• Barros, A. P., , and Tao K. , 2008: A space-filling algorithm to extrapolate narrow-swath instantaneous TRMM microwave rain-rate estimates using thermal IR imagery. J. Atmos. Oceanic Technol., 25, 19011920, doi:10.1175/2008JTECHA1019.1.

    • Search Google Scholar
    • Export Citation
  • Barros, A. P., , Joshi M. , , Putkonen J. , , and Burbank D. W. , 2000: A study of the 1999 monsoon rainfall in a mountainous region in central Nepal using TRMM products and rain gauge observations. Geophys. Res. Lett., 27, 36833686, doi:10.1029/2000GL011827.

    • Search Google Scholar
    • Export Citation
  • Bitew, M. M., , and Gebremichael M. , 2011: Assessment of satellite rainfall products for streamflow simulation in medium watersheds of the Ethiopian highlands. Hydrol. Earth Syst. Sci., 15, 11471155, doi:10.5194/hess-15-1147-2011.

    • Search Google Scholar
    • Export Citation
  • Bitew, M. M., , Gebremichael M. , , Ghebremichael L. T. , , and Bayissa Y. A. , 2012: Evaluation of high-resolution satellite rainfall products through streamflow simulation in a hydrological modeling of a small mountainous watershed in Ethiopia. J. Hydrometeor., 13, 338350, doi:10.1175/2011JHM1292.1.

    • Search Google Scholar
    • Export Citation
  • Chen, S., and et al. , 2013: Performance evaluation of radar and satellite rainfalls for Typhoon Morakot over Taiwan: Are remote-sensing products ready for gauge denial scenario of extreme events? J. Hydrol., 506, 413, doi:10.1016/j.jhydrol.2012.12.026.

    • Search Google Scholar
    • Export Citation
  • Chow, V. T., , Maidment D. R. , , and Mays L. W. , 1988: Applied Hydrology. McGraw-Hill, New York, 572 pp.

  • Derin, Y., , and Yilmaz K. K. , 2014: Evaluation of multiple satellite-based precipitation products over complex topography. J. Hydrometeor., 15, 14981516, doi:10.1175/JHM-D-13-0191.1.

    • Search Google Scholar
    • Export Citation
  • Dinku, T., , Chidzambwa S. , , Ceccato P. , , Connor S. J. , , and Ropelewski C. F. , 2008: Validation of high‐resolution satellite rainfall products over complex terrain. Int. J. Remote Sens., 29, 40974110, doi:10.1080/01431160701772526.

    • Search Google Scholar
    • Export Citation
  • Dinku, T., , Ruiz F. , , Connor S. J. , , and Ceccato P. , 2010: Validation and intercomparison of satellite rainfall estimates over Colombia. J. Appl. Meteor. Climatol., 49, 10041014, doi:10.1175/2009JAMC2260.1.

    • Search Google Scholar
    • Export Citation
  • Fulton, R. A., , Breidenbach J. P. , , Seo D. J. , , Miller D. A. , , and Miller T. , 1998: The WSR-88D rainfall algorithm. Wea. Forecasting, 13, 377395, doi:10.1175/1520-0434(1998)013<0377:TWRA>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Gao, Y. C., , and Liu M. F. , 2013: Evaluation of high-resolution satellite precipitation products using rain gauge observations over the Tibetan Plateau. Hydrol. Earth Syst. Sci., 17, 837, doi:10.5194/hess-17-837-2013.

    • Search Google Scholar
    • Export Citation
  • Gochis, D., and et al. , 2015: The great Colorado flood of September 2013. Bull. Amer. Meteor. Soc., 96, 14611487, doi:10.1175/BAMS-D-13-00241.1.

    • Search Google Scholar
    • Export Citation
  • Gourley, J. J., , Hong Y. , , Flamig Z. L. , , Wang J. , , Vergara H. , , and Anagnostou E. N. , 2011: Hydrologic evaluation of rainfall estimates from radar, satellite, gauge, and combinations on Ft. Cobb basin, Oklahoma. J. Hydrometeor., 12, 973988, doi:10.1175/2011JHM1287.1.

    • Search Google Scholar
    • Export Citation
  • Heidke, P., 1926: Berechnung des Erfolges und der Güte der Windstärkevorhersagen im Sturmwarnungsdienst. Geogr. Ann., 8, 301349.

  • Hirpa, F. A., , Gebremichael M. , , and Hopson T. , 2010: Evaluation of high-resolution satellite precipitation products over very complex terrain in Ethiopia. J. Appl. Meteor. Climatol., 49, 10441051, doi:10.1175/2009JAMC2298.1.

    • Search Google Scholar
    • Export Citation
  • Hou, A. Y., and et al. , 2014: The Global Precipitation Measurement Mission. Bull. Amer. Meteor. Soc., 95, 701722, doi:10.1175/BAMS-D-13-00164.1.

    • Search Google Scholar
    • Export Citation
  • Janowiak, J. E., , and Xie P. , 1999: CAMS–OPI: A global satellite–rain gauge merged product for real-time precipitation monitoring applications. J. Climate, 12, 33353342, doi:10.1175/1520-0442(1999)012<3335:COAGSR>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Joyce, R. J., , Janowiak J. E. , , Arkin P. A. , , and Xie P. , 2004: CMORPH: A method that produces global precipitation estimates from passive microwave and infrared data at high spatial and temporal resolution. J. Hydrometeor., 5, 487503, doi:10.1175/1525-7541(2004)005<0487:CAMTPG>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Koren, V., , Smith M. , , and Duan Q. , 2003: Use of a priori parameter estimates in the derivation of spatially consistent parameter sets of rainfall–runoff models. Calibration of Watershed Models, Water Science and Application Series, Vol. 6, Amer. Geophys. Union, 239–254.

  • Lang, T. J., , and Barros A. P. , 2002: An investigation of the onsets of the 1999 and 2000 monsoons in central Nepal. Mon. Wea. Rev., 130, 12991316, doi:10.1175/1520-0493(2002)130<1299:AIOTOO>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Lin, Y., , and Mitchell K. E. , 2005: The NCEP stage II/IV hourly precipitation analyses: Development and applications. 19th Conf. on Hydrology, San Diego, CA, Amer. Meteor. Soc., 1.2. [Available online at http://www.emc.ncep.noaa.gov/mmb/ylin/pcpanl/refs/stage2-4.19hydro.pdf.]

  • Mei, Y., , Anagnostou E. N. , , Nikolopoulos E. I. , , and Borga M. , 2014: Error analysis of satellite precipitation products in mountainous basins. J. Hydrometeor., 15, 17781793, doi:10.1175/JHM-D-13-0194.1.

    • Search Google Scholar
    • Export Citation
  • Mendoza, P. A., , Rajagopalan B. , , Clark M. P. , , Ikeda K. , , and Rasmussen R. M. , 2015: Statistical postprocessing of high-resolution regional climate model output. Mon. Wea. Rev., 143, 15331553, doi:10.1175/MWR-D-14-00159.1.

    • Search Google Scholar
    • Export Citation
  • NCAR, 2012: Weather Research and Forecasting ARW: Version 3 modeling system user’s guide. NCAR, 384 pp.[Available online at http://www2.mmm.ucar.edu/wrf/users/docs/user_guide_V3.4/ARWUsersGuideV3.pdf.]

  • Nikolopoulos, E. I., , Bartsotas N. , , Anagnostou E. N. , , and Kallos G. , 2015: Using high-resolution numerical weather forecasts to improve remotely sensed rainfall estimates: The case of the 2013 Colorado flash flood. J. Hydrometeor., 16, 17421751, doi:10.1175/JHM-D-14-0207.1.

    • Search Google Scholar
    • Export Citation
  • Pan, M., , Li H. , , and Wood E. , 2010: Assessing the skill of satellite‐based precipitation estimates in hydrologic applications. Water Resour. Res., 46, W09535, doi:10.1029/2009WR008290.

    • Search Google Scholar
    • Export Citation
  • Prat, O. P., , and Barros A. P. , 2010: Assessing satellite-based precipitation estimates in the southern Appalachian mountains using rain gauges and TRMM PR. Adv. Geosci., 25, 143153, doi:10.5194/adgeo-25-143-2010.

    • Search Google Scholar
    • Export Citation
  • Romilly, T. G., , and Gebremichael M. , 2011: Evaluation of satellite rainfall estimates over Ethiopian river basins. Hydrol. Earth Syst. Sci., 15, 15051514, doi:10.5194/hess-15-1505-2011.

    • Search Google Scholar
    • Export Citation
  • Skamarock, W. C., and et al. , 2008: A description of the Advanced Research WRF version 3. NCAR Tech. Note NCAR/TN-475+STR, 113 pp., doi:10.5065/D68S4MVH.

  • Stampoulis, D., , and Anagnostou E. N. , 2012: Evaluation of global satellite rainfall products over continental Europe. J. Hydrometeor., 13, 588603, doi:10.1175/JHM-D-11-086.1.

    • Search Google Scholar
    • Export Citation
  • Stampoulis, D., , Anagnostou E. N. , , and Nikolopoulos E. I. , 2013: Assessment of high-resolution satellite-based rainfall estimates over the Mediterranean during heavy precipitation events. J. Hydrometeor., 14, 15001514, doi:10.1175/JHM-D-12-0167.1.

    • Search Google Scholar
    • Export Citation
  • Thompson, G., , Field P. R. , , Hall W. D. , , and Rasmussen R. M. , 2006: A new bulk microphysical parameterization for WRF (& MM5). Proc. Seventh Weather Research and Forecasting Model Workshop, Boulder, CO, NCAR, 11 pp. [Available online at http://www2.mmm.ucar.edu/wrf/users/workshops/WS2006/abstracts/Session05/5_3_Thompson.pdf.]

  • Vergara, H., , Hong Y. , , Gourley J. J. , , Anagnostou E. N. , , Maggioni V. , , Stampoulis D. , , and Kirstetter P. E. , 2014: Effects of resolution of satellite-based rainfall estimates on hydrologic modeling skill at different scales. J. Hydrometeor., 15, 593613, doi:10.1175/JHM-D-12-0113.1.

    • Search Google Scholar
    • Export Citation
  • Wang, J., and et al. , 2011: The Coupled Routing and Excess Storage (CREST) distributed hydrological model. Hydrol. Sci. J., 56, 8498, doi:10.1080/02626667.2010.543087.

    • Search Google Scholar
    • Export Citation
  • Wilk, J., , Kniveton D. , , Andersson L. , , Layberry R. , , Todd M. C. , , Hughes D. , , Ringrose S. , , and Vanderpost C. , 2006: Estimating rainfall and water balance over the Okavango River basin for hydrological applications. J. Hydrol., 331, 1829, doi:10.1016/j.jhydrol.2006.04.049.

    • Search Google Scholar
    • Export Citation
  • Xie, P., , Yoo S. H. , , Joyce R. , , and Yarosh Y. , 2011: Bias-corrected CMORPH: A 13-year analysis of high-resolution global precipitation. Geophysical Research Abstracts, Vol. 13, Abstract EGU2011-1809. [Available online at http://meetingorganizer.copernicus.org/EGU2011/EGU2011-1809.pdf.]

  • Zeweldi, D. A., , and Gebremichael M. , 2009: Evaluation of CMORPH precipitation products at fine space–time scales. J. Hydrometeor., 10, 300307, doi:10.1175/2008JHM1041.1.

    • Search Google Scholar
    • Export Citation
  • Zhang, X., , Anagnostou E. N. , , Frediani M. , , Solomos S. , , and Kallos G. , 2013: Using NWP simulations in satellite rainfall estimation of heavy precipitation events over mountainous areas. J. Hydrometeor., 14, 18441858, doi:10.1175/JHM-D-12-0174.1.

    • Search Google Scholar
    • Export Citation
  • Zulkafli, Z., , Buytaert W. , , Onof C. , , Manz B. , , Tarnavsky E. , , Lavado W. , , and Guyot J. L. , 2014: A comparative performance analysis of TRMM 3B42 (TMPA) versions 6 and 7 for hydrological applications over Andean–Amazon river basins. J. Hydrometeor., 15, 581592, doi:10.1175/JHM-D-13-094.1.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    Study area with 20 basins. (Elevation data source: Aster Global DEM; https://asterweb.jpl.nasa.gov/gdem.asp.) The map also depicts WRF’s inner domain at 2-km resolution (390 × 324 grids).

  • View in gallery

    Quantile–quantile plot of WRF vs stage IV hourly rain rates for the six hurricane events.

  • View in gallery

    Event-total precipitation accumulation maps from stage IV, WRF, original CMORPH, WRF-adjusted CMORPH, and gauge-adjusted CMORPH, for hurricanes (a)–(e) Bill, (f)–(j) Gaston, (k)–(o) Frances, (p)–(t) Ivan, (u)–(y) Cindy, and (z)–(δ) Fay.

  • View in gallery

    Quantile–quantile plots of original CMORPH vs WRF and stage IV hourly rain rates for the six hurricane events.

  • View in gallery

    BS of (left) original CMORPH and WRF and (right) WRF-adjusted and gauge-adjusted CMORPH vs rain-rate threshold.

  • View in gallery

    As in Fig. 5, but for the HSS metric.

  • View in gallery

    Scatterplots of stage IV vs estimated (original, WRF-adjusted, and gauge-adjusted CMORPH) basin-average accumulated rainfall.

  • View in gallery

    Basin B6 runoff time series from CREST simulations.

  • View in gallery

    Scatterplots of stage IV vs estimated rainfall-driven simulated (top) basin outlet peak runoff and (bottom) basin outlet accumulated runoff.

  • View in gallery

    The NRMSE ratio of (a) accumulated rainfall and (b) accumulated runoff for each event.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 36 36 4
PDF Downloads 26 26 1

Hydrologic Evaluation of NWP-Adjusted CMORPH Estimates of Hurricane-Induced Precipitation in the Southern Appalachians

View More View Less
  • 1 Department of Civil and Environmental Engineering, University of Connecticut, Storrs, Connecticut
  • | 2 School of Civil Engineering and Environmental Science, University of Oklahoma, Norman, Oklahoma
© Get Permissions
Full access

Abstract

Satellite-retrieved precipitation has the potential to support flood modeling in mountainous areas. However, to reach this potential satellite estimates need to be corrected for the severe underestimation exhibited in orography-induced heavy precipitation events (HPEs). This paper assesses an existing satellite precipitation error correction technique driven by high-resolution numerical weather prediction (NWP) simulations of HPEs in complex terrain. The study is based on NOAA Climate Prediction Center morphing technique (CMORPH) high-resolution precipitation estimates of six such events induced by hurricane landfalls in the southern Appalachian mountainous region. A distributed hydrological model (Coupled Routing and Excess Storage model) is applied to evaluate the impact of the proposed satellite precipitation error correction on flood simulations for 20 basins of various sizes in this mountainous region. The results demonstrate significant improvements due to the NWP-based adjustment technique in terms of both the precipitation error characteristics and corresponding runoff simulations. These improvements are shown to be comparable to those from the postprocessed gauge-adjusted CMORPH precipitation product, which is promising for advancing hydrologic uses of satellite rainfall in mountainous areas lacking ground observations.

Corresponding author address: Emmanouil Anagnostou, Civil and Environmental Engineering, University of Connecticut, Unit 2037, Storrs, CT 06269. E-mail: manos@engr.uconn.edu

Abstract

Satellite-retrieved precipitation has the potential to support flood modeling in mountainous areas. However, to reach this potential satellite estimates need to be corrected for the severe underestimation exhibited in orography-induced heavy precipitation events (HPEs). This paper assesses an existing satellite precipitation error correction technique driven by high-resolution numerical weather prediction (NWP) simulations of HPEs in complex terrain. The study is based on NOAA Climate Prediction Center morphing technique (CMORPH) high-resolution precipitation estimates of six such events induced by hurricane landfalls in the southern Appalachian mountainous region. A distributed hydrological model (Coupled Routing and Excess Storage model) is applied to evaluate the impact of the proposed satellite precipitation error correction on flood simulations for 20 basins of various sizes in this mountainous region. The results demonstrate significant improvements due to the NWP-based adjustment technique in terms of both the precipitation error characteristics and corresponding runoff simulations. These improvements are shown to be comparable to those from the postprocessed gauge-adjusted CMORPH precipitation product, which is promising for advancing hydrologic uses of satellite rainfall in mountainous areas lacking ground observations.

Corresponding author address: Emmanouil Anagnostou, Civil and Environmental Engineering, University of Connecticut, Unit 2037, Storrs, CT 06269. E-mail: manos@engr.uconn.edu

1. Introduction

The quantification of heavy precipitation events (HPEs) over mountainous areas has been a challenge for all types of satellite products. Many past research studies have focused on the quantitative evaluation of satellite precipitation over different complex terrain regions including the Appalachian mountainous area (Prat and Barros 2010), the South American Andes (Dinku et al. 2010; Zulkafli et al. 2014), the Alps and Massif Central mountain range (Stampoulis et al. 2013), the western Black Sea region of Turkey (Derin and Yilmaz 2014), the Ethiopia highlands (Hirpa et al. 2010; Romilly and Gebremichael 2011), and the Tibetan Plateau (Gao and Liu 2013). These studies have mainly focused on three quasi-global satellite products: the National Oceanic and Atmospheric Administration (NOAA) Climate Prediction Center morphing technique (CMORPH), the Tropical Rainfall Measuring Mission (TRMM) Multisatellite Precipitation Analysis (TMPA), and the Precipitation Estimation from Remotely Sensed Information Using Artificial Neural Networks (PERSIANN). Major findings from these complex terrain error analyses include the following:

  1. The estimates are fairly accurate in capturing the precipitation spatial variability, but the quantification of rainfall exhibits strong underestimation of high rain rates and overestimation of light precipitation.
  2. The satellite products tend to underestimate rainfall values over regions with higher elevation. This fact has been discussed over different regions, including Nepal (Barros et al. 2000; Lang and Barros 2002; Barros and Tao 2008), Taiwan (Chen et al. 2013), Ethiopia (Dinku et al. 2008), and continental Europe (Stampoulis and Anagnostou 2012).
  3. The CMORPH and TMPA estimates exhibited lower bias than PERSIANN in most of the study regions, indicative of the fact that passive-microwave-based products (used in CMORPH and TMPA) can better represent precipitation processes than the infrared-based precipitation retrievals (PERSIANN).

Even though satellite precipitation products are strongly underestimating HPEs in mountainous areas, given their unrivalled advantage of spatial coverage over these data-poor regions of Earth, and the advent in precipitation remote sensing from the Global Precipitation Measurement satellite (Hou et al. 2014), there is great interest in advancing uses in hydrological applications. This entails the understanding of errors and investigation of correction techniques at different spatial and temporal scales, for example, from daily to monthly for deriving water budgets and subdaily for modeling floods and flash floods. Gauge-adjusted satellite products are usually considered as a more reliable data source for hydrological applications than the corresponding unadjusted estimates (Janowiak and Xie 1999; Pan et al. 2010; Mei et al. 2014). However, gauge adjustment requires a relatively dense gauge network and high-quality ground measurements (Wilk et al. 2006; Gourley et al. 2011). In most mountainous areas, lack of surface stations, or the weak characterization of spatial precipitation variability, challenges the reliability of satellite adjustment procedures. Research has even shown that in some cases gauge adjustment estimated with sparse gauge distributions may worsen the accuracy of satellite rainfall estimates (Bitew and Gebremichael 2011; Bitew et al. 2012).

To address this issue, Zhang et al. (2013) has proposed a bias-correction technique based solely on high-resolution numerical weather prediction (NWP) simulations of mountainous HPEs. The technique was demonstrated on the high-resolution (8 km and hourly) CMORPH precipitation product based on major flood-inducing HPEs over an Alpine region in northern Italy. The authors showed improved error statistics resulting from the NWP-adjusted CMORPH relative to the original CMORPH precipitation estimates by comparing with a high-resolution (1 km and hourly) gauge-adjusted radar rainfall product. It was argued that such a method can be extended over different data-poor mountainous regions to derive error corrections for high-resolution satellite products. In a recent study, Nikolopoulos et al. (2015) tested the above technique on three near-real-time remotely sensed precipitation datasets (two satellite datasets and a radar-only dataset) that severely underestimated the 2013 Colorado flood event (Gochis et al. 2015). They confirmed significant reduction of the precipitation underestimation for all examined remotely sensed precipitation products.

This study is built upon previous results demonstrating the feasibility of the NWP-based satellite precipitation adjustment technique for a different type of HPE (i.e., Atlantic tropical cyclone) and evaluating impacts on flood simulation. Specifically, the study is based on heavy precipitation–induced flooding events associated with six tropical cyclones over the southern Appalachian mountainous area. Like in the previous studies, we focused on CMORPH precipitation because of its high spatiotemporal resolution (8 km and hourly) and the stronger correlation it exhibits in terms of precipitation patterns relative to other high-resolution products (Zeweldi and Gebremichael 2009). Runoff simulations for 20 medium- to large-sized basins within the study area were conducted with a regional distributed hydrologic model currently used at NOAA/NSSL for issuing flash floods within the continental United States (http://flash.ou.edu).

This paper is organized as follows. Section 2 describes the study area and precipitation datasets. Section 3 presents the NWP and hydrological model setups for the study area and storm events. The methodology is explained in section 4, and results based on the six hurricane cases are shown in section 5. Section 6 provides the conclusions and discussion.

2. Study area and data

a. Study area

The study area (Fig. 1) is centered in the southern Appalachians and spans into the Piedmont and Coastal Plain regions of North Carolina. The region is located between 33° and 38°N latitude and 78° and 86°W longitude. It is characterized by complex mountainous terrain in the upper reaches with average annual precipitation ranging between 1200 and 1500 mm. Twenty medium- to large-sized basins (shown in Table 1 and Fig. 1) were used to evaluate the error propagation on flood simulations. The basin areas range between 5847 and 64 395 km2.

Fig. 1.
Fig. 1.

Study area with 20 basins. (Elevation data source: Aster Global DEM; https://asterweb.jpl.nasa.gov/gdem.asp.) The map also depicts WRF’s inner domain at 2-km resolution (390 × 324 grids).

Citation: Journal of Hydrometeorology 17, 4; 10.1175/JHM-D-15-0088.1

Table 1.

Case study basins information.

Table 1.

Six past hurricane landfall events were selected as case studies of this paper: Hurricanes Bill, Gaston, Frances, Ivan, Cindy, and Fay. Table 2 summarizes the period of each event and the associated rainfall characteristics (peak rainfall rate and rain accumulation) over the study region.

Table 2.

Hurricane landfall precipitation events used in the study.

Table 2.

b. Precipitation data

There are four precipitation datasets involved in the study: 1) high-resolution (8 km and 30 min) CMORPH rainfall product (ftp://ftp.cpc.ncep.noaa.gov/precip/CMORPH_V1.0/; Joyce et al. 2004), 2) gauge-adjusted CMORPH rainfall product (Xie et al. 2011), 3) 2 km and hourly rainfall from Weather Research and Forecasting (WRF) Model simulations (Skamarock et al. 2008) with initial and boundary conditions derived from the 0.5° Global Forecast System (GFS) analysis (the WRF Model setup is explained in section 3a), and 4) 4 km and hourly gauge-corrected stage IV WSR-88D precipitation data (http://data.eol.ucar.edu/codiac/dss/id=21.093; Fulton et al. 1998; Lin and Mitchell 2005). Stage IV precipitation data are considered as the reference for evaluating the CMORPH and WRF rainfall estimations. To apply a common data comparison, all datasets were scaled and projected into the CMORPH (stage IV) rainfall spatial (temporal) resolutions, that is, 8 km and hourly.

c. Runoff data

Hourly runoff data were simulated by the Coupled Routing and Excess Storage (CREST) distributed hydrological model (Wang et al. 2011) using the above precipitation datasets. Information on CREST model setup is provided in section 3b.

3. Model setup

a. NWP model

The NWP rainfall simulations are provided by WRF Model, version 3.4 (NCAR 2012). The WRF simulations of hurricane events are completed in a two-way interactive mode with three-domain nesting (18-, 6-, and 2-km resolution) and 27 vertical level configurations. WRF’s 2-km inner domain extends to 390 × 324 grids and fully covers the southern Appalachian area (Fig. 1). To cover the entire period of rainfall from the hurricanes, each WRF simulation lasted 72 h—that is, simulations started one day before the hurricane landfall day as a warm-up period and ended the day after. Model output files were recorded at hourly time intervals. The rainfall used in this study comes from the entire simulation because there is almost no rain during the warm-up period.

The selection of WRF parameterization schemes is shown in Table 3. The Thompson et al. (2006) bulk parameterization scheme was used to describe the microphysical processes. This scheme uses parameters determining autoconversion rate calculated by presetting cloud water droplet concentration. The rainfall from WRF simulations with the current parameterization have been compared to stage IV by quantile–quantile plots for the six hurricane events (Fig. 2), and general event characteristics are summarized in Table 2. Results show that rainfall magnitudes simulated by WRF are consistent with the stage IV values, which is the prerequisite of using these estimates to guide the adjustment procedure of CMORPH precipitation.

Table 3.

WRF Model setup.

Table 3.
Fig. 2.
Fig. 2.

Quantile–quantile plot of WRF vs stage IV hourly rain rates for the six hurricane events.

Citation: Journal of Hydrometeorology 17, 4; 10.1175/JHM-D-15-0088.1

b. Hydrological model setup

An implementation of the CREST distributed hydrological model over the conterminous United States (CONUS) was employed for the flood simulations of this paper. The water balance component of CREST consists of a variable infiltration curve and a conceptual mechanism for surface and subsurface partitioning of excess rainfall. In this version of CREST, the subsurface portion of excess rainfall is routed with a linear reservoir model, while the surface portion is routed with the kinematic wave approximation of the Saint-Venant equations of one-dimensional open channel flow (Chow et al. 1988). CREST is a model that represents the spatiotemporal variation of water fluxes and storages on a regular grid. An important feature of this model is its versatility for working on different user-defined spatial and temporal scales, which enables multiscale applications. CREST can be easily configured for various forcing data, which facilitated the analysis in this study.

CREST was implemented herein with the configuration employed in the Flooded Locations and Simulated Hydrographs Project (FLASH; http://nssl.noaa.gov/projects/flash/) for its real-time flash flood monitoring system (http://flash.ou.edu). It is configured on a 1-km grid over CONUS. Given the scale of the analysis in this study, the model is integrated using a time step of 1 h. Model parameters were estimated through an a priori approach using raster-based data from soil datasets, land cover, and digital elevation model derivatives (Vergara et al. 2015, manuscript submitted to J. Hydrol.). The use of a priori estimates can reduce uncertainty in model simulations (Koren et al. 2003) and enables an unbiased comparison of multiple QPE products (Vergara et al. 2014).

4. Methodology

a. Adjustment procedure

In this paper, we follow the adjustment technique described in Zhang et al. (2013). Specifically, the original CMORPH rainfall rates were adjusted using a power-law function [Eq. (1)] between CMORPH and WRF precipitation rainfall rates derived for each storm event:
e1
where X and Y correspond to the original CMORPH and WRF hourly rain rates, respectively. The parameters of the power-law relationship are determined using quantile values of the original CMORPH and WRF rainfall rates derived for different cumulative probability values (i.e., 0.05, 0.1, 0.15, …, 0.95). The least squares method was used to fit the power-law function to the CMORPH–WRF quantile–quantile data of each hurricane event. The power-law function with optimal parameter values was then applied on the original CMORPH rain rates to produce the WRF-adjusted CMORPH rainfall rates.

b. Precipitation error evaluation method

The evaluation is conducted for two temporal scales (storm-length period and hourly), and two spatial scales, that is, satellite product resolution over the entire study domain and for basin-average storm-total accumulation values.

We first verified the power-law relationship derived based on CMORPH–WRF quantiles against the power-law relationships derived using the reference stage IV hourly rain rates.

Subsequently, we evaluated the hourly rain rates of four precipitation products over the study domain (i.e., WRF simulations, original CMORPH, WRF-adjusted CMORPH, and gauge-adjusted CMORPH) against the reference rainfall data from stage IV using 1) qualitative comparison of event-total rainfall accumulation maps and 2) quantitative comparisons based on two statistical error metrics: bias ratio score (BS) and Heidke skill score (HSS; Heidke 1926).
e2
e3
where A, B, C, and D are the following occurrences, determined based on five different rain-rate thresholds (1, 2, 4, 8, and 12 mm h−1):
  • A = estimator > threshold and stage IV > threshold;
  • B = estimator > threshold and stage IV < threshold;
  • C = estimator < threshold and stage IV > threshold; and
  • D = estimator < threshold and stage IV < threshold.
A BS of 1 is considered as an unbiased estimation, while above or below 1 represents overestimation or underestimation, respectively. The HSS metric tests the occurrences of exceeding or failing to reach a certain rain-rate threshold; its values range from −∞ to 1, where 1 indicates a perfect estimation and less than or equal to zero indicates a random estimation.

Note that although WRF-based adjustment technique is practical in terms of modifying the CMORPH precipitation magnitude, it cannot improve the spatial rainfall patterns or rainfall areas. Therefore, the degree of improvement from this technique largely depends on the quality of spatial rainfall patterns from the original satellite precipitation product.

c. Hydrological impact analysis

As mentioned above we have identified 20 basins within the study domain to evaluate the impact of the WRF-based CMORPH precipitation error adjustment in terms of basin hydrology and flood simulations (Fig. 1). Specifically, we used the stage IV reference rainfall, original CMORPH, WRF-adjusted CMORPH, and gauge-adjusted CMORPH hourly rain rates to derive hourly basin-average precipitation. These precipitation datasets were then used as input in the CREST model to simulate basin outlet runoff. Evaluation at basin scale is based on scatterplots of basin-average precipitation accumulation, peak runoff simulations, and accumulated runoff between the various products and reference (stage IV). The events’ mean RMSE ratio of accumulated rainfall and runoff are also evaluated to compare the performance of WRF-adjusted CMORPH datasets relative to the gauge-adjusted CMORPH datasets.

Three statistical metrics, namely, bias ratio, Pearson correlation, and central normalized root-mean-square error [NRMSE; Eq. (4)], determined for storm-total rainfall accumulation, peak runoff, and accumulated runoff values, are used to demonstrate the error structure of basin-scale precipitation and runoff. The NRMSE metric definition is shown below:
e4
The NRMSE represents the relative (to the reference mean rainfall) random error component variability.

5. Results

a. CMORPH precipitation adjustment

An overview of the accumulated precipitation maps (Fig. 3) show similar spatial rainfall patterns between stage IV, original CMORPH, and WRF data for all hurricane events. However, significant magnitude differences (primarily underestimation) are exhibited for the original CMORPH rainfall relative to the reference stage IV rainfall. By contrast, WRF-simulated rainfall exhibits overestimation for most cases except Hurricane Gaston (slightly underestimated), where the main rainband is not located at the mountain range but at the eastern side of Appalachians.

Fig. 3.
Fig. 3.

Event-total precipitation accumulation maps from stage IV, WRF, original CMORPH, WRF-adjusted CMORPH, and gauge-adjusted CMORPH, for hurricanes (a)–(e) Bill, (f)–(j) Gaston, (k)–(o) Frances, (p)–(t) Ivan, (u)–(y) Cindy, and (z)–(δ) Fay.

Citation: Journal of Hydrometeorology 17, 4; 10.1175/JHM-D-15-0088.1

The quantile–quantile plots of Fig. 4 for the quantile ranges of 0.05–0.95 show an approximate power-law relationship between WRF and CMORPH, which is in close agreement with the relationship derived between stage IV and CMORPH. This agreement supports the argument that WRF simulations can be used as proxy to derive the power-law adjustment relationship parameters for CMORPH when ground reference is lacking. Furthermore, it should be noted that for all hurricane events, the power-law lines are enclosed in a relatively narrow range in which the CMORPH–WRF relationships tend to have slightly steeper slopes than the CMORPH–stage IV relationships, which is attributed to the WRF overestimation shown in Figs. 2 and 4. The differences between WRF and stage IV are more significant for low rain rates (<8 mm h−1) than for high rain rates.

Fig. 4.
Fig. 4.

Quantile–quantile plots of original CMORPH vs WRF and stage IV hourly rain rates for the six hurricane events.

Citation: Journal of Hydrometeorology 17, 4; 10.1175/JHM-D-15-0088.1

As shown by the maps of Fig. 3, the WRF-adjusted CMORPH precipitation exhibits better consistency with the reference (stage IV) than the original CMORPH product. A point to note is that WRF-adjusted and gauge-adjusted CMORPH estimates seem to perform similarly in terms of rainfall accumulations for the six hurricane cases examined in this study. However, the gauge-adjusted CMORPH uses gauges that were also used in the stage IV radar rainfall adjustment, while the WRF-based adjustment is independent of any ground rainfall measurement.

Next we provide quantitative error analysis (Fig. 5) based on the error metrics of Eqs. (2) and (3). As shown, the BS of the original CMORPH decreases sharply from a slight overestimation in low rainfall threshold (<2 mm h−1) to significant underestimation at high rainfall rate thresholds (>8 mm h−1). On the other hand, the WRF simulations tend to bias positively rainfall with overestimation ranging from moderate in low thresholds (<4 mm h−1) to high in thresholds exceeding 8 mm h−1. The WRF-adjusted CMORPH exhibits a more consistent BS (~1) and less dependence on rainfall magnitude. The gauge-adjusted CMORPH also exhibits improvements relative to the original CMORPH BS statistic, but it still has a strong magnitude-dependent bias. For example, BS values of gauge-adjusted CMORPH are around 0.5 for the 12 mm h−1 threshold, while the corresponding WRF-adjusted CMORPH BS values are around 1. Evidently, WRF-adjusted CMORPH is the best estimation among the four estimators for these events. In terms of the HSS error metric (Fig. 6), we show that the original CMORPH and WRF data exhibit lower scores than the two adjusted CMORPH estimates, especially at thresholds exceeding 8 mm h−1. This indicates that adjustment not only reduces the bias score, but also improves the random component of the error. Comparison of the WRF-adjusted to the gauge-adjusted CMORPH error statistics shows similar level for all threshold rainfall values.

Fig. 5.
Fig. 5.

BS of (left) original CMORPH and WRF and (right) WRF-adjusted and gauge-adjusted CMORPH vs rain-rate threshold.

Citation: Journal of Hydrometeorology 17, 4; 10.1175/JHM-D-15-0088.1

Fig. 6.
Fig. 6.

As in Fig. 5, but for the HSS metric.

Citation: Journal of Hydrometeorology 17, 4; 10.1175/JHM-D-15-0088.1

In terms of the storm-total precipitation accumulation at basin scale, the two adjusted CMORPH datasets also perform much better than the original CMORPH. Figure 7 shows the scatterplot of storm-total rainfall for the six hurricane events over the 20 basins in our study area. The original CMORPH tends to underestimate the basin-average rainfall accumulation in all cases. Arguably, the WRF-adjusted CMORPH effectively moderates this underestimation. Qualitatively, the scatter shown in Fig. 7 for the gauge-adjusted CMORPH data is better than the WRF-adjusted CMORPH, but as stated earlier gauge-adjustment in CMORPH is based on the same gauges used in stage IV, while WRF-adjusted CMORPH is independent of ground reference data.

Fig. 7.
Fig. 7.

Scatterplots of stage IV vs estimated (original, WRF-adjusted, and gauge-adjusted CMORPH) basin-average accumulated rainfall.

Citation: Journal of Hydrometeorology 17, 4; 10.1175/JHM-D-15-0088.1

Table 4 shows quantitative error statistics for the basin-average rainfall accumulations. Results indicate that WRF-adjusted CMORPH rainfall values exhibit improved bias ratios and normalized RMSE values relative to the original CMORPH data. In terms of correlation, it is shown that the WRF-based adjustment tends to slightly reduce the score in rainfall. Consistent to the scatterplot of Fig. 7, the gauge-adjusted CMORPH exhibits better error scores than the WRF-adjusted CMORPH. This aspect is also apparent in the plot of event rainfall accumulation NRMSE ratios (Fig. 10a; described in greater detail below). Specifically, the NRMSE ratios of gauge-adjusted CMORPH to original CMORPH are lower than the NRMSE ratios of WRF-adjusted CMORPH to original CMORPH for all events except Hurricane Bill. It is also worth noting that the gauge-adjusted CMORPH data provide much lower NRMSE values than the original CMORPH for all events, while the WRF-based adjustment does not reduce the CMORPH estimates NRMSE as consistently as the gauge-adjustment.

Table 4.

Basin-scale statistics of the different CMORPH estimates vs stage IV (statistics are based on six events over the 20 basins).

Table 4.

In summary, the WRF-based adjustment is effective in reducing the systematic and magnitude-dependent error, but it is not as efficient in improving the random error component as exhibited by the central NRMSE and correlation statistics. Three events out of the six used in this study exhibited similar NRMSE values as the original CMORPH rainfall (Fig. 10a; described in greater detail below); in two events, though, we showed significant reduction, while in one event WRF-based adjustment worsened the NRMSE. This probably stems from inaccuracies in WRF simulations of high rainfall rates.

b. Hydrological impacts

Results on hydrological impacts are summarized in Figs. 8 and 9. Overall, this analysis shows that the WRF-based adjustment improves the accuracy of CMORPH rainfall–driven hydrological simulations. Specifically, we note close agreement between the WRF-adjusted CMORPH rainfall–driven CREST simulations and the reference runoff simulations driven by stage IV data. An example of runoff simulations for basin B6 based on the various precipitation-forcing datasets is shown in Fig. 8. We note significant reduction of the underestimation of original CMORPH-driven runoff simulations because of the two adjustment procedures (gauge based and WRF based). However, the time series runoff results do not demonstrate any clear preference between the two datasets. Consistent with the time series plots, the scatterplots of basin outlet runoff (Fig. 9) show that the original CMORPH-derived runoff exhibits strong underestimation. On the other hand, both WRF-adjusted and gauge-adjusted CMORPH-derived runoff have values closer to those derived from stage IV-driven simulations. As in the basin-scale rainfall error analysis (Fig. 7), the WRF-adjusted CMORPH reflects slightly higher biases than the gauge-adjusted CMORPH.

Fig. 8.
Fig. 8.

Basin B6 runoff time series from CREST simulations.

Citation: Journal of Hydrometeorology 17, 4; 10.1175/JHM-D-15-0088.1

Fig. 9.
Fig. 9.

Scatterplots of stage IV vs estimated rainfall-driven simulated (top) basin outlet peak runoff and (bottom) basin outlet accumulated runoff.

Citation: Journal of Hydrometeorology 17, 4; 10.1175/JHM-D-15-0088.1

Table 4 summarizes bulk statistics for the runoff simulations. As shown in the results of Table 4, the WRF-adjusted CMORPH runoff performs better than the original CMORPH in terms of all error scores. The bias ratios of WRF-adjusted CMORPH runoff are close to 1 for both peak runoff and accumulated runoff values. WRF-adjusted CMORPH runoff also has higher correlation and lower NRMSE values than the original CMORPH. Comparing results to the rainfall error statistics, the hydrological simulations moderate the differences of WRF-adjusted CMORPH to the gauge-adjusted CMORPH data. Figure 10b shows a comparison between WRF-adjusted and gauge-adjusted CMORPH datasets. Both datasets reduce the runoff NRMSE values of the original CMORPH; gauge adjustment provides a marginally better correction than the WRF-based correction. However, the point of pursuing this study is that the gauge adjustment is not a scenario that is always feasible, particularly when considering mountainous areas with limited in situ observations. Therefore, the close similarity using WRF-adjusted CMORPH is a promising approach to improving the hydrologic use of satellite rainfall in global data-poor mountainous areas.

Fig. 10.
Fig. 10.

The NRMSE ratio of (a) accumulated rainfall and (b) accumulated runoff for each event.

Citation: Journal of Hydrometeorology 17, 4; 10.1175/JHM-D-15-0088.1

6. Conclusions and discussion

The study assessed the performance of NWP-based CMORPH adjustment at high spatiotemporal resolution based on WRF simulations of six hurricane landfall events in the southern Appalachians region. The error analysis was based on two aspects: 1) the evaluation of adjusted CMORPH precipitation error properties across the study domain and at basin scale and 2) the hydrologic evaluation of the adjusted products in terms of rainfall and flood simulations derived from a distributed hydrological model. The major findings can be summarized as follows.

Original CMORPH data are likely to provide similar rainfall patterns as the reference radar rainfall (stage IV) dataset, but with a strong magnitude-dependent systematic error. WRF simulations, on the other hand, overestimated rainfall magnitude, especially the higher rain rates. The WRF-adjusted CMORPH rainfall based on the Zhang et al. (2013) technique was shown to provide a significantly improved estimate. The technique could effectively moderate the magnitude-dependent systematic error; it particularly reduced the strong underestimation of high rain rates apparent in the original CMORPH estimates. Furthermore, WRF-adjusted CMORPH rain rates exhibited improved correlation scores relative to the original CMORPH as demonstrated by the HSS error metric.

Based on the entire study domain error analysis, the WRF-based adjustment on CMORPH performed similarly to the postprocessing gauge adjustment. However, the gauge-adjusted CMORPH product is not available in real time, and its accuracy is prone to the density and availability of surface observations. On the other hand, WRF-adjusted CMORPH rainfall can be applied for any mountainous region on Earth based on NWP analysis or forecasts once the CMORPH product is available (usually 12–18 h after the observation time). Note that the current WRF-adjustment method will have limited success when the satellite product has 1) rain detection errors (both false detection and missing rainfall areas) and 2) varying error characteristics according to geographic and storm development stages. These aspects will require extensions of the technique that is the subject of future research.

The basin outlet runoff derived from WRF-adjusted CMORPH rainfall exhibited improved statistics relative to the ones derived from original CMORPH rainfall fields. In general, the gauge-adjusted CMORPH product performed better in terms of the hydrological analysis (i.e., basin-average rainfall and flood simulations). Although the flash flood prediction based on WRF-adjusted CMORPH data does not show consistently superior performance relative to the gauge-adjusted product, this adjustment method presents an innovation with potential to advance real-time flash flood forecasting by satellite precipitation, especially in regions that are ungauged, where the community is likely to get most societal benefit.

The current study used site-specific power-law parameter values for each rainfall event, meaning that WRF analysis had to be performed for each event to apply the method, which may not be considered an efficient way to correct satellite precipitation data at the global scale. Furthermore, application of the proposed technique on other precipitation products, particularly the newly available multiagency high-resolution precipitation dataset [Integrated Multisatellite Retrievals for GPM (IMERG)], is needed. Future continuation of this work would focus on different WRF postprocessing methods to improve WRF rainfall prediction (Mendoza et al. 2015) and the investigation of global applicability of this power-law technique over a variety of mountainous areas, including the Andes (South America), the upper Blue Nile (East Africa), the Olympic Peninsula (northwestern United States), and Taiwan.

Acknowledgments

This work was supported by NASA Precipitation Measurement Mission Award NNX07AE31G. We acknowledge and appreciate valuable input from three anonymous reviewers.

REFERENCES

  • Barros, A. P., , and Tao K. , 2008: A space-filling algorithm to extrapolate narrow-swath instantaneous TRMM microwave rain-rate estimates using thermal IR imagery. J. Atmos. Oceanic Technol., 25, 19011920, doi:10.1175/2008JTECHA1019.1.

    • Search Google Scholar
    • Export Citation
  • Barros, A. P., , Joshi M. , , Putkonen J. , , and Burbank D. W. , 2000: A study of the 1999 monsoon rainfall in a mountainous region in central Nepal using TRMM products and rain gauge observations. Geophys. Res. Lett., 27, 36833686, doi:10.1029/2000GL011827.

    • Search Google Scholar
    • Export Citation
  • Bitew, M. M., , and Gebremichael M. , 2011: Assessment of satellite rainfall products for streamflow simulation in medium watersheds of the Ethiopian highlands. Hydrol. Earth Syst. Sci., 15, 11471155, doi:10.5194/hess-15-1147-2011.

    • Search Google Scholar
    • Export Citation
  • Bitew, M. M., , Gebremichael M. , , Ghebremichael L. T. , , and Bayissa Y. A. , 2012: Evaluation of high-resolution satellite rainfall products through streamflow simulation in a hydrological modeling of a small mountainous watershed in Ethiopia. J. Hydrometeor., 13, 338350, doi:10.1175/2011JHM1292.1.

    • Search Google Scholar
    • Export Citation
  • Chen, S., and et al. , 2013: Performance evaluation of radar and satellite rainfalls for Typhoon Morakot over Taiwan: Are remote-sensing products ready for gauge denial scenario of extreme events? J. Hydrol., 506, 413, doi:10.1016/j.jhydrol.2012.12.026.

    • Search Google Scholar
    • Export Citation
  • Chow, V. T., , Maidment D. R. , , and Mays L. W. , 1988: Applied Hydrology. McGraw-Hill, New York, 572 pp.

  • Derin, Y., , and Yilmaz K. K. , 2014: Evaluation of multiple satellite-based precipitation products over complex topography. J. Hydrometeor., 15, 14981516, doi:10.1175/JHM-D-13-0191.1.

    • Search Google Scholar
    • Export Citation
  • Dinku, T., , Chidzambwa S. , , Ceccato P. , , Connor S. J. , , and Ropelewski C. F. , 2008: Validation of high‐resolution satellite rainfall products over complex terrain. Int. J. Remote Sens., 29, 40974110, doi:10.1080/01431160701772526.

    • Search Google Scholar
    • Export Citation
  • Dinku, T., , Ruiz F. , , Connor S. J. , , and Ceccato P. , 2010: Validation and intercomparison of satellite rainfall estimates over Colombia. J. Appl. Meteor. Climatol., 49, 10041014, doi:10.1175/2009JAMC2260.1.

    • Search Google Scholar
    • Export Citation
  • Fulton, R. A., , Breidenbach J. P. , , Seo D. J. , , Miller D. A. , , and Miller T. , 1998: The WSR-88D rainfall algorithm. Wea. Forecasting, 13, 377395, doi:10.1175/1520-0434(1998)013<0377:TWRA>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Gao, Y. C., , and Liu M. F. , 2013: Evaluation of high-resolution satellite precipitation products using rain gauge observations over the Tibetan Plateau. Hydrol. Earth Syst. Sci., 17, 837, doi:10.5194/hess-17-837-2013.

    • Search Google Scholar
    • Export Citation
  • Gochis, D., and et al. , 2015: The great Colorado flood of September 2013. Bull. Amer. Meteor. Soc., 96, 14611487, doi:10.1175/BAMS-D-13-00241.1.

    • Search Google Scholar
    • Export Citation
  • Gourley, J. J., , Hong Y. , , Flamig Z. L. , , Wang J. , , Vergara H. , , and Anagnostou E. N. , 2011: Hydrologic evaluation of rainfall estimates from radar, satellite, gauge, and combinations on Ft. Cobb basin, Oklahoma. J. Hydrometeor., 12, 973988, doi:10.1175/2011JHM1287.1.

    • Search Google Scholar
    • Export Citation
  • Heidke, P., 1926: Berechnung des Erfolges und der Güte der Windstärkevorhersagen im Sturmwarnungsdienst. Geogr. Ann., 8, 301349.

  • Hirpa, F. A., , Gebremichael M. , , and Hopson T. , 2010: Evaluation of high-resolution satellite precipitation products over very complex terrain in Ethiopia. J. Appl. Meteor. Climatol., 49, 10441051, doi:10.1175/2009JAMC2298.1.

    • Search Google Scholar
    • Export Citation
  • Hou, A. Y., and et al. , 2014: The Global Precipitation Measurement Mission. Bull. Amer. Meteor. Soc., 95, 701722, doi:10.1175/BAMS-D-13-00164.1.

    • Search Google Scholar
    • Export Citation
  • Janowiak, J. E., , and Xie P. , 1999: CAMS–OPI: A global satellite–rain gauge merged product for real-time precipitation monitoring applications. J. Climate, 12, 33353342, doi:10.1175/1520-0442(1999)012<3335:COAGSR>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Joyce, R. J., , Janowiak J. E. , , Arkin P. A. , , and Xie P. , 2004: CMORPH: A method that produces global precipitation estimates from passive microwave and infrared data at high spatial and temporal resolution. J. Hydrometeor., 5, 487503, doi:10.1175/1525-7541(2004)005<0487:CAMTPG>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Koren, V., , Smith M. , , and Duan Q. , 2003: Use of a priori parameter estimates in the derivation of spatially consistent parameter sets of rainfall–runoff models. Calibration of Watershed Models, Water Science and Application Series, Vol. 6, Amer. Geophys. Union, 239–254.

  • Lang, T. J., , and Barros A. P. , 2002: An investigation of the onsets of the 1999 and 2000 monsoons in central Nepal. Mon. Wea. Rev., 130, 12991316, doi:10.1175/1520-0493(2002)130<1299:AIOTOO>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Lin, Y., , and Mitchell K. E. , 2005: The NCEP stage II/IV hourly precipitation analyses: Development and applications. 19th Conf. on Hydrology, San Diego, CA, Amer. Meteor. Soc., 1.2. [Available online at http://www.emc.ncep.noaa.gov/mmb/ylin/pcpanl/refs/stage2-4.19hydro.pdf.]

  • Mei, Y., , Anagnostou E. N. , , Nikolopoulos E. I. , , and Borga M. , 2014: Error analysis of satellite precipitation products in mountainous basins. J. Hydrometeor., 15, 17781793, doi:10.1175/JHM-D-13-0194.1.

    • Search Google Scholar
    • Export Citation
  • Mendoza, P. A., , Rajagopalan B. , , Clark M. P. , , Ikeda K. , , and Rasmussen R. M. , 2015: Statistical postprocessing of high-resolution regional climate model output. Mon. Wea. Rev., 143, 15331553, doi:10.1175/MWR-D-14-00159.1.

    • Search Google Scholar
    • Export Citation
  • NCAR, 2012: Weather Research and Forecasting ARW: Version 3 modeling system user’s guide. NCAR, 384 pp.[Available online at http://www2.mmm.ucar.edu/wrf/users/docs/user_guide_V3.4/ARWUsersGuideV3.pdf.]

  • Nikolopoulos, E. I., , Bartsotas N. , , Anagnostou E. N. , , and Kallos G. , 2015: Using high-resolution numerical weather forecasts to improve remotely sensed rainfall estimates: The case of the 2013 Colorado flash flood. J. Hydrometeor., 16, 17421751, doi:10.1175/JHM-D-14-0207.1.

    • Search Google Scholar
    • Export Citation
  • Pan, M., , Li H. , , and Wood E. , 2010: Assessing the skill of satellite‐based precipitation estimates in hydrologic applications. Water Resour. Res., 46, W09535, doi:10.1029/2009WR008290.

    • Search Google Scholar
    • Export Citation
  • Prat, O. P., , and Barros A. P. , 2010: Assessing satellite-based precipitation estimates in the southern Appalachian mountains using rain gauges and TRMM PR. Adv. Geosci., 25, 143153, doi:10.5194/adgeo-25-143-2010.

    • Search Google Scholar
    • Export Citation
  • Romilly, T. G., , and Gebremichael M. , 2011: Evaluation of satellite rainfall estimates over Ethiopian river basins. Hydrol. Earth Syst. Sci., 15, 15051514, doi:10.5194/hess-15-1505-2011.

    • Search Google Scholar
    • Export Citation
  • Skamarock, W. C., and et al. , 2008: A description of the Advanced Research WRF version 3. NCAR Tech. Note NCAR/TN-475+STR, 113 pp., doi:10.5065/D68S4MVH.

  • Stampoulis, D., , and Anagnostou E. N. , 2012: Evaluation of global satellite rainfall products over continental Europe. J. Hydrometeor., 13, 588603, doi:10.1175/JHM-D-11-086.1.

    • Search Google Scholar
    • Export Citation
  • Stampoulis, D., , Anagnostou E. N. , , and Nikolopoulos E. I. , 2013: Assessment of high-resolution satellite-based rainfall estimates over the Mediterranean during heavy precipitation events. J. Hydrometeor., 14, 15001514, doi:10.1175/JHM-D-12-0167.1.

    • Search Google Scholar
    • Export Citation
  • Thompson, G., , Field P. R. , , Hall W. D. , , and Rasmussen R. M. , 2006: A new bulk microphysical parameterization for WRF (& MM5). Proc. Seventh Weather Research and Forecasting Model Workshop, Boulder, CO, NCAR, 11 pp. [Available online at http://www2.mmm.ucar.edu/wrf/users/workshops/WS2006/abstracts/Session05/5_3_Thompson.pdf.]

  • Vergara, H., , Hong Y. , , Gourley J. J. , , Anagnostou E. N. , , Maggioni V. , , Stampoulis D. , , and Kirstetter P. E. , 2014: Effects of resolution of satellite-based rainfall estimates on hydrologic modeling skill at different scales. J. Hydrometeor., 15, 593613, doi:10.1175/JHM-D-12-0113.1.

    • Search Google Scholar
    • Export Citation
  • Wang, J., and et al. , 2011: The Coupled Routing and Excess Storage (CREST) distributed hydrological model. Hydrol. Sci. J., 56, 8498, doi:10.1080/02626667.2010.543087.

    • Search Google Scholar
    • Export Citation
  • Wilk, J., , Kniveton D. , , Andersson L. , , Layberry R. , , Todd M. C. , , Hughes D. , , Ringrose S. , , and Vanderpost C. , 2006: Estimating rainfall and water balance over the Okavango River basin for hydrological applications. J. Hydrol., 331, 1829, doi:10.1016/j.jhydrol.2006.04.049.

    • Search Google Scholar
    • Export Citation
  • Xie, P., , Yoo S. H. , , Joyce R. , , and Yarosh Y. , 2011: Bias-corrected CMORPH: A 13-year analysis of high-resolution global precipitation. Geophysical Research Abstracts, Vol. 13, Abstract EGU2011-1809. [Available online at http://meetingorganizer.copernicus.org/EGU2011/EGU2011-1809.pdf.]

  • Zeweldi, D. A., , and Gebremichael M. , 2009: Evaluation of CMORPH precipitation products at fine space–time scales. J. Hydrometeor., 10, 300307, doi:10.1175/2008JHM1041.1.

    • Search Google Scholar
    • Export Citation
  • Zhang, X., , Anagnostou E. N. , , Frediani M. , , Solomos S. , , and Kallos G. , 2013: Using NWP simulations in satellite rainfall estimation of heavy precipitation events over mountainous areas. J. Hydrometeor., 14, 18441858, doi:10.1175/JHM-D-12-0174.1.

    • Search Google Scholar
    • Export Citation
  • Zulkafli, Z., , Buytaert W. , , Onof C. , , Manz B. , , Tarnavsky E. , , Lavado W. , , and Guyot J. L. , 2014: A comparative performance analysis of TRMM 3B42 (TMPA) versions 6 and 7 for hydrological applications over Andean–Amazon river basins. J. Hydrometeor., 15, 581592, doi:10.1175/JHM-D-13-094.1.

    • Search Google Scholar
    • Export Citation
Save