algorithm is presented, and results from case studies and from global applications are shown.

The Italian radar network is currently composed by 20 C-band and 2 X-band systems, managed by 11 administra- tions, able to cover most of the country. The product generation at national level is carried out by the Department of Civil Protection (DPC) that currently manages 7 C-band and 2 X-band systems, all with dual-polarization capa- bility. The national mosaic provides an estimation of the precipitation intensity with a five minutes time resolution over a equi-spaced 1 km grid. A Quality Index (QI) map is associated to the rainfall rate map, taking into account the radar artifacts, the partial beam blockage, the height of measurements with respect to the freezing layer, the beam broadening and the rain path attenuation. The geostationary Meteosat Second Generation (MSG) Spinning Enhanced Visible and Infrared Imager (SEVIRI) provides high-quality measurements across twelve visible (VIS) and infrared spectral (IR) bands. Several tech- niques have been developed exploiting VIS or IR observations (individually or in a combined way) to relate the cloud features to the surface precipitation. One of them is the blending Rapid Update (RU) technique combining IR observations with precipitation estimates obtained from spaceborne passive microwave radiometers, and used operationally within the EUMETSAT H SAF program. Within the collaborative agreement between DPC and the Institute of Atmospheric Sciences and Climate (ISAC) of the National Research Council (CNR) a new blended algorithm has been developed that combines the ground radar precipitation estimates and the IR observations from SEVIRI to provide a precipitation map over the whole Italian territory every 15 minutes, at 1 km spatial resolution. The goal is to integrate the low QI ground radar mea- surements with the precipitation estimates inferred from IR observations. The IR precipitation estimates are based on regression curves built from a training dataset based on coincident IR and high QI radar observations. The present work shows the preliminary results of the blended algorithm. The validation is carried out through a comparative study with the Italian rain gauges network.

Precipitation is an essential element of the global hydrological and energy cycles, and its measurements are of great importance in a variety of research areas, such as climate studies, management of water resources, natural hazards, and hydrology. Despite the measurement of the precipitation on a global scale is therefore crucial in many aspects of economic and social life on Earth, yet it still has many problems to overcome to meet the demands of hydrolog- ical and climate research, and of operational applications. An important step forward towards the improvement of global precipitation monitoring has been achieved with the advent of the Global Precipitation Measurement (GPM) mission thanks to the availability of the NASA/JAXA GPM Core Observatory (GPM-CO) (equipped with the GPM Microwave Imager (GMI) and the Dual frequency (Ku and Ka) Precipitation Radar (DPR)), expected to provide ac- curate precipitation estimates, and contributing to the constellation of pre-existing and future radiometers to ensure 3-hourly global coverage between 65oS and 65oN. In this direction, the EUMETSAT program H-SAF (Satellite Application Facility on Support to Operational Hydrology and Water Management) designed to deliver satellite products for hydrological applications (precipitation, soil moisture and snow parameters) to research and opera- tion users worldwide, is being carried out in Europe.H-SAF precipitation products are based on the exploitation of all overpasses of present and future Low Earth Orbit (LEO) satellites carrying cross-track and conically scan- ning passive microwave radiometers orbiting around the globe. Within this framework we have developed a new algorithm, based on the Passive microwave Neural network Precipitation Retrieval (PNPR) approach, designed to work with the conically scanning GMI radiometer. As opposed to the PNPR algorithm originally developed for cross-track scanning radiometers (AMSU/MHS and ATMS), and based on a training database obtained from cloud resolving model simulations coupled to a radiative transfer model, the new PNPR for GMI is based on ob- servational databases. The purpose was to develop a Neural Network (NN) based algorithm, able to handle the extremely large and rich observational datasets available from combined GPM-CO observations and to exploit the advanced observational capabilities of the GMI both in terms of channel assortment, and spatial resolution. The PNPR-GMI algorithm uses a single NN to retrieve the surface precipitation over different background surfaces of the GPM global coverage area. For this purpose the algorithm uses GMI TB derived variables in a new Principal Component Analysis (PCA) approach to detect the precipitation signal over different backgrounds reducing the number of inputs to the NN. Ancillary meteorological parameters (including ECMWF Total Precipitable Water (TPW) and 2m temperature) are also used as input. The details concerning the design and development of the NN, and the algorithm input selection procedure will be presented. Results of the algorithm performance for some case studies of interest over the GPM coverage area will be shown. Furthermore results from an extensive comparison of PNPR-GMI with the GMI Goddard profiling algorithm (GPROF) performance will be also presented.

Precipitation retrievals exploiting the available passive microwave (PMW) observations by cross-track and conically scanning satellite-borne radiometers in the Global Precipitation Measurement (GPM) mission era are used to monitor and characterize heavy precipitation events that occurred during the Fall 2014 in Italy. Different physically based PMW precipitation retrieval algorithms are used: the Cloud Dynamics and Radiation Database (CDRD) and Passive microwave Neural network Precipitation Retrieval (PNPR), used operationally in the European Organisation for the Exploitation of Meteorological Satellites (EUMETSAT) Satellite Application Facility on support to Operational Hydrology and Water Management (H-SAF), and the National Aeronautics and Space Administration (NASA) Goddard PROFiling algorithm (GPROF). Results show that PMW precipitation retrievals from theGPMconstellation of radiometers provide a reliable and quantitative description of the precipitation (instantaneous and on the daily scale) throughout the evolution of the precipitation systems in the Mediterranean region. The comparable relative errors among gauges, radar, and combination of radiometer overpasses legitimize the use of PMW estimates as a valuable and independent tool for monitoring precipitation. The pixel-based comparison with dual-polarization radars and raingauges indicates the ability of the different sensors to identify different precipitation areas and regimes (0.60 < POD < 0.76; 0.28 < FAR < 0.45; 0.42 < ETS < 0.59;−1.6 mm/h < ME < 1.1 mm/h, with values depending on the radiometer and on the precipitation product). This is particularly relevant in the presence of complex orography in proximity of coastal areas, as for the analyzed cases. The different characteristics of the radiometers (i.e., viewing geometry, spatial resolution, channel assortment) and of retrieval techniques, as well as the limitations of the ground-based reference datasets, are taken into consideration in the evaluation of the accuracy and consistency of the retrievals.

On 5 September 2015 a violent hailstorm hit the Gulf and the city of Naples in Italy. The storm originated over the Tyrrhenian Sea dropping 7-10 cm diameter hailstones along its path. During its mature phase, at 08:47 UTC, the hailstorm was captured by one overpass of the Global Precipitation Measurement mission Core Observatory (GPM-CO) embarking the GPM Microwave Imager (GMI) and the Ka/Ku-band Dual-frequency Precipitation Radar (DPR). In this paper, observations by both GMI and DPR are thoroughly analyzed in conjunction with other spaceborne and ground-based measurements, to show how the GPM-CO integrates established observational tools in monitoring, understanding, and characterizing severe weather. Rapid-scan MSG SEVIRI images show an extremely rapid development, with 10.8 mu m cloud-top temperatures dropping by 65 K in 40 min down to 198 K. The Lightning NETwork registered over 37,000 strokes in 5 h, with intracloud positive stroke fraction increasing during the regeneration phases, when ground-based polarimetric radar and DPR support the presence of large graupel/hail particles. DPR Ku 40 dBZ and 20 dBZ echo top heights at 14 km and 16 km, respectively, indicate strong updraft and deep overshooting. GMI extremely low brightness temperatures (TBs) in correspondence of the convective core (158, 97, 67, and 87 K at 18.7, 36.5, 89 and 166 GHz) are compatible with the presence of massive ice particles. In two years of GPM global observations the storm ranks as fourth and first in terms of minimum 36.5 and 18.7 GHz (V-pol) TBs, respectively. This study illustrates GPM-CO sensing capabilities for characterizing the structure of such severe hailstorm, while providing observational evidence of its intensity and rarity, both globally and over the Mediterranean area. (C) 2017 Elsevier B.V. All rights reserved.

The objective of this paper is to describe the development and evaluate the performance of a completely new version of the Passive microwave Neural network Precipitation Retrieval (PNPR v2), an algorithm based on a neural network approach, designed to retrieve the instantaneous surface precipitation rate using the cross-track Advanced Technology Microwave Sounder (ATMS) radiometer measurements. This algorithm, developed within the EUMETSAT H-SAF program, represents an evolution of the previous version (PNPR v1), developed for AMSU/MHS radiometers (and used and distributed operationally within H-SAF), with improvements aimed at exploiting the new precipitation-sensing capabilities of ATMS with respect to AMSU/MHS. In the design of the neural network the new ATMS channels compared to AMSU/MHS, and their combinations, including the brightness temperature differences in the water vapor absorption band, around 183 GHz, are considered. The algorithm is based on a single neural network, for all types of surface background, trained using a large database based on 94 cloud-resolving model simulations over the European and the African areas. The performance of PNPR v2 has been evaluated through an intercomparison of the instantaneous precipitation estimates with co-located estimates from the TRMM Precipitation Radar (TRMM-PR) and from the GPM Core Observatory Ku-band Precipitation Radar (GPM-KuPR). In the comparison with TRMM-PR, over the African area the statistical analysis was carried out for a 2-year (2013–2014) dataset of coincident observations over a regular grid at 0.5° × 0.5° resolution. The results have shown a good agreement between PNPR v2 and TRMM-PR for the different surface types. The correlation coefficient (CC) was equal to 0.69 over ocean and 0.71 over vegetated land (lower values were obtained over arid land and coast), and the root mean squared error (RMSE) was equal to 1.30 mm h−1 over ocean and 1.11 mm h−1 over vegetated land. The results showed a slight tendency to underestimate moderate to high precipitation, mostly over land, and overestimate moderate to light precipitation over ocean. Similar results were obtained for the comparison with GPM-KuPR over the European area (15 months, from March 2014 to May 2015 of coincident overpasses) with slightly lower CC (0.59 over vegetated land and 0.57 over ocean) and RMSE (0.82 mm h−1 over vegetated land and 0.71 mm h−1 over ocean), confirming a good agreement also between PNPR v2 and GPM-KuPR. The performance of PNPR v2 over the African area was also compared to that of PNPR v1. PNPR v2 has higher R over the different surfaces, with generally better estimation of low precipitation, mostly over ocean, thanks to improvements in the design of the neural network and also to the improved capabilities of ATMS compared to AMSU/MHS. Both versions of PNPR algorithm have shown a general consistency with the TRMM-PR.

This paper describes a new rainfall rate retrieval algorithm, developed within the EUMETSAT H SAF program, based on the Passive microwave Neural network Precipitation Retrieval approach (PNPR v3), designed to work with the conically scanning Global Precipitation Measurement (GPM) Microwave Imager (GMI). A new rain/no-rain classification scheme, also based on the NN approach, which provides different rainfall masks for different minimum thresholds and degree of reliability, is also described. The algorithm is trained on an extremely large observational database, built from GPM global observations between 2014 and 2016, where the NASA 2B-CMB (V04) rainfall rate product is used as reference. In order to assess the performance of PNPR v3 over the globe, an independent part of the observational database is used in a verification study. The good results found over all surface types (CC > 0.90, ME < −0.22 mm h−1, RMSE < 2.75 mm h−1 and FSE% < 100% for rainfall rates lower than 1 mm h−1 and around 30–50% for moderate to high rainfall rates), demonstrate the good outcome of the input selection procedure, as well as of the training and design phase of the neural network. For further verification, two case studies over Italy are also analysed and a good consistency of PNPR v3 retrievals with simultaneous ground radar observations and with the GMI GPROF V05 estimates is found. PNPR v3 is a global rainfall retrieval algorithm, able to optimally exploit the GMI multi-channel response to different surface types and precipitation structures, that provide global rainfall retrieval in a computationally very efficient way, making the product suitable for near-real time operational applications.

Spaceborne microwave cross-track scanning radiometers, originally developed for temperature and humidity sounding, have shown great capabilities to provide a significant contribution in precipitation monitoring both in terms of measurement quality and spatial/temporal coverage. The Passive microwave Neural network Precipi- tation Retrieval (PNPR) algorithm for cross-track scanning radiometers, originally developed for the Advanced Microwave Sounding Unit/Microwave Humidity Sounder (AMSU-A/MHS) radiometers (on board the European MetOp and U.S. NOAA satellites), was recently newly designed to exploit the Advanced Technology Microwave Sounder (ATMS) on board the Suomi-NPP satellite and the future JPSS satellites. The PNPR algorithm is based on the Artificial Neural Network (ANN) approach. The main PNPR-ATMS algorithm changes with respect to PNPR-AMSU/MHS are the design and implementation of a new ANN able to manage the information derived from the additional ATMS channels (respect to the AMSU-A/MHS radiometer) and a new screening procedure for not-precipitating pixels. In order to achieve maximum consistency of the retrieved surface precipitation, both PNPR algorithms are based on the same physical foundation. The PNPR is optimized for the European and the African area. The neural network was trained using a cloud-radiation database built upon 94 cloud-resolving simulations over Europe and the Mediterranean and over the African area and radiative transfer model simulations of TB vectors consistent with the AMSU-A/MHS and ATMS channel frequencies, viewing angles, and view-angle dependent IFOV sizes along the scan projections. As opposed to other ANN precipitation retrieval algorithms, PNPR uses a unique ANN that retrieves the surface precipitation rate for all types of surface backgrounds represented in the training database, i.e. land (vegetated or arid), ocean, snow/ice or coast. This approach prevents different precipitation estimates from being inconsistent with one another when an observed precipitation system extends over two or more types of surfaces. As input data, the PNPR algorithm incorporates the TBs from selected channels, and various additional TBs- derived variables. Ancillary geographical/geophysical inputs (i.e. latitude, terrain height, surface type, season) are also considered during the training phase. The PNPR algorithm outputs consist of both the surface precip- itation rate (along with the information on precipitation phase: liquid, mixed, solid) and a pixel-based quality index. We will illustrate the main features of the PNPR algorithm and will show results of a verification study over Europe and Africa. The study is based on the available ground-based radar and/or rain gauge network observations over the European area. In addition, results of the comparison with rainfall products available from the NASA/JAXA Tropical Rainfall Measuring Mission (TRMM) Precipitation Radar (PR) (over the African area) and Global Precipitation Measurement (GPM) Dual frequency Precipitation Radar (DPR) will be shown. The analysis is built upon a two-years coincidence dataset of AMSU/MHS and ATMS observations with PR (2013-2014) and DPR (2014-2015). The PNPR is developed within the EUMETSAT H/SAF program (Satellite Application Facility for Operational Hydrology and Water Management), where it is used operationally towards the full exploitation of all microwave radiometers available in the GPM era. The algorithm will be tailored to the future European Microwave Sounder (MWS) onboard the MetOp-Second Generation (MetOp-SG) satellites.

Validation study of CDRD and PNPR precipitation microwave retrieval (Level 2) over West Africa using X-pol radar and rain gauges observations

Precipitation retrieval from space has seen great advances thanks to the improving quality of PMW measurements, the refinement of precipitation retrieval techniques, and the increasing number of microwave radiometers on board LEO satellites orbiting around the Earth. With the recent advent of the Global Precipitation Measurement (GPM) mission the constellation of cross-track and conically scanning microwave radiometers with precipitation-sensing capabilities currently ensures 1- to 3-hourly coverage at mid/high latitudes. Traditionally passive microwave (PMW) retrieval algorithms are based on the principle that surface precipitation can be estimated from the mul- tichannel brightness temperature (TB) measurements because these are affected (in different ways depending on channel frequency, viewing geometry, spatial resolution, and surface background conditions) by the microphysical properties and 3-D distribution of liquid and frozen hydrometeors within the precipitating cloud, and, therefore, can be related to surface precipitation. These approaches can be categorized as top-down approaches and they provide instantaneous precipitation rate estimate at the surface at the time of the satellite observation. Recently a new perspective for surface precipitation estimate has been proposed, the bottom-up approach, based on the principle that the soil moisture can be considered as a "natural raingauge" and can be employed for "measuring" rainfall. The algorithm, called SM2RAIN, allows estimating rainfall directly from soil moisture retrieved from spaceborne sensors (i.e. ASCAT). Several recent studies have demonstrated that the approach is very effective for precipitation estimation from the daily to 5-daily scale, even though not applicable in regions where soil moisture retrieval is not feasible (i.e. highly vegetated areas, frozen surfaces, oceans). This study shows that the precipitation estimates obtained by PMW observations using the two approaches (top-down and bottom-up) can be combined to provide accurate estimates of daily rainfall. For the top-down approach two different precipitation retrieval algorithms are used: the Cloud Dynamics Radiation Database algorithm (CDRD), a physically ¬based Bayesian algorithm applied to the conically scanning radiometer SSMIS, and the Passive microwave Neural network Precipitation Retrieval (PNPR) algorithm for cross-track scanning radiometers AMSU/MHS. Four-year (2011-2014) assessment over Italy (complex territory with orography, sea-land interface, vegetation, frozen soils) is carried out for the daily rainfall estimates obtained from SM2RAIN, combined CDRD-PNPR, and merged SM2RAIN-CDRD-PNPR products, all regridded at 0.25ox0.25o spatial resolution. Rainfall measurements from the Italian raingauge network are used as reference. Results show that both SM2RAIN and CDRD-PNPR rainfall products are as accurate as state-of-the-art products, e.g., the TRMM product 3B42RT. The CDRD-PNPR product shows sensitivity to the number and frequency of SSMIS and AMSU/MHS overpasses available, and evidences worse results along the coast, while SM2RAIN shows criticalities mostly related to soil moisture retrieval feasibility. The merged SM2RAIN-CDRD-PNPR product shows better results in terms of correlation (R=0.7), accuracy (RMSE=6.6 mm), and rainfall detection capability (POD=0.75, FAR=0.32, TS=0.5 for 1 mm threshold). The study demonstrates that daily rainfall estimate over a complex territory such as Italy can benefit from the exploitation of multiple PMW radiometers, and from the integration of the top-down and bottom-up approaches.

In this study, precipitation retrievals exploiting the available overpasses of passive microwave (PMW) cross-track and conically scanning radiometers in the GPM era are used to monitor the evolution of heavy precipitation systems occurred during the fall 2014 in Italy. Two different physically-based retrieval algorithms (CDRD for SSMIS and PNPR for AMSU/MHS and ATMS) are used in conjunction with official NASA/JAXA GPM instantaneous precipitation products (for AMSR-2 and GMI). The comparison with dual-polarization radar observations at ground evidences the ability of the different sensors to identify different precipitation areas and regimes. This is particularly relevant in presence of complex orography, often found in proximity of coastal areas for the analyzed cases. Analysis of the accuracy and consistency of the retrievals is carried out taking into account the different spatial resolution and viewing geometry of the different radiometers and the different approaches used for the precipitation retrieval.

The ever increasing availability and accuracy of satellite rainfall and soil moisture products is opening new perspectives in the hydrology and civil protection fields, providing new valuable sources of information for flood and landslide prediction. Despite the abundance of this kind of data, the use of satellite products for such purposes in well gauged regions is still very scarce, most likely due to measurement (bias, accuracy) and resolution issues (both spatial and temporal). Satellite rainfall and soil moisture products have been employed for flood and landslide prediction throughout the Italian territory. For an assessment of performances obtained by satellite products, hydrometeorological ground observations of rainfall, temperature and discharge derived from the monitoring network of the Italian Civil Protection Department (period 2010 - 2014) are used. The satellite rainfall products here considered are: 1) the Tropical Rainfall Measurement Mission (TRMM) Multi - satellite Precipitation Analysis (TMPA) 3B42 - RT product, 2) the H05 product from the H - SAF project, and 3) an experimental daily satellite rainfall product obtained by merging instantaneous rainfall rate estimates from cross - track and conical scanning microwave radiometers. Additionally, the satellite soil moisture products provided by the Advanced SCATterometer (ASCAT) and the Soil Moisture and Ocean Salinity (SMOS) mission are used. In this work, on one hand, satellite rainfall observations are used to drive a continuous rainfall - runoff model (MISDc, Brocca et al., 2011). On the other hand, satellite soil moisture data are used to define the initial conditions for a physically - based slope stability model (TRIGRS, Baum et al., 2008), applied in the Umbria region (central Italy), allowing the estimation of the landslide susceptibility over the territory. The modelled landslide susceptibility, obtained with ground and satellite observations, has been compared with a landslide inventory available in the study area. For the investigated period, MISDc model has been used for flood simulation over different basins in Italy (e.g., Brenta, Tanaro, Tevere, Volturno, ...). For landslide susceptibility evaluation, TRIGRS model has been initialized in two different ways. The first considers modelled soil moisture data computed by a Soil Water Balance model forced with in situ meteorological observations (rainfall and air temperature). The second uses satellite soil moisture products in place of modelled data. The analysis has encompassed some significant storm events occurred in the recent years. As far as the floods prediction is concerned, results highlighted that both bias correction and model recalibration are preliminary steps to be carried out for using properly satellite rainfall data. After performing these steps, satellite rainfall data have allowed to achieve results comparable with those obtained by forcing MISDc model with observed data, confirming the feasibility of using satellite rainfall estimation in hydrological applications. As regards the landslide susceptibility analysis, 102 although spatial and temporal resolution of satellite soil moisture data could be considered too coarse, satisfactory results are obtained as demonstrated from the comparison with the landslide inventory map.

The sensitivity of Global Precipitation Measurement (GPM) Microwave Imager (GMI) high-frequency channels to snowfall at higher latitudes (around 60°N/S) is investigated using coincident CloudSat observations. The 166 GHz channel is highlighted throughout the study due to its ice scattering sensitivity and polarization information. The analysis of three case studies evidences the important combined role of total precipitable water (TPW), supercooled cloud water, and background surface composition on the brightness temperature (TB) behavior for different snow-producing clouds. A regression tree statistical analysis applied to the entire GMI-CloudSat snowfall dataset indicates which variables influence the 166 GHz polarization difference (166 ∆TB) and its relation to snowfall. Critical thresholds of various parameters (sea ice concentration (SIC), TPW, ice water path (IWP)) are established for optimal snowfall detection capabilities. The 166 ∆TB can identify snowfall events over land and sea when critical thresholds are exceeded (TPW > 3.6 kg·m−2, IWP > 0.24 kg·m−2 over land, and SIC > 57%, TPW > 5.1 kg·m−2 over sea). The complex combined 166 ∆TB-TB relationship at higher latitudes and the impact of supercooled water vertical distribution are also investigated. The findings presented in this study can be exploited to improve passive microwave snowfall detection algorithms.