Towards combining GPM and MFG observations to monitor near real time heavy precipitation at fine scale over India and nearby oceanic regions
Introduction
Near Real time precipitation information at fine scale is crucial for various applications ranging from flash flood monitoring to weather forecasting. Changes in precipitation extremes over India and other parts of the world have been reported in the context of changing climate (Goswami et al., 2006, Rajeevan et al., 2008, Dash et al., 2009, O’Gorman, 2012, Villarini et al., 2013, Mishra and Liu, 2014). These changes include increase in heavy precipitation causing floods and decrease in low and moderate precipitation causing droughts. India’s economy depends on agriculture which in turn is greately affected by floods and droughts. In recent years, many parts of Indian region suffered from severe floods (Mishra and Srinivasan, 2013, Mishra and Liu, 2014, Mishra, 2015, Mishra, 2016). Near Real time precipitation information at fine resolution is required to monitor flood. Unfortunately Indian region have very poor ground based rain gauge and Radar Density. Moreover, rain gauge stations stop working during severe flood situatuions (Mishra, 2015). Satellite remote sensing from space offers a unique opportunity to monitor rainy storms. Microwaves due to their larger wavelength may penetrate through the clouds and interact with cloud droplets. Thus Passive Microwave observations offer opportunity to provide more accurate precipitation information as compared to IR observations. However, these observations suffer from poor temporal and spatial resolution (Stephens and Kummerow, 2007). Flood events are associated with a large spatial and temporal variation of rainfall and hence continuous near real time high resolution hourly satellite data is essential to monitor such events (Mishra and Srinivasan, 2013). This kind of estimates can be achieved by merging microwave precipitation estimates with observations from Geostationary satellites. Past researches show that cold observations from IR are associated with convective clouds and thus heavy precipitation (Arkin, 1979, Arkin and Meisner, 1987). In the last decade, various satellite precipitation products have become widely available for users. These data sets integrate different estimates of precipitation from different sensors and satellites into a precipitation product. These data sets include the Tropical Rainfall Measuring Mission (TRMM) Multisatellite Precipitation Analysis (TMPA) near real time product (Huffman et al., 2007), the Global Satellite Mapping of Precipitation (GSMaP) (Kubota et al., 2007, Aonashi et al., 2010), Climate Prediction Centre MORPHing (CMORPH) (Joyce et al., 2004), Precipitation Estimation from Remote Sensing Information using Artificial Neural Network (PERSIANN) (Hsu et al., 1997), Hydro-Estimator (H-E) (Scofield and Kuligowski, 2003), and Integrated Multi-satellitE Retrievals for GPM (IMERG) (Huffman et al., 2015). Validation results over India show that most of these products have large errors (Mishra et al., 2010, Mishra, 2013, Prakash and Sathiyamoorthi, 2014). Mishra et al. (2009) reported that regional rain signatures derived for India outperform global rainfall signatures for their application over India. Few efforts have been made to monitor rainfall over India and nearby region by synergistic use of multisatellite sensors (Mishra et al., 2009, Mishra et al., 2010, Mishra et al., 2011, Mishra, 2012a, Mishra, 2012b, Mishra, 2013). Furthermore, rainfall estimates over Mountain terrains suffer large errors because heavy rainfall associated with shallow orographic rainfall storms can be underestimated due to weak scattering and emission signatures from Microwave and IR sensors, respectively. These errors become crucial over Indian region which is rich in mountain terrains. Recent studies over Indian region focus on the errors in rainfall estimates (Sunilkumar et al., 2015, Prakash et al., 2016). Techniques to improve rainfall estimates over complex terrains are discussed in previous studies (Shige et al., 2013, Shige et al., 2014, Taniguchi et al., 2013, Yamamoto and Shige, 2015). Very recently, INSAT Multispectral Rainfall Algorithm (IMSRA), an operational rainfall algorithm developed over India, has been improved by applying topographic corrections (Upadhyaya and Ramsankaran, 2016). Mishra et al. (2010) merged observations for Precipitation Radar (PR) onboard TRMM and Meteosat to develop a rainfall estimation algorithm to monitor rainfall over India. The technique consists of two steps. First, a cloud classification approach is used to filter out cirrus clouds using multispectral measurements from Meteosat. Next, the collocated measurements from PR and Meteosat are used to derive a relationship between rainfall and brightness temperature. This relationship is used to estimate rainfall from Meteosat brightness temperature observations. However, PR suffers from several problem due to single frequency (Ku band). These include errors arising due to radar system calibration, Rayleigh fading and quantization, drop size distribution (DSD) and non-uniformity of rain distribution. The most crucial one is uncertainty of rain distribution. These errors are discussed in Iguchi et al. (2009). Availability of additional microwave measurements with broader swath and high frequency ice scattering channels from Global Precipitation Measurement (GPM) provide a unique opportunity to merge accurate microwave rainfall information with Infrared observations from Meteosat over India. The GPM Core Observatory measures precipitation using two sensors: the GPM Microwave Imager (GMI) and the Dual-frequency Precipitation Radar (DPR). The combination of observations from dual frequency DPR provides more accurate estimates of DSD parameters than PR (Iguchi 2003). Additional Ka band observations offer high sensitivity to weak rain and snow. Additional frequency at 157 GHz from GMI is useful to monitor high level ice cloud systems associated with heavy precipitation. The GMI supplies information on cloud structure and on the type (i.e., liquid or ice) of cloud particles. Data from the DPR provides insights into the three-dimensional structure of precipitation, along with layer-by-layer estimates of the sizes of raindrops, within and below the cloud. Combined observation from GMI and DPR is expected to provide most accurate precipitation observations. A recent preliminary study reports that rainfall estimates from Dual-frequency Precipitation Radar (DPR) onboard GPM are closer to the gauge estimates than those from Precipitation Radar (PR) onboard TRMM. It may be noted that GPM improves on the capabilities of Tropical Rainfall Measuring Mission (TRMM) in numerous ways. GPM’s radar is the only dual frequency radar in space, and is capable of creating 3-D profiles and intensity estimates of precipitation. GPM’s radiometer has a greater frequency range than TRMM’s (13 channels versus 9 channels), which allows GPM to measure precipitation intensity. One of the most significant evolutions in GPM data is its broader global coverage. While TRMM collected data in tropical and subtropical regions between roughly 35° north and south latitude, GPM collects data between approximately 65° north and south latitude. This allows GPM’s instruments to collect data on storms as they form in the tropics and move into the middle and high latitudes. Combined observations of GMI and DPR also offers a wide swath of about 904 km as compared to 215 km of PR used in Mishra et al. (2010). DPR observations are accurate but offers poor coverage (due to very narrow swath; 245 Km maximum). The major role of the GMI is to improve accuracy of rainfall/snowfall estimates by simultaneous observation with the DPR (Oki et al., 2010). Combined observations of GMI and DPR also offers a wide swath of about 904 km.
The increased sensitivity of the Dual-frequency Precipitation Radar (DPR) and the high-frequency channels on the GPM Microwave Imager (GMI) will enable GPM to improve forecasting by estimating light rain and falling snow outside the tropics, even in the winter seasons, which other satellites are unable to measure (Huffman et al., 2015).
In this study, we have merged observations from GPM and Meteosat to monitor near real time precipitation over Indian region and nearby ocean. Validation has been performed using rain gauge based product to test the accuracy of the present approach for its application in heavy rainfall cases. We have also tested the performance of the satellite based rainfall estimates with hourly rainfall observation from rain gauges over southern part of India.
Section snippets
Data used and study area
For the present study Meteosat data of Meteosat First Generation (MFG) is used. MFG provides images of the full Earth disc, and data for weather forecasts. Meteosat provides observations in Thermal Infra Red (TIR) and Water Vapor (WV) absorption band at half-hourly interval, with a spatial resolution of 4 km. Data has been downloaded from EUMETSAT Portal (https://www.eumetsat.int/website/home/Data/DataDelivery/EUMETSATDataCentre/index.html). We have used Meteosat 7 data from Meteosat First
Methodology
First of all, accuracy of GPM rainfall was validated against rain gauge observations over Indian region (Fig. 1b). For this purpose, 3879 data points during monsoon season of 2016 have been used during rainy spell (July 06–09, 16–18; August 12–14, 19–23; September 15–19, 28–30; October 03–09, 14–18, 26–29; November 10–15, 18–23, 27–28). Fig. 2 shows the scatter plot between hourly rainfall from rain gauges and that from GPM observations.
It can be seen that rainfall observations from rain gauges
Results and discussion
Primary aim of the present algorithm is to monitor near real time heavy rainy systems. For this reason, performance of this technique was tested by applying it to few cyclonic cases during landfall. We have compared the results with daily IMD gridded rainfall available at 0.25 ° resolution. The IMD product uses gauge observations from stations to estimate daily accumulated rainfall in the 24 h ending 0300 GMT. Hourly rainfall obtained from the present algorithm is integrated (24 images) and
Conclusion
GMI-DPR combined precipitation presents an opportunity to measure accurate precipitation but suffers from limited availability and sampling gaps. Present study merges rainfall information from combined GMI-DPR with Meteosat observations by utilizing the high accuracy of GMI-DPR based rainfall estimates and continuous Meteosat observations to monitor heavy precipitation. It offers an opportunity to explore the climatic aspect of heavy precipitation at finer scale since MFG has long past records.
Acknowledgements
We acknowledge the funding for this work from MoES under grant MoES/16/27/2014-RDEAS, from SERB DST under grant SR/FTP/ES-116/2014 and from ISRO under grant No. B. 19012/174/2016-Sec.-2. Meteosat data from EUMETSET and GPM data used in this study is also thankfully acknowledged. Useful suggestions from two anonymous reviewers and editor are thankfully acknowledged.
References (49)
- et al.
Rainfall Retrieval over Indian land and oceanic regions from SSM/I Microwave data
Adv. Space Res.
(2009) - et al.
Retrieving microphysical properties near the tops of potential rain clouds by multispectral analysis of AVHRR data
Atmos. Res.
(1994) - et al.
Modified-INSAT multi-Spectral rainfall algorithm (M-IMSRA) at climate region scale: development and validation
Remote Sens. Environ.
(2016) - et al.
Implementation of an orographic/nonorographic rainfall classification scheme in the GSMaP algorithm for microwave radiometers
Atmos. Res.
(2015) - et al.
A satellite infrared technique to estimate tropical convective and stratiform rainfall
J. Appl. Meteorol.
(1988) - et al.
GSMaP passive microwave precipitation retrieval algorithm: algorithm description and validation
J. Meteorol. Soc. of Japan
(2010) - et al.
The relationship between large-scale convective rainfall and cold cloud over the western hemisphere during 1982-84
Mon. Weather Rev.
(1987) The relationship between fractional coverage of high cloud and rainfall accumulations during GATE over the B-scale array
Mon. Weather Rev.
(1979)- et al.
GOES multispectral rainfall algorithm (GMSRA)
J. Appl. Meteorol.
(2001) - et al.
Rainfall estimation from a combination of TRMM precipitation radar and GOES multispectral satellite imagery through the use of an artificial neural network
J. Appl. Meteorol.
(2000)
Changes in the characteristics of rain events in India
J. Geophys. Res.
Increasing trends of extreme rain events over India in a warming environment
Science
Combining Satellite Microwave Radiometer and Radar observations to estimate atmospheric heating profiles
J. Clim.
The GPM combined algorithm
J. Atmos. Oceanic Technol.
Precipitation estimation from remotely sensed information using artificial neural networks
J. Appl. Meteorol.
The TRMM multisatellite precipitation analysis (TMPA): quasi-global, multiyear, combined-sensor precipitation estimates at fine scales
J. Hydrometeorol.
NASA Global Precipitation Measurement (GPM) Integrated Multi-satellitE Retrievals for GPM (IMERG), Algorithm Theoretical Basis Document (ATBD) Version 4.5
Uncertainties in the rain profiling algorithm for the TRMM precipitation radar
J. Meteorolog. Soc. Jpn.
Spaceborne precipitation radars in TRMM and GPM
CMORPH: A method that produces global precipitation estimates from passive microwave and infrared data at high spatial and temporal resolution
J. Hydromet.
Global precipitation map using satellite-borne microwave radiometers by the GSMaP project: production and validation
IEEE T. Geosci. Remote
Investigating a SSM/I microwave algorithm to calibrate METEOSAT infrared instantaneous rainrate estimates
Meteorol. Appl.
Changes in precipitation pattern and risk of drought over India in the context of global warming
J. Geophys. Res.
Cited by (8)
Forecasting water level of Glacial fed perennial river using a genetically optimized hybrid Machine learning model
2021, Materials Today: ProceedingsCitation Excerpt :The world is not only witnessing the increase in the flood events but also the intensity or we can say high-impact floods. Flooding is a convoluted anomaly which depends on many factors such as heavy precipitation [21,28,29], snow-melt [15,30], landslides [40,41], GLOF (Glacial Lake Outburst Flood) (Ives et al. 2010, [39,17] dike and dam breaches [43], storm and coastal flooding. Not out of all these factors, most of them cause rivers to flood while others have a pathway or a definite course.
Improving precipitation estimates over orographic regions by integrating multi-spectral satellite observations with rain gauge in India
2019, Remote Sensing Applications: Society and EnvironmentCitation Excerpt :Bias adjustment technique adopted in present study utilizes excellent temporal and spatial (hourly rainfall at 0.1° resolution) resolution of multi-spectral satellite observations and accurate rain gauge estimates to improve heavy precipitation estimates over orographic region. Precipitation estimates derived from multi-spectral satellite from MFG and GPM using technique adopted by Mishra and Rafiq (2017) have been used in this study. A description of the technique and inputs used for deriving precipitation has been provided in Mishra and Rafiq (2017).
Refined Evaluation of Satellite Precipitation Products against Rain Gauge Observations along the Sichuan—Tibet Railway
2022, Journal of Meteorological ResearchObserving a severe flooding over southern part of India in monsoon season of 2019
2021, Journal of Earth System Science