Algorithm Theoretical Definition Document (ATDD) for product …hsaf.meteoam.it/documents/ATDD/ATDD-03.pdf · 2014-11-12 · Algorithms Theoretical Definition Document, 31 August

Algorithms Theoretical Definition Document, 31 August 2010 - ATDD-03 (Product PR-OBS-3) Page 1

Italian Meteorological Service

Italian Department of Civil Defence

Algorithm Theoretical Definition Document (ATDD) for product

PR-OBS-3 - Precipitation rate at ground by GEO/IR

supported by LEO/MW

Zentralanstalt für Meteorologie und

Geodynamik

Vienna University of Technology Institut für Photogrammetrie

und Fernerkundung Royal Meteorological Institute of Belgium

European Centre for Medium-Range Weather Forecasts

Finnish Meteorological Institute

Finnish Environment Institute

Helsinki University of Technology

Météo-France CNRS Laboratoire Atmosphères,

Milieux, Observations Spatiales CNRS Centre d'Etudes

Spatiales de la BIOsphere Bundesanstalt für Gewässerkunde

Hungarian Meteorological Service

CNR - Istituto Scienze dell’Atmosfera

e del Clima Università di Ferrara

Institute of Meteorology and Water Management

Romania National Meteorological Administration

Slovak Hydro-Meteorological Institute

Turkish State Meteorological Service

Middle East Technical University

Istanbul Technical University Anadolu University

31 August 2010


Algorithm Theoretical Definition Document ATDD-03 Product PR-OBS-3

Precipitation rate at ground by GEO/IR supported by LEO/MW

INDEX

Page

Acronyms 03

1. The EUMETSAT Satellite Application Facilities and H-SAF 05

2. Introduction to product PR-OBS-3 06 2.1 Sensing principle 06 2.2 Main operational characteristics 06 2.3 Architecture of the products generation chain 06 2.4 Product development team 07

3. Processing concept 08

4. Algorithms description 09 4.1 The ‘Rapid-update’ processing chain 09 4.2 Processing steps 11 4.3 Additional developments 11 4.4 Algorithm validation/heritage 13

5. Examples of PR-OBS-3 products 15

References 16

List of Tables Table 01 List of H-SAF products 05Table 02 Development team for product PR-OBS-3 07

List of Figures Fig. 01 Conceptual scheme of the EUMETSAT application ground segment 05Fig. 02 Current composition of the EUMETSAT SAF network (in order of establishment) 05Fig. 03 The H-SAF required coverage in the Meteosat projection 06Fig. 04 Flow chart of the LEO/MW-GEO/IR-blending precipitation rate processing chain 07Fig. 05 Rain rate vs. brightness temperature average relationships for the days and the

geographical areas marked. (D) presents the zero-rain thresholds as function of time 10

Fig. 06 Main functions of the RU software package 12Fig. 07 SEVIRI image in the 10.8 µm channel with indicated processing area. Meteosat-9,

day 03 Feb 2008, time 08:15 UTC 15

Fig. 08 Example of map of precipitation rate from SEVIRI + PR-OBS-2. Meteosat-9, day 03 Feb 2008, time 08:15 UTC

15


Acronyms

AMSU Advanced Microwave Sounding Unit (on NOAA and MetOp) AMSU-A Advanced Microwave Sounding Unit - A (on NOAA and MetOp) AMSU-B Advanced Microwave Sounding Unit - B (on NOAA up to 17) ATDD Algorithms Theoretical Definition Document AU Anadolu University (in Turkey) BfG Bundesanstalt für Gewässerkunde (in Germany) CAF Central Application Facility (of EUMETSAT) CDOP Continuous Development-Operations Phase CESBIO Centre d'Etudes Spatiales de la BIOsphere (of CNRS, in France) CM-SAF SAF on Climate Monitoring CNMCA Centro Nazionale di Meteorologia e Climatologia Aeronautica (in Italy) CNR Consiglio Nazionale delle Ricerche (of Italy) CNRS Centre Nationale de la Recherche Scientifique (of France) DMSP Defense Meteorological Satellite Program DPC Dipartimento Protezione Civile (of Italy) EARS EUMETSAT Advanced Retransmission Service ECMWF European Centre for Medium-range Weather Forecasts EDC EUMETSAT Data Centre, previously known as U-MARF EUM Short for EUMETSAT EUMETCast EUMETSAT’s Broadcast System for Environmental Data EUMETSAT European Organisation for the Exploitation of Meteorological Satellites FMI Finnish Meteorological Institute FTP File Transfer Protocol GEO Geostationary Earth Orbit GRAS-SAF SAF on GRAS Meteorology HDF Hierarchical Data Format HRV High Resolution Visible (one SEVIRI channel) H-SAF SAF on Support to Operational Hydrology and Water Management IDL© Interactive Data Language IFOV Instantaneous Field Of View IMWM Institute of Meteorology and Water Management (in Poland) IPF Institut für Photogrammetrie und Fernerkundung (of TU-Wien, in Austria) IPWG International Precipitation Working Group IR Infra Red IRM Institut Royal Météorologique (of Belgium) (alternative of RMI) ISAC Istituto di Scienze dell’Atmosfera e del Clima (of CNR, Italy) ITU İstanbul Technical University (in Turkey) LATMOS Laboratoire Atmosphères, Milieux, Observations Spatiales (of CNRS, in France) LEO Low Earth Orbit LSA-SAF SAF on Land Surface Analysis Météo France National Meteorological Service of France METU Middle East Technical University (in Turkey) MHS Microwave Humidity Sounder (on NOAA 18 and 19, and on MetOp) MSG Meteosat Second Generation (Meteosat 8, 9, 10, 11) MVIRI Meteosat Visible and Infra Red Imager (on Meteosat up to 7) MW Micro Wave NESDIS National Environmental Satellite, Data and Information Services NMA National Meteorological Administration (of Romania) NOAA National Oceanic and Atmospheric Administration (Agency and satellite) NWC-SAF SAF in support to Nowcasting & Very Short Range Forecasting NWP Numerical Weather Prediction


NWP-SAF SAF on Numerical Weather Prediction O3M-SAF SAF on Ozone and Atmospheric Chemistry Monitoring OMSZ Hungarian Meteorological Service ORR Operations Readiness Review OSI-SAF SAF on Ocean and Sea Ice PDF Probability Density Function PEHRPP Pilot Evaluation of High Resolution Precipitation Products Pixel Picture element PMW Passive Micro-Wave PP Project Plan PR Precipitation Radar (on TRMM) PUM Product User Manual PVR Product Validation Report RMI Royal Meteorological Institute (of Belgium) (alternative of IRM) RR Rain Rate RU Rapid Update SAF Satellite Application Facility SEVIRI Spinning Enhanced Visible and Infra-Red Imager (on Meteosat from 8 onwards) SHMÚ Slovak Hydro-Meteorological Institute SSM/I Special Sensor Microwave / Imager (on DMSP up to F-15) SSMIS Special Sensor Microwave Imager/Sounder (on DMSP starting with S-16) SYKE Suomen ympäristökeskus (Finnish Environment Institute) TBB Equivalent Blackbody Temperature (used for IR) TKK Teknillinen korkeakoulu (Helsinki University of Technology) TMI TRMM Microwave Imager (on TRMM) TRMM Tropical Rainfall Measuring Mission UKMO TSMS Turkish State Meteorological Service TU-Wien Technische Universität Wien (in Austria) U-MARF Unified Meteorological Archive and Retrieval Facility UniFe University of Ferrara (in Italy) URD User Requirements Document UTC Universal Coordinated Time VIS Visible ZAMG Zentralanstalt für Meteorologie und Geodynamik (of Austria)


1. The EUMETSAT Satellite Application Facilities and H-SAF The “EUMETSAT Satellite Application Facility on Support to Operational Hydrology and Water Management (H-SAF)” is part of the distributed application ground segment of the “European Organisation for the Exploitation of Meteorological Satellites (EUMETSAT)”. The application ground segment consists of a “Central Application Facility (CAF)” and a network of eight “Satellite Application Facilities (SAFs)” dedicated to development and operational activities to provide satellite-derived data to support specific user communities. See Fig. 01.

Fig. 01 - Conceptual scheme of the EUMETSAT application ground segment.

Fig. 02 reminds the current composition of the EUMETSAT SAF network (in order of establishment).

Nowcasting & Very

Short Range Forecasting Ocean and Sea Ice Ozone & Atmospheric Chemistry Monitoring Climate Monitoring Numerical Weather

Prediction GRAS Meteorology Land Surface Analysis Operational Hydrology & Water Management

Fig. 02 - Current composition of the EUMETSAT SAF network (in order of establishment).

The H-SAF was established by the EUMETSAT Council on 3 July 2005; its Development Phase started on 1st September 2005 and ends on 31 August 2010. The list of H-SAF products is shown in Table 01.

Table 01 - List of H-SAF products Code Acronym Product name H01 PR-OBS-1 Precipitation rate at ground by MW conical scanners (with indication of phase) H02 PR-OBS-2 Precipitation rate at ground by MW cross-track scanners (with indication of phase) H03 PR-OBS-3 Precipitation rate at ground by GEO/IR supported by LEO/MW H04 PR-OBS-4 Precipitation rate at ground by LEO/MW supported by GEO/IR (with flag for phase) H05 PR-OBS-5 Accumulated precipitation at ground by blended MW and IR H06 PR-ASS-1 Instantaneous and accumulated precipitation at ground computed by a NWP model H07 SM-OBS-1 Large-scale surface soil moisture by radar scatterometer H08 SM-OBS-2 Small-scale surface soil moisture by radar scatterometer H09 SM-ASS-1 Volumetric soil moisture (roots region) by scatterometer assimilation in NWP model H10 SN-OBS-1 Snow detection (snow mask) by VIS/IR radiometry H11 SN-OBS-2 Snow status (dry/wet) by MW radiometry H12 SN-OBS-3 Effective snow cover by VIS/IR radiometry H13 SN-OBS-4 Snow water equivalent by MW radiometry

Decentralised processing and generation of products

EUM Geostationary Systems

Systems of the EUM/NOAA Cooperation

Centralised processing and generation of products

Data Acquisition and Control

Data Processing EUMETSAT HQ

Meteorological Products Extraction

EUMETSAT HQ

Archive & Retrieval Facility (Data Centre)

EUMETSAT HQ

USERS

Application Ground Segment

other data sources

Satellite Application

Facilities (SAFs)


2. Introduction to product PR-OBS-3

2.1 Sensing principle

Product PR-OBS-3 (Precipitation rate at ground by GEO/IR supported by LEO/MW) is based on the IR images from the SEVIRI instrument onboard Meteosat satellites. The whole H-SAF area is covered (see Fig. 03, same as for PR-OBS-4 and PR-OBS-5), but the resolution degrades with increasing latitude. The equivalent blackbody temperatures (TBB) are converted to precipitation rate via lookup tables updated at intervals by precipitation rate determinations generated from MW instruments (in H-SAF: PR-OBS-1 and PR-OBS-2). The product is generated at the 15-min imaging rate of SEVIRI, and the spatial resolution is consistent with the SEVIRI pixel. The processing method is called “Rapid Update”.

The SEVIRI channel utilised for PR-OBS-3 is the one at 10.8 µm. The calibration of TBB’s in term of precipitation rate by means of MW measurements (supposedly accurate) implies the existence of good correlation between TBB and precipitation rate. This is fairly acceptable for convective precipitation, less for non-convective. Nevertheless, Rapid Update is currently the only operational algorithm enabling precipitation rate estimates with the time resolution required for nowcasting. In addition, frequent sampling is a prerequisite for computing accumulated precipitation (product PR-OBS-5).

For more information, please refer to the Products User Manual (specifically, PUM-03).

2.2 Main operational characteristics

The operational characteristics of PR-OBS-3 are discussed in PUM-03. Here are the main highlights.

The horizontal resolution (∆x). The IFOV of SEVIRI images is 4.8 km at nadir, and degrades moving away from nadir, becoming about 8 km in the H-SAF area. A figure representative of the PR-OBS-3 resolution is: ~ 8 km. Sampling is made at ~ 5 km intervals, consistent with the SEVIRI pixel over Europe. Conclusion: • resolution ∆x ~ 8 km - sampling distance: ~ 5 km.

The observing cycle (∆t) is defined as the average time interval between two measurements over the same area. In the case of PR-OBS-3 the product is generated soon after each SEVIRI new acquisition, Thus: • observing cycle ∆t = 15 min - sampling time: 15 min. The timeliness (δ). For PR-OBS-3, the time of observations is 1-5 min before each quarter of an hour, ending at the full hour. To this, ~ 5 min have to be added for acquisition through EUMETCast and ~ 5 min for processing at CNMCA, thus: • timeliness δ ~ 15 min.

The accuracy is evaluated a-posteriori by means of the validation activity. See Product Validation Report PVR-03.

2.3 Architecture of the products generation chain

The architecture of the PR-OBS-3 product generation chain is shown in Fig. 04.

Fig. 03 - The H-SAF required coverage in the Meteosat projection.


Actually, Fig. 04 refers to the architecture of the coupled products PR-OBS-3 and PR-OBS-4, that includes: • the Rapid Update process based on (frequent) SEVIRI IR images “calibrated” by the (infrequent)

MW-derived precipitation data as retrieved from SSM/I and SSMIS (PR-OBS-1) or from AMSU-A, AMSU-B or MHS (PR-OBS-2);

• the Morphing process based on (infrequent) MW-derived precipitation maps, and MW precipitation pseudo-maps interpolated at frequent intervals by exploiting the dynamic information provided by the SEVIRI images.

It is noted that, at the time of the ORR in mid-2010, • the Morphing-based product (PR-OBS-4) has not yet reached a potentially pre-operational status; • PR-OBS-3 does not yet make use of MW precipitation data coming from the PR-OBS-1 chain.

The reason for using PR-OBS-2 only (from AMSU/MHS) stems from the delay occurred in developing PR-OBS-1 (from SSM/I-SSMIS). In the descriptions that follow, reference is generally made to SSM/I, since that was used for the early phase of the PR-OBS-3 product development. However, the Rapid Update method works regardless of the source of precipitation measurement used for calibration.

2.4 Product development team

Names and coordinates of the main actors for PR-OBS-3 algorithm development and integration are listed in Table 02.

Table 02 - Development team for product PR-OBS-3

Vincenzo Levizzani (Leader) [email protected] Sante Laviola [email protected] Elsa Cattani

CNR Istituto di Scienze dell’Atmosfera e del Clima (ISAC) Italy

[email protected] Francesco Zauli (Co-Leader) [email protected] Davide Melfi [email protected] Daniele Biron

Centro Nazionale di Meteorologia e Climatologia Aeronautica (CNMCA) Italy

[email protected]

SSM/I-SSMIS

AMSU-MHS

SEVIRI 15-min images

~ 3-hourly sequence of MW observations

Lookup tables updating Rapid-update

algorithm Extraction of

dynamical info

Morphing algorithm

PRECIPITATION RATE

Fig. 04 - Flow chart of the LEO/MW-GEO/IR-blending precipitation rate processing chain.


3. Processing concept The PR-OBS-03 product is based on MW-derived precipitation measurements generated as PR-OBS-1 and PR-OBS-2, and IR images from the geostationary Meteosat satellites namely SEVIRI on Meteosat ≥ 8 (Meteosat Second Generation, MSG). In respect of its precursor instrument MVIRI, SEVIRI, the current basis for products release, provides significantly increased information due to an imaging-repeat cycle of 15 min (30 min for MVIRI) and 12 spectral channels (3 channels for MVIRI), quantisation with 10 bits per pixel (8 for MVIRI), and image sampling distances of 3 km at nadir for all channels except the high-resolution visible with 1 km (for MVIRI, 5 and 2.5 km, respectively). These enhanced time-space resolution performances allow for retrieving timely information on rapid weather development, and are especially instrumental to rainfall retrieval algorithms.

SEVIRI is a 12-channel imager observing the Earth–atmosphere system (Schmetz et al. 2002). Eleven channels observe the Earth’s full disk with a 15-min repeat cycle. A high resolution visible (HRV) channel covers half of the full disk in the east–west direction and a full disk in the north–south direction. The high resolution visible channel has an IFOV of 1.67 km, and the oversampling factor is 1.67 that corresponds to a sampling distance of 1 km at nadir. The corresponding values for the eight thermal IR and the other three solar channels are 4.8 km IFOV, with an oversampling factor of 1.6 and a sampling distance of 3 km for nadir view.

The imaging is performed by combining satellite spin and rotation (stepping) of the scan mirror. The images are taken from south to north and east to west. The E-W scan is achieved through the rotation of the satellite with a nominal spin rate of 100 revolutions/min. The spin axis is nominally parallel to the north–south axis of the Earth. The scan from south to north is achieved through a scan mirror covering the Earth’s disk with about 1250 scan lines; this results in 3750 image lines for channels 1-11 since three detectors for each channel are used for the imaging three parallel lines. A complete image, that is, the full disk of the Earth, consists of nominally 3712 × 3712 pixels for channels 1-11. A nominal repeat cycle is a full-disk imaging of about 12 min, followed by the calibration of thermal IR channels. Then the scan mirror returns to its initial scanning position.

The instantaneous field of view (IFOV) corresponds to the area of sensitivity for each picture element. Since the aperture angle for each IFOV is constant, it follows that the corresponding area at the surface varies with satellite-viewing angle. The image acquisition is based on a constant angular stepping, that is, the subtended angle for each pixel remains constant; hence, the spatial resolution of a pixel at the surface degrades with increasing off-nadir viewing angle.

For the aim of ingesting SEVIRI data into the processing chain of blended algorithms it is necessary to extract from the available image files the information needed by the algorithm. Very briefly, the essential data needed to process SEVIRI image data are: • the starting and ending date and time of acquisition (UTC) for each image data or area subset; • the latitude and longitude coordinates for the geo-location of each pixel in the area; • the Channel 9 (10.8 µm) equivalent blackbody temperature (K) for each pixel. The digital counts are

converted into radiances by means of the calibration coefficients distributed along with each image data. Then, radiances are converted into equivalent blackbody temperature (TBB) by means of regression relationships and the corresponding coefficients;

• the satellite zenith angle of observation for each pixel (deg); • the acquisition time of each pixel (or line of pixels) measured in seconds from the starting time of

acquisition; • nominal line and column number for each pixel.

The IDL© (Interactive Data Language) based ToolBox routines supplied by EUMETSAT has been utilised to extract Level 1.5 data form the reference software and supply the documentation to develop the interface routines and to compare the results.


4. Algorithms description The following Sections describe the algorithms used in the various modules of the precipitation products generation chain. The degree of detail is consistent with the requirement of a manageable document. Detailed algorithm descriptions are available within the H-SAF project inside an electronic forum at the site: ftp://ftp.meteoam.it - username: hsaf - password: 00Hsaf ⇒ hsaf ⇒ algorithm-forum.

4.1 The ‘Rapid-update’ processing chain

The adopted blended technique (Turk et al. 2000 a and b), called RU, has been originally developed at the Naval Research Laboratory in Monterey (CA). In the past, ISAC adapted the original operational set-up of the software (global, automatic, real time, using a suite of MW and IR observations) to the task of analysing test case studies (Torricella et al. 2007).

Key to the RU blended satellite technique is a real time, underlying collection of time and space-intersecting pixels from operational geostationary IR imagers and LEO MW sensors. Rain intensity maps derived from MW measurements are used to create global, geo-located rain rate (RR) and TB (brightness temperature) relationships that are renewed as soon as new co-located data are available from both geostationary and MW instruments. In the software package these relationships are called histograms. To the end of geo-locating histogram relationships, the globe (or the study area) is subdivided in equally spaced lat-lon boxes (2.5° × 2.5°).

As new input datasets (MW and IR) are available in the processing chain, the MW-derived rainrate pixels are paired with their time and space-coincident geostationary 11-µm IR equivalent blackbody temperature (TBB) data, using a 15-minute maximum allowed time offset between the pixel observation times. Each co-located data increments histograms of TBB and RR in the nearest 2.5° latitude-longitude box, as well as the eight surrounding boxes (this overlap ensures a fairly smooth transition in the histogram shape between neighbouring boxes). The rational behind these threshold values for time-collocation and box size is discussed by Turk et al. 2002.

In order to set-up a meaningful statistical ensemble, the method can look at older MW-IR slot intersections, until a certain (75 %) box coverage is reached and a minimum number of coincident observations are gathered for a 2.5 × 2.5 region (at present 400 points, this is a tunable parameter in the procedure). The RU thus requires an initial start-up time period (at present 24 h), to allow for establishing meaningful, initial relationships all over the study area.

As soon as a box is refreshed with new data, a probabilistic histogram matching relationship (Calheiros and Zawadzki 1987) is updated using the MW rainrate and IR TBB probability distribution functions (PDF), and an updated lookup table (histogram file) is created. The matching is performed as follows:

( ) ( )( )

( )

BB

iT

TTBB

R

R

dTTpdRRpBB

BB

i

T

∫∫ = ,

where p(R) and p(TBB) are the PDF or RR and TBB respectively, and RT and TBB(T) are the threshold values.

Examples of TBB vs. RR relationships (histogram) are presented in Fig. 05.


Fig. 05 - Rain rate vs. brightness temperature average relationships for the days and

the geographical areas marked. (D) presents the zero-rain thresholds as function of time.

The global histograms update process is constantly ongoing along with the operational input of MW and geostationary datasets. The transfer of this “background” information to the stream of steadily arriving GEO data involves a computationally fast lookup table and interpolation process for each pixel in the geostationary datasets. Bi-cubic interpolation of the four rainrates surrounding each pixel in the GEO dataset ensures smooth transitions in rain rates across box boundaries. If any histogram is more than 24-hours old relative to the IR dataset time, that histogram is not used. In this case a conventional rain rate value = -1 is assigned to each pixel. However, in ordinary operations, the case is only theoretical since, considering the suite of MW data in input to the algorithm, a histogram more recent than 24 hors is nearly always available. In case of a prolonged interruption of the input data stream (either MW or IR data) the blended product can not be produced and delivered. Moreover, it will require a start-up period of several hours to restart properly.

Within the RU the rain intensities can be derived, in principle, from any source, provided they are geo-located rain intensities measured in mm/h, and contain some useful information (orbit, date, start time, sensor, satellite) to co-locate the information with IR data, as explained above. The state-of-the-art version of the algorithm available at ISAC derives rain maps from SSM/I data. From the brightness temperatures measured in SSM/I channels from 19.2 to 85.5 GHz, rain rates are derived by means of the NOAA-NESDIS operational algorithm (Ferraro and Marks 1995; Ferraro 1997). The NESDIS algorithm derives rain rates at the A-scan resolution of the SSM/I (~ 25 km) by means of non-linear relationships involving the instrument channels (vertical and horizontal polarisation) that have been calibrated using large sets of ground reference data collected by radar networks in different countries. The physical basis of such relationships are the scattering of MW radiation due to large ice particles above the freezing level occurring in precipitating clouds, and the emission from liquid water. This latter phenomenon can be sensed only above oceanic surfaces owing to the high and largely unknown emissivity of land surfaces in the MW spectral range. Relying on MW


measurements only (without the need for a large input-database of physical properties) and on simple but well-founded relationships, this algorithm is very robust and lends itself to global applications.

4.2 Processing steps

From a technical point of view, the RU package comprises executable C and FORTRAN 77 programs, C-shells, and include files containing the tunable parameters of the software. Some of the programs require the linking to HDF standard libraries, if SSM/I rain data are used as input. The main inputs to the RU procedure are: • geolocated equivalent blackbody temperatures observed by the GEO platform (in Kelvin) • rain-rate maps that, in principle, can arise from any satellite based PMW data and algorithm.

For both types of input the following is required: • detailed information on starting and ending date and time of the orbit (or in general of the

observation); • timing of pixel acquisition; • geolocation information (latitude and longitude of the pixel); • spatial resolution; • observation geometry (satellite zenith angle).

The package can be subdivided into four main parts, namely: 1) pre-processing: preparation and pre-processing of GEO data; preparation and pre-processing of

PMW data; computation of rain rate maps at the LEO space-time resolution. To allow for the proper initialization of the statistical relationships the input data must be collected for a time window that start NHOURS before the study period. According to the present constellation of SSM/I sensors onboard DMSP platforms, the parameter NHOURS is currently set to 24 h.

2) co-location: co-located GEO and LEO observations are collected for the selected study area and accumulated from oldest to newest;

3) set-up of geo-located statistical relationships applying the probability matching technique; 4) assign rain rate to each GEO pixel: production of rain-rate maps at the GEO space-time

resolution.

The diagram in Fig. 06 describes the main functions of pre-processing and the general structure of the RU software package. The names of the executable programs and/or C shells that drive the data processing are emphasised in red.

4.3 Additional developments

In the course of the H-SAF Development Phase, the RU algorithm was further developed in CNMCA at instances. Several implementations in the source code occurred, in particular: • pre screening of the IR data using the “Cloud Type” from NWC-SAF was added, that gives the

possibility to clean the meteorological scene from cirrus, broken clouds and semitransparent clouds in general;

• the possibility to include the parallax error correction using the “Cloud Top Height” product from NWC-SAF was added: however, the actual implementation of the feature was delayed to CDOP-1;

• a parallel operational chain with a delay of 3 hours to include the PR-OBS-1 outputs in the RU processing was built. By using PR-OBS-1 in addition to PR-OBS-2, thus improving the frequency of look-up table updating in PR-OBS-3, the accuracy of compute accumulated precipitation (product PR-OBS-5) is expected to improve. However, the actual implementation of the feature was delayed to CDOP-1.


Fig. 06 - Main functions of the RU software package.


4.4 Algorithm validation/heritage

To the aim of assessing the RU performances, one of the first steps that need to be taken before dataset blending, is to account for the characteristics of the MW estimated precipitation as retrieved by the individual MW sensors within the satellite constellation. Owing to different sensor frequencies, scanning modes (conical or cross-track) and polarization states, different precipitation-retrieval algorithms are applied to different sensors. This yields different precipitation retrieval characteristics and possible biases between sensors. One way to accommodate these differences in the RU technique is to select one MW sensor as a reference, and to frequency-match the satellite derived rainfall histograms of the other satellite sensors to the reference histogram (Joyce et al. 2004, 2006). This procedure ensures that with enough observations over a sufficiently long period of time, each sensor will contribute equivalent rainfall statistics to the overall merged rainfall product. In previous RU applications, the TRMM-PR was used as the reference estimate (since it best captures the high rainfall rates, especially over land), and the SSMI at latitudes above 40°N (below 40°S) where TRMM does not orbit. The results showed that while there are differences in the adjustment depending on the local observation time, for over-ocean pixels and latitudes between 20°S and 20°N, the AMSU-B and TMI estimates should be scaled upwards in order to match the TRMM-PR rain histogram.

Recently, the results of a validation study of RU during a three month summer interval indicative of summer monsoon conditions. have been presented (Turk et al. 2009). The validation was based on the Automated Weather Station network operated by the Korean Meteorological Administration over the South Korean peninsula. The motivation for this effort was the belief that an analysis of the performance of the RU should be initiated with a validation system whose overall time sampling can resolve the precipitation scale and intensity over short accumulation windows (hours), and be appropriately centred about the observation time of instantaneous satellite overpasses. The space-time root mean square error, mean bias, and correlation matrices were computed using various time windows for the gauge averaging, centred about the satellite observation time. For ±10 minute time window, a correlation of 0.6 was achieved at 0.1-degree spatial scale by averaging over 3 days; coarsening the spatial scale to 1.8 degrees produced the same correlation by averaging over one hour. Finer than approximately 24-hours and 1-degree time and space scales, respectively, a rapid decay of the error statistics were obtained by trading off either spatial or time resolution. Beyond a daily time scale, the blended estimates were nearly unbiased and with an RMS error of no worse than 1 mm/day.

This kind of validation is fundamentally constrained by the nature of sporadic and intermittent rain falling over a limited number of gauges at short time scales. By analyzing the three month period, there are many short time-scale periods that are averaged together, some with intermittent, sporadic rainfall and others with more widespread rainfall, therefore effects related to rain inhomogeneity across the box size should be averaged to some extent.

To fully examine the overall characteristics and performance of RU, a longer validation time interval is needed, and the analysis should be pooled into tropical and mid-latitude rainfall regimes, summer and winter seasons, and 3-hourly local time windows (to examine if the method is capturing a diurnal cycle). The effort being proposed by the International Precipitation Working Group (IPWG) and its Program to Evaluate High Resolution Precipitation Products (PEHRPP, http://essic.umd.edu/~msapiano/PEHRPP) is focused upon not only pooled analyses, but also regional-scale validation (e.g., Sapiano and Arkin 2009), and comparisons against model-forecasted precipitation from several NWP models (Ebert 2005, 2007; Turk et al. 2006). Even though authors did not examine it, it seems plausible to assume that there is a dependence between the overall MW constellation revisit time and the performance of the blended technique, below some minimum combination of space and time scales.

The RU algorithm in its official, operational set-up, is subject to a continuous effort of validation and intercomparison in different validation sites all over the world, comprising also a site devoted to the validation of precipitation estimates over Europe. The description of the operational set-up of the RU being validated and up-to-date results of the validation activity are available at the IPWG web site at http://www.isac.cnr.it/~ipwg .


It is noted that, although the RU algorithm is designed to assimilate any sort of MW-derived precipitation data, and even any sort of precipitation data coded in BUFR, it is necessary to carefully interface any source with the processing software in order not to create discontinuities due the different nature of the various sources. For instance, PR-OBS-1 and PR-OBS-2 do not observe precipitation in the same way: PR-OBS-1, based on SSM/I-SSMIS, mostly exploits atmospheric windows, whereas PR-OBS-2, based on AMSU-A and AMSU-B/MHS, exploits absorption bands. Therefore, in overlapping areas, the precipitation data could be not consistent, and in the time sequence inclusive of both sources there may be spurious gradients in space and time. During the H-SAF Development Phase only PR-OBS-2 has been entered in PR-OBS-3, in order not to jeopardise the benefit of more frequent update of the lookup tables by introducing possible inconsistency between PR-OBS-1 and PR-OBS-2. PR-OBS-1 will be input in PR-OBS-3 during CDOP-1, initially in a parallel chain optimised to improve PR-OBS-5 (accumulated precipitation) rather than PR-OBS-3 per-sé.


5. Examples of PR-OBS-3 products Fig. 07 shows a SEVIRI image, in its native projection, and the processing area of product PR-OBS-3. The input area includes 900 lines x 1900 columns, from 70°N southwards. However, the algorithm stops processing above 60°N, thus does not cover the full H-SAF area (it could, but the product quality would sharply deteriorate).

Fig. 07 - SEVIRI image in the 10.8 µm channel with indicated processing area. Meteosat-9, day 03 Feb 2008, time 08:15 UTC.

The corresponding precipitation map is shown in Fig. 08. The represented area is a fraction of the total processed area. The map sequences are generally visualised as animations at 15-min intervals.

Fig. 08 - Example of map of precipitation rate from SEVIRI + PR-OBS-2. Meteosat-9, day 03 Feb 2008, time 08:15 UTC.


References

Calheiros R.V. and I. Zawadzki, 1987: “Reflectivity rain-rate relationship for radar hydrology and Brazil”. J. Clim. Appl. Meteor., 26, 118-132.

Ebert E.E., 2005: “Monitoring the quality of operational and semi-operational satellite precipitation estimates: The IPWG validation/intercomparison study”. Proc. 2nd International Precipitation Working Group, 25-28 October, Monterey, pp. 190-199.

Ebert E.E., J. Janowiak and C. Kidd, 2007: “Comparison of near-real-time precipitation estimates from satellite observations and numerical models”. Bull. Amer. Meteor. Soc., 88, 47-64.

Ferraro R.R. and G.F. Marks, 1995: “The development of SSM/I rain-rate retrieval algorithms using ground-based radar measurements”. J. Atmos. Oceanic Technol., 12, 755–770.

Ferraro R.R., 1997: “Special sensor microwave imager derived global rainfall estimates for climatological applications”. J. Geophys. Res., 102 (D14), 16715-16735.

Joyce R.J., J.E. Janowiak, P.A Arkin and P. Xie, 2004: “CMORPH: A method that produces global precipitation estimates from passive microwave and infrared data at high spatial and temporal resolution”. J. Hydrometeor., 5, 487-503.

Joyce R.J., J.E. Janowiak, P.A. Arkin, and P. Xie, 2006: “The combination of a passive microwave based satellite rainfall estimation algorithm with an IR-based algorithm”. 14th AMS Conf. Sat. Meteor. Ocean., 29 Jan-3 Feb, Atlanta.

Sapiano M.R.P. and P.A. Arkin, 2009: “An intercomparison and validation of high-resolution satellite precipitation estimates with 3-hourly gauge data”. J. Hydrometeor., 10, 149-166.

Schmetz J., P. Pili, S. Tjemkes, D. Just, J. Kerkmann, S. Rota and A. Ratier, 2002: “An introduction to METEOSAT Second Generation (MSG)”. Bull. Amer. Meteor. Soc., 83, 977-992.

Torricella F., V. Levizzani and F.J. Turk, 2007: “Application of a blended MW-IR rainfall algorithm to the Mediterranean”. In: Measuring precipitation from space – EURAINSAT and the future. V. Levizzani, P. Bauer, and F. J. Turk, Eds., Springer, 497-507.

Turk F.J., E.E. Ebert, B.-J. Sohn, H.-J. Oh, V. Levizzani, E.A. Smith and R.R. Ferraro, 2002: “Validation of an operational global precipitation analysis at short time scales”. Proc. 1st IPWG Workshop, Madrid, 23-27 Sept., 225-248.

Turk J.F., B.-J. Sohn, H.-J. Oh, E.E. Ebert, V. Levizzani, and E.A. Smith, 2009: “Validating a rapid-update satellite precipitation analysis across telescoping space and time scales”. Submitted to Meteor. Atmos. Phys.

Turk J.F., G. Rohaly, J. Hawkins, E.A. Smith, F.S. Marzano, A. Mugnai and V. Levizzani, 2000a: “Meteorological applications of precipitation estimation from combined SSM/I, TRMM and geostationary satellite data”. Microwave Radiometry and Remote Sensing of the Earth’s Surface and Atmosphere, P. Pampaloni and S. Paloscia Eds., VSP Int. Sci. Publisher, Utrecht (The Netherlands), 353-363.

Turk J.F., G. Rohaly, J. Hawkins, E.A. Smith, F.S. Marzano, A. Mugnai and V. Levizzani, 2000b: “Analysis and assimilation of rainfall from blended SSMI, TRMM and geostationary satellite data”. Proc. 10th AMS Conf. Sat. Meteor. and Ocean., 9, 66-69.

Turk J.F., P. Bauer, E.E. Ebert, and P.A. Arkin, 2006: “Satellite-derived precipitation verification activities within the International Precipitation Working Group”. 14th AMS Conf. Sat. Meteor. Ocean., 29 Jan-3 Feb, Atlanta.

Documents

Algorithm Theoretical Definition Document (ATDD) for product …hsaf.meteoam.it/documents/ATDD/ATDD-03.pdf · 2014-11-12 · Algorithms Theoretical Definition Document, 31 August