• Arnault, J., S. Wagner, T. Rummler, B. Fersch, J. Bliefernicht, S. Andresen, and H. Kunstmann, 2016: Role of runoff–infiltration partitioning and resolved overland flow on land–atmosphere feedbacks: A case study with the WRF-Hydro coupled modeling system for West Africa. J. Hydrometeor., 17, 14891516, https://doi.org/10.1175/JHM-D-15-0089.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bannister, R. N., 2017: A review of operational methods of variational and ensemble-variational data assimilation. Quart. J. Roy. Meteor. Soc., 143, 607633, https://doi.org/10.1002/qj.2982.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Barker, D., and et al. , 2012: The Weather Research and Forecasting Model’s Community Variational/Ensemble Data Assimilation System: WRFDA. Bull. Amer. Meteor. Soc., 93, 831843, https://doi.org/10.1175/BAMS-D-11-00167.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bosilovich, M. G., and W. Y. Sun, 1999: Numerical simulation of the 1993 Midwestern flood: Land–atmosphere interactions. J. Climate, 12, 14901505, https://doi.org/10.1175/1520-0442(1999)012<1490:NSOTMF>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bouttier, F., and G. Kelly, 2001: Observing-system experiments in the ECMWF 4D-Var data assimilation system. Quart. J. Roy. Meteor. Soc., 127, 14691488, https://doi.org/10.1002/qj.49712757419.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bukovsky, M. S., and D. J. Karoly, 2009: Precipitation simulations using WRF as a nested regional climate model. J. Appl. Meteor. Climatol., 48, 21522159, https://doi.org/10.1175/2009JAMC2186.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Chen, F., and J. Dudhia, 2001: Coupling an advanced land surface-hydrology model with the Penn State–NCAR MM5 modeling system. Part I: Model implementation and sensitivity. Mon. Wea. Rev., 129, 569585, https://doi.org/10.1175/1520-0493(2001)129<0569:CAALSH>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Courtier, P., J. N. Thépaut, and A. Hollingsworth, 1994: A strategy for operational implementation of 4D-Var, using an incremental approach. Quart. J. Roy. Meteor. Soc., 120, 13671387, https://doi.org/10.1002/qj.49712051912.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Cui, C. Y., and et al. , 2018: Soil moisture mapping from satellites: An intercomparison of SMAP, SMOS, FY3B, AMSR2, and ESA CCI over two dense network regions at different spatial scales. Remote Sens., 10, 33, https://doi.org/10.3390/rs10010033.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dagon, K., and D. P. Schrag, 2016: Exploring the effects of solar radiation management on water cycling in a coupled land–atmosphere model. J. Climate, 29, 26352650, https://doi.org/10.1175/JCLI-D-15-0472.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • DelSole, T., M. Zhao, P. A. Dirmeyer, and B. P. Kirtman, 2008: Empirical correction of a coupled land–atmosphere model. Mon. Wea. Rev., 136, 40634076, https://doi.org/10.1175/2008MWR2344.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dirmeyer, P. A., and M. Zhao, 2004: Flux replacement as a method to diagnose coupled land–atmosphere model feedback. J. Hydrometeor., 5, 10341048, https://doi.org/10.1175/JHM-384.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dudhia, J., 1989: Numerical study of convection observed during the winter monsoon experiment using mesoscale two-dimensional model. J. Atmos. Sci., 46, 30773107, https://doi.org/10.1175/1520-0469(1989)046<3077:NSOCOD>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ferguson, C. R., E. F. Wood, and R. K. Vinukollu, 2012: A global intercomparison of modeled and observed land–atmosphere coupling. J. Hydrometeor., 13, 749784, https://doi.org/10.1175/JHM-D-11-0119.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Gao, J. Q., M. Qiao, X. Qiu, Y. Zeng, H. Hua, and M. Adamu, 2018: Estimation of actual evapotranspiration distribution in the Huaihe River upstream basin based on the generalized complementary principle. Adv. Meteor., 2018, 2158168, https://doi.org/10.1155/2018/2158168.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Grell, G. A., and D. Devenyi, 2002: A generalized approach to parameterizing convection combining ensemble and data assimilation techniques. Geophys. Res. Lett., 29, 1693, https://doi.org/10.1029/2002GL015311.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hong, S.-Y., Y. Noh, and J. Dudhia, 2006: A new vertical diffusion package with an explicit treatment of entrainment processes. Mon. Wea. Rev., 134, 23182341, https://doi.org/10.1175/MWR3199.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Jasper, K., J. Gurtz, and H. Lang, 2002: Erratum to “Advanced flood forecasting in Alpine watersheds by coupling meteorological observations and forecasts with a distributed hydrological model.” J. Hydrol., 269, 253254, https://doi.org/10.1016/S0022-1694(02)00345-1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kim, Y., and G. L. Wang, 2007: Impact of initial soil moisture anomalies on subsequent precipitation over North America in the coupled land–atmosphere model CAM3–CLM3. J. Hydrometeor., 8, 513533, https://doi.org/10.1175/JHM611.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kleczek, M. A., G. J. Steeneveld, and A. A. M. Holtslag, 2014: Evaluation of the weather research and forecasting mesoscale model for GABLS3: Impact of boundary-layer schemes, boundary conditions and spin-up. Bound.-Layer Meteor., 152, 213243, https://doi.org/10.1007/s10546-014-9925-3.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Li, Q. L., W. C. Zhang, L. Yi, J. P. Liu, and H. Chen, 2018: Accuracy evaluation and comparison of GPM and TRMM precipitation product over Mainland China. Adv. Water Sci., 29, 303313, https://doi.org/10.14042/j.cnki.32.1309.2018.03.001.

    • Search Google Scholar
    • Export Citation
  • Li, Y. P., Z. Li, Z. Zhang, L. Chen, S. Kurkute, L. Scaff, and X. Pan, 2019: High-resolution regional climate modeling and projection over western Canada using a weather research forecasting model with a pseudo-global warming approach. Hydrol. Earth Syst. Sci., 23, 46354659, https://doi.org/10.5194/hess-23-4635-2019.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lim, K. S. S., and S. Y. Hong, 2010: Development of an effective double-moment cloud microphysics scheme with prognostic cloud condensation nuclei (CCN) for weather and climate models. Mon. Wea. Rev., 138, 15871612, https://doi.org/10.1175/2009MWR2968.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lin, L. F., A. M. Ebtehaj, R. L. Bras, A. N. Flores, and J. F. Wang, 2015: Dynamical precipitation downscaling for hydrologic applications using WRF 4D-var data assimilation: Implications for GPM era. J. Hydrometeor., 16, 811829, https://doi.org/10.1175/JHM-D-14-0042.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Liu, S., X. Mo, W. Zhao, V. Naeimi, D. Dai, C. Shu, and L. Mao, 2009: Temporal variation of soil moisture over the Wuding River basin assessed with an eco-hydrological model, in-situ observations and remote sensing. Hydrol. Earth Syst. Sci., 13, 13751398, https://doi.org/10.5194/hess-13-1375-2009.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Liu, Y. B., G. P. Wu, and C. Q. Ke, 2016: Hydrological Remote Sensing. Science Press, 96–108.

  • Liu, Z., 2016: Comparison of Integrated Multisatellite Retrievals for GPM (IMERG) and TRMM Multisatellite Precipitation Analysis (TMPA) monthly precipitation products: Initial results. J. Hydrometeor., 17, 777790, https://doi.org/10.1175/JHM-D-15-0068.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lorenc, A. C., 2003: Modelling of error covariances by 4D-Var data assimilation. Quart. J. Roy. Meteor. Soc., 129, 31673182, https://doi.org/10.1256/qj.02.131.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Maggioni, V., and C. Massari, 2018: On the performance of satellite precipitation products in riverine flood modeling: A review. J. Hydrol., 558, 214224, https://doi.org/10.1016/j.jhydrol.2018.01.039.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Mahfouf, J. F., P. Bauer, and V. Marecal, 2005: The assimilation of SSM/I and TMI rainfall rates in the ECMWF 4D-Var system. Quart. J. Roy. Meteor. Soc., 131, 437458, https://doi.org/10.1256/qj.04.17.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Mlawer, E. J., S. J. Taubman, P. D. Brown, M. J. Iacono, and S. A. Clough, 1997: Radiative transfer for inhomogeneous atmospheres: RRTM, a validated correlated-k model for the longwave. J. Geophys. Res., 102, 16 66316 682, https://doi.org/10.1029/97JD00237.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Pan, X. D., X. Li, G. D. Cheng, and Y. Hong, 2017: Effects of 4D-Var data assimilation using remote sensing precipitation products in a WRF Model over the complex terrain of an arid region river basin. Remote Sens., 9, 963, https://doi.org/10.3390/rs9090963.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Parrish, D. F., and J. C. Derber, 1992: The National Meteorological Center spectral statistical interpolation analysis system. Mon. Wea. Rev., 120, 17471763, https://doi.org/10.1175/1520-0493(1992)120<1747:TNMCSS>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Pei, L. S., N. Moore, S. Zhong, L. Luo, D. W. Hyndam, W. E. Heilman, and Z. Gao, 2014: WRF Model sensitivity to land surface model and cumulus parameterization under short-term climate extremes over the southern Great Plains of the United States. J. Climate, 27, 77037724, https://doi.org/10.1175/JCLI-D-14-00015.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Pennelly, C., G. Reuter, and T. Flesch, 2014: Verification of the WRF Model for simulating heavy precipitation in Alberta. Atmos. Res., 135–136, 172192, https://doi.org/10.1016/j.atmosres.2013.09.004.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Reichle, R. H., and et al. , 2017: Assessment of the SMAP Level-4 surface and root-zone soil moisture product using in situ measurements. J. Hydrometeor., 18, 26212645, https://doi.org/10.1175/JHM-D-17-0063.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Rogelis, M. C., and M. Werner, 2018: Streamflow forecasts from WRF precipitation for flood early warning in mountain tropical areas. Hydrol. Earth Syst. Sci., 22, 853870, https://doi.org/10.5194/hess-22-853-2018.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Santanello, J. A., S. V. Kumar, C. D. Peters-Lidard, K. Harrison, and S. J. Zhou, 2013: Impact of land model calibration on coupled land–atmosphere prediction. J. Hydrometeor., 14, 13731400, https://doi.org/10.1175/JHM-D-12-0127.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Santi, E., S. Paloscia, S. Pettinato, C. Notarnicola, L. Pasolli, and A. Pistocchi, 2013: Comparison between SAR soil moisture estimates and hydrological model simulations over the Scrivia test site. Remote Sens., 5, 49614976, https://doi.org/10.3390/rs5104961.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Sharifi, E., R. Steinacker, and B. Saghafian, 2016: Assessment of GPM-IMERG and other precipitation products against gauge data under different topographic and Climatic conditions in Iran: Preliminary results. Remote Sens., 8, 135, https://doi.org/10.3390/rs8020135.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Shrestha, D. L., D. E. Robertson, Q. J. Wang, T. C. Pagano, and H. A. P. Hapuarachchi, 2013: Evaluation of numerical weather prediction model precipitation forecasts for short-term streamflow forecasting purpose. Hydrol. Earth Syst. Sci., 17, 19131931, https://doi.org/10.5194/hess-17-1913-2013.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Smirnova, T. G., J. M. Brown, S. G. Benjamin, and J. S. Kenyon, 2016: Modifications to the Rapid Update Cycle Land Surface Model (RUC LSM) available in the Weather Research and Forecasting (WRF) Model. Mon. Wea. Rev., 144, 18511865, https://doi.org/10.1175/MWR-D-15-0198.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Srinivas, D., and D. V. B. Rao, 2014: Implications of vortex initialization and model spin-up in tropical cyclone prediction using Advanced Research Weather Research and Forecasting Model. Nat. Hazards, 73, 10431062, https://doi.org/10.1007/s11069-014-1125-4.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Srivastava, P. K., D. Han, M. A. Rico-Ramirez, P. O’Neill, T. Islam, M. Gupta, and Q. Dai, 2015: Performance evaluation of WRF-Noah Land surface model estimated soil moisture for hydrological application: Synergistic evaluation using SMOS retrieved soil moisture. J. Hydrol., 529, 200212, https://doi.org/10.1016/j.jhydrol.2015.07.041.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Sun, R. C., H. L. Yuan, and Y. Z. Yang, 2018: Using multiple satellite-gauge merged precipitation products ensemble for hydrologic uncertainty analysis over the Huaihe River basin. J. Hydrol., 566, 406420, https://doi.org/10.1016/j.jhydrol.2018.09.024.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Tian, J. Y., J. Liu, J. Wang, C. Li, F. Lu, and Z. Chu, 2017a: A spatio-temporal evaluation of the WRF physical parameterisations for numerical rainfall simulation in semi-humid and semi-arid catchments of Northern China. Atmos. Res., 191, 141155, https://doi.org/10.1016/j.atmosres.2017.03.012.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Tian, J. Y., J. Liu, D. Yan, C. Li, Z. Chu, and F. Yu, 2017b: An assimilation test of Doppler radar reflectivity and radial velocity from different height layers in improving the WRF rainfall forecasts. Atmos. Res., 198, 132144, https://doi.org/10.1016/j.atmosres.2017.08.004.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Veerse, F., and J. N. Thepaut, 1998: Multiple-truncation incremental approach for four-dimensional variational data assimilation. Quart. J. Roy. Meteor. Soc., 124, 18891908, https://doi.org/10.1002/qj.49712455006.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Verbunt, M., M. Zappa, J. Gurtz, and P. Kaufmann, 2006: Verification of a coupled hydrometeorological modelling approach for alpine tributaries in the Rhine basin. J. Hydrol., 324, 224238, https://doi.org/10.1016/j.jhydrol.2005.09.036.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wang, Z. L., R. D. Zhong, C. G. Lai, and J. C. Chen, 2017: Evaluation of the GPM IMERG satellite-based precipitation products and the hydrological utility. Atmos. Res., 196, 151163, https://doi.org/10.1016/j.atmosres.2017.06.020.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wu, L. Y., and J. Y. Zhang, 2013: Role of land–atmosphere coupling in summer droughts and floods over eastern China for the 1998 and 1999 cases. Chin. Sci. Bull., 58, 39783985, https://doi.org/10.1007/s11434-013-5855-6.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Xia, J., D. X. She, Y. Y. Zhang, and H. Du, 2012: Spatio-temporal trend and statistical distribution of extreme precipitation events in Huaihe River Basin during 1960–2009. J. Geogr. Sci., 22, 195208, https://doi.org/10.1007/s11442-012-0921-6.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Xiong, Z., S. Y. Wang, Z. M. Zeng, and C. B. Fu, 2003: Analysis of simulated heavy rain over the Yangtze River valley during 11–30 June 1998 using RIEMS. Adv. Atmos. Sci., 20, 815824, https://doi.org/10.1007/BF02915407.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yang, W., Y. Andréasson, L. P. Graham, J. Olsson, J. Rosberg, and F. Wetterhall, 2010: Distribution-based scaling to improve usability of regional climate model projections for hydrological climate change impacts studies. Hydrol. Res., 41, 211229, https://doi.org/10.2166/nh.2010.004.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yi, L., W. C. Zhang, and X. Y. Li, 2018a: Assessing hydrological modelling driven by different precipitation datasets via the SMAP soil moisture product and gauged streamflow data. Remote Sens., 10, 1872, https://doi.org/10.3390/rs10121872.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yi, L., W. C. Zhang, and K. Wang, 2018b: Evaluation of heavy precipitation simulated by the WRF Model using 4D-Var data assimilation with TRMM 3B42 and GPM IMERG over the Huaihe River Basin, China. Remote Sens., 10, 646, https://doi.org/10.3390/rs10040646.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yong, B., L.-L. Ren, X. Chen, Y. Zhang, W.-C. Zhang, C.-B. Fu, and G.-Y. Niu, 2009: Development of a large-scale hydrological model TOPX and its coupling with regional integrated environment modeling system RIEMS. Chin. J. Geophys., 8, 19541965, https://doi.org/10.1002/cjg2.1399.

    • Search Google Scholar
    • Export Citation
  • Yuan, Z., Z. Y. Yang, X. D. Zheng, and Y. Yuan, 2012: Spatial and temporal variations of precipitation in Huaihe river basin in recent 50 years. South-to-North Water Diversion Water Sci. Technol., 10, 98103.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    (a) Location of the Wangjiaba watershed. (b) Locations of the precipitation, evaporation, and hydrological stations in the Wangjiaba watershed, and its altitude (m) above sea level.

  • View in gallery

    Daily precipitation (mm) observed at each CMWR station in the WJB watershed in August 2015 and their mean daily values (mm).

  • View in gallery

    Daily watershed means of precipitation and potential evaporation from the WRF Model, the WRF 4D-Var system, and the CMWR in situ stations.

  • View in gallery

    Linear fittings among the potential evaporation (mm day−1) observed by the CMWR stations, the WRF Model, and the WRF 4D-Var system.

  • View in gallery

    The 9-km grid data of the average daily soil moisture in the Wangjiaba watershed extracted from (a) the SMAP and the experiments of (b) CMWR_PreEvp, (c) WRF_Pre, (d) WRF_PreEvp, (e) 4D-Var_Pre, and (f) 4D-Var_PreEvp during the study period.

  • View in gallery

    The Wangjiaba watershed average of the daily soil moisture obtained from the SMAP (mm3 mm−3) and the different land–atmosphere modeling experiments (mm).

  • View in gallery

    The 9-km grid data of the Pearson’s correlation coefficient (CC) between the daily soil moisture from the SMAP and the land–atmosphere coupling experiments of (a) CMWR_PreEvp, (b) WRF_Pre, (c) 4D-Var_Pre, (d) WRF_PreEvp, and (e) 4D-Var_PreEvp during the study period in the Wangjiaba watershed. The black filled circles denote the CC value that is statistically significant at the level of 0.05.

  • View in gallery

    The simulated and observed discharges at the Wangjiaba watershed outlet: (a) hourly and (b) daily.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 164 164 19
PDF Downloads 125 125 18

Impact of 4D-Var Data Assimilation on Performance of the Coupled Land–Atmosphere Model WRF–TOPX: A Case Study of a Flood Event in the Wangjiaba Watershed, China

View More View Less
  • 1 Key Laboratory of Coastal Environment and Resources of Zhejiang Province, School of Engineering, Westlake University, Hangzhou, China
  • | 2 China State Key Laboratory of Hydrology-Water Resources and Hydraulic Engineering, Hohai University, Nanjing, China
  • | 3 School of Resource Environment and Earth Science, Yunnan University, Kunming, China
  • | 4 Key Laboratory of Regional Climate-Environment Research for East Asia, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, China
© Get Permissions
Open access

Abstract

To assess the impact of four-dimensional variational (4D-Var) data assimilation on the performance of a land–atmosphere coupled model, the satellite precipitation of the Integrated Multisatellite Retrievals for Global Precipitation Measurement (IMERG) was assimilated into the Weather Research Forecast (WRF) Model, and the WRF was coupled to the hydrological model TOPX. Precipitation and evaporation were both investigated as connecting elements in the coupled model WRF–TOPX. Differing in whether the 4D-Var data assimilation and evaporation were applied, one control experiment and four experiments were performed to simulate a historical flood event that happened in the Wangjiaba watershed in eastern China. The key hydrological variables of precipitation, potential evaporation, soil moisture, and discharge in the studied flood process were evaluated. The results showed that 1) the 4D-Var data assimilation with the IMERG could reduce both the overestimations of the WRF-predicted precipitation and potential evaporation; 2) the applied 4D-Var data assimilation could improve considerably the accuracy of the soil moisture and discharges from the coupled model WRF–TOPX; and 3) evaporation was also an important factor to influence the net precipitation to affect the performance of the coupled land–atmosphere model. With the two connecting elements of precipitation and evaporation, the 4D-Var assimilation based on IMERG could improve the Nash–Sutcliffe coefficient of the coupled model WRF–TOPX from 0.483 to 0.521 at the hourly scale. These investigations can provide important implications for the land–atmosphere coupling with both the precipitation and evaporation and using the 4D-Var data assimilation with IMERG for flood simulation at a large scale.

Denotes content that is immediately available upon publication as open access.

© 2021 American Meteorological Society. For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).

Corresponding author: Lu Yi, yilu@westlake.edu.cn

Abstract

To assess the impact of four-dimensional variational (4D-Var) data assimilation on the performance of a land–atmosphere coupled model, the satellite precipitation of the Integrated Multisatellite Retrievals for Global Precipitation Measurement (IMERG) was assimilated into the Weather Research Forecast (WRF) Model, and the WRF was coupled to the hydrological model TOPX. Precipitation and evaporation were both investigated as connecting elements in the coupled model WRF–TOPX. Differing in whether the 4D-Var data assimilation and evaporation were applied, one control experiment and four experiments were performed to simulate a historical flood event that happened in the Wangjiaba watershed in eastern China. The key hydrological variables of precipitation, potential evaporation, soil moisture, and discharge in the studied flood process were evaluated. The results showed that 1) the 4D-Var data assimilation with the IMERG could reduce both the overestimations of the WRF-predicted precipitation and potential evaporation; 2) the applied 4D-Var data assimilation could improve considerably the accuracy of the soil moisture and discharges from the coupled model WRF–TOPX; and 3) evaporation was also an important factor to influence the net precipitation to affect the performance of the coupled land–atmosphere model. With the two connecting elements of precipitation and evaporation, the 4D-Var assimilation based on IMERG could improve the Nash–Sutcliffe coefficient of the coupled model WRF–TOPX from 0.483 to 0.521 at the hourly scale. These investigations can provide important implications for the land–atmosphere coupling with both the precipitation and evaporation and using the 4D-Var data assimilation with IMERG for flood simulation at a large scale.

Denotes content that is immediately available upon publication as open access.

© 2021 American Meteorological Society. For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).

Corresponding author: Lu Yi, yilu@westlake.edu.cn

1. Introduction

The coupled land–atmosphere model based on the regional climate model and hydrological model is an important tool to extend the forecast period of local flood (Bosilovich and Sun 1999; Wu and Zhang 2013). In a coupled land–atmosphere model, the regional climate model can provide a hydrological model with continuous spatiotemporal variation fields of hydrological variables such as precipitation, evaporation, temperature, and radiation. The hydrological model has more refined mechanisms for the generation and confluence of the vertical and lateral streamflow, which can make up for consideration of only the vertical streamflow in the regional climate model (Arnault et al. 2016). Many related studies have been carried out to examine the impacts of various factors on the coupled land–atmosphere model, including solar radiation, initial soil moisture, sea surface temperature, land surface anomalies, runoff infiltration, physical parameters, land model calibration, and bias correction (Dagon and Schrag 2016; DelSole et al. 2008; Dirmeyer and Zhao 2004; Ferguson et al. 2012; Kim and Wang 2007; Santanello et al. 2013; Tian et al. 2017a). However, as an important connecting element in the land–atmosphere coupling, the accuracy of the precipitation simulated by the regional climate model is especially influential to determine the success of flood simulation (Jasper et al. 2002; Tian et al. 2017b; Verbunt et al. 2006; Yang et al. 2010).

The performance of a regional climate model to predict precipitation can be enhanced by reducing its system error through solving the stochastic problem of atmospheric motion, optimizing the physical parameterization scheme, and improving its spatial resolution. Moreover, good initial condition is also a key factor to improve the accuracy of precipitation predicted by a climate model, since the essence of numerical weather prediction is to solve mathematical equations.

Data assimilation is an effective way to improve the initial condition of the regional climate model. It mainly includes the methods of polynomial interpolation, optimal interpolation, spectral method, response function method, Kalman filter, three-dimensional variational (3D-Var) assimilation, and four-dimensional variational (4D-Var) assimilation (Bannister 2017). In spite of costing huge computational resources, 4D-Var is still specially favored in improving the initial conditions of climate models, since it can directly assimilate not only conventional but also unconventional observed precipitation in continuous time and with better dynamic constraints (Bannister 2017; Bouttier and Kelly 2001). As the technology of remote sensing observation develops, more and more satellite precipitation products with wide coverage and high resolution are freely open to the public, such as TRMM, GPM, CMORPH, GSMaP, PERSIANN, and MSWEP (Maggioni and Massari 2018). This provides plenty of unconventional observation operators for the 4D-Var data assimilation.

Assimilating unconventional observations of the satellite-based precipitation product based on the 4D-Var data assimilation method provides another possible way to improve the performance of a regional climate model for precipitation simulation (Lin et al. 2015; Mahfouf et al. 2005; Pan et al. 2017; Yi et al. 2018b). However, the impact of 4D-Var assimilation with precipitation observation on the accuracy of evaporation, which is also another important input in most hydrology models, is seldom analyzed, and the performance of the land–atmosphere coupled with both precipitation and evaporation is seldom investigated as well. Therefore, we looked into the possible influence of 4D-Var assimilation with the satellite-based precipitation on the performance of the coupled land–atmosphere model, through detailed analysis of its impacts on the two connecting elements (precipitation and evaporation) and the two hydrological variables (soil moisture and discharge) in one flood process. In the rest of this paper, section 2 will introduce the study data and methods applied in the 4D-Var assimilation, land–atmosphere model coupling, and evaluation. Section 3 will discuss the accuracy changes of the precipitation and evaporation predicted by the atmosphere model due to the application of 4D-Var data assimilation, and the accuracy variations of the soil moisture and discharge simulated by the coupled land–atmosphere models due to the application of 4D-Var and different coupling elements. Section 4 will give the conclusions and highlight the limitations of this study.

2. Data and methods

a. Study area and event

1) Study area

This research was carried out in the Wangjiaba (WJB) watershed, covering an area of 30 630 km2. It is an upstream subbasin of the Huaihe River basin (HRB), which is one of the seven major basins in China and lies between the Yellow River and the Yangze River (Fig. 1a). The HRB is in the climate transition zone between north and south China; it has a warm temperature and semihumid monsoon climate. Its annual mean precipitation and actual evaporation are approximately 900 and 720 mm, respectively (Gao et al. 2018; Sun et al. 2018). The elevation difference in the WJB is about 170 m, which is much higher than that in the middle (~16 m) and lower (~6 m) streams of the HRB. Such a big decrease of elevation in the WJB watershed results in rapid flow transfer from the upstream to the middle and lower reaches of the HRB. This may increase the risk of flood occurrence in the HRB. Therefore, accurate simulation for the rainfall–runoff process in the WJB is crucial for flood forecasting and management in the HRB.

Fig. 1.
Fig. 1.

(a) Location of the Wangjiaba watershed. (b) Locations of the precipitation, evaporation, and hydrological stations in the Wangjiaba watershed, and its altitude (m) above sea level.

Citation: Journal of Hydrometeorology 22, 3; 10.1175/JHM-D-20-0161.1

2) Study period and event

A regional climate model commonly demands vast computational resources and computing time, particularly when it applies 4D-Var data assimilation and runs at a very fine grid spacing of 1 km (Bukovsky and Karoly 2009; Lin et al. 2015; Pennelly et al. 2014; Rogelis and Werner 2018). Therefore, considering the limitation of computational resources, the data availability, the spinup problem (Kleczek et al. 2014; Srinivas and Rao 2014; Veerse and Thepaut 1998), and the fact that a single heavy precipitation event is the main contributor to the amount of annual precipitation in the WJB watershed (Xia et al. 2012; Yuan et al. 2012), we selected the flood process from 17 to 30 August 2015 triggered by a 2-day continuous precipitation as the study event.

In the study period, precipitation observations were collected from the 215 precipitation stations (Fig. 1b) of China Ministry of Water Resources (CMWR). As shown in Fig. 2, the mean daily precipitation based on the CMWR stations reached 27.4 and 21.5 mm on 18 and 19 August, respectively; there are 26 and 12 CMWR stations that observed heavy precipitation (more than 50 mm day−1) on these two consecutive rainy days. The highest daily precipitation reached 97.8 mm on 18 August. Then on the second day, the mean rainfall decreased, and the detected highest precipitation dropped to around 68.7 mm.

Fig. 2.
Fig. 2.

Daily precipitation (mm) observed at each CMWR station in the WJB watershed in August 2015 and their mean daily values (mm).

Citation: Journal of Hydrometeorology 22, 3; 10.1175/JHM-D-20-0161.1

b. Data

1) Precipitation

In this investigation, precipitation from the in situ observation, climate model prediction, and remote sensing were collected. The in situ observed precipitation was applied to evaluate the accuracy of the precipitation predicted by the regional climate model and to drive the hydrological model for a control experiment. The daily gauged precipitation data were obtained from the dense network of 215 CMWR stations in the WJB watershed (Fig. 1b). They were gathered from the book of Annual Hydrological Reports for China published by the CMWR (http://www.mwr.gov.cn; hereafter referenced as the CMWR data). Integrated Multisatellite Retrievals for Global Precipitation Measurement (IMERG) data with spatiotemporal resolutions of 0.1° and 30 min (Liu 2016; Sharifi et al. 2016; Wang et al. 2017) were applied as observation operators in the 4D-Var data assimilation. According to Li et al. (2018) the Pearson’s correlation coefficient (CC) of the daily IMERG in the HRB reaches 0.76, and its root-mean-square error (RMSE) is 6.48 mm day−1 when comparing with daily gauged precipitation.

2) Evaporation

Evaporation is another important hydrological variable in water cycle. It is also a direct input of most hydrological models that can be obtained from the regional climate model. Therefore, evaporation can be another connecting element during the land–atmosphere coupling. The daily evaporation data were collected from the 10 evaporation stations (Fig. 1b) located in the WJB watershed and recorded by the CMWR. The data were applied to evaluate the potential evaporation calculated by the employed regional climate model.

3) Soil moisture and outlet streamflow discharge

As the measurement network for soil moisture is sparse in the WJB watershed and the gauged data are generally not available, the Soil Moisture Active Passive (SMAP) product launched by the National Aeronautics and Space Administration (NASA) was taken as the benchmark to evaluate the soil moisture simulations (Liu et al. 2009; Santi et al. 2013). Taking advantage of the relative strengths of both active (radar) and passive (radiometer) microwave remote sensing, the SMAP product has high spatiotemporal resolutions of 9 km and 3 h. Generally, the remote sensing product with finer resolution is considered to have higher accuracy (Liu et al. 2016). The accuracy of the SMAP soil moisture reaches an unbiased RMSE of 0.04 m3 m−3 or better (Reichle et al. 2017). In this study, the SMAP L4_SM product at the root zone (0–100 cm) was used since the vegetation root was considered during the calculation of the soil moisture in the TOPX. The hourly streamflow discharge of the hydrological station (Fig. 1b) at the WJB outlet was provided by the China Institute of Water Resources and Hydropower Research (IWHR).

c. Method

1) Regional climate model WRF

The universally applied regional climate models include MM5, National Meteorological Center (NMC), and the WRF Model from the United States, the UKMO model from the United Kingdom, the JMA mesoscale model from Japan, the ECMWF model from Europe, and the Regional Integrated Environment Modeling System (RIEMS) model from China (Shrestha et al. 2013; Xiong et al. 2003). In this study, the WRF Model was employed since it has merged many of the latest developments in atmosphere research (Li et al. 2019) and can provide a high spatial resolution of 1 km, which is also feasible in a hydrological model, thus overcoming the spatial-scale mismatch problem between the regional climate model and hydrological model when coupling.

In the WRF Model, the final (FNL) analysis data obtained from the National Centers for Environmental Prediction (NCEP) were used as the initial and boundary conditions. A four-layer nested domain around the WJB watershed was set (Fig. 1a). The outer domain (D01) was set as large as possible to cover the important weather system influencing the precipitation in the WJB, which mainly includes the East Asian subtropical monsoon, the Asian monsoon, and the Indian southwestern monsoon. The D01, D02, D03, and D04 domains contained 80 × 155, 322 × 271, 604 × 433, and 700 × 700 grids, respectively; their spatial resolutions were 27, 9, 3, and 1 km, respectively; and their temporal resolutions were 150, 50, 17, and 6 s, respectively. Twenty-seven layers were set in the vertical direction. The physical configuration of the WRF Model and its related dominant parameters were set by taking the reference of successful forecast case (Yi et al. 2018b). The main physical configuration of the WRF Model is listed in Table 1.

Table 1.

The main configuration of the WRF Model.

Table 1.

2) Hydrological model TOPX

The hydrological model was selected to meet the following two needs: 1) it is constructed on a grid that facilitates the coupling with atmosphere model, and 2) it can be applied to large-scale watershed and reflect the different runoff generation mechanisms in areas with different topographies as hill and plains. Based on these criteria, the distributed hydrological model TOPX was employed in this study. The TOPX model is constructed on the basis of the topographic index (TOP) and the water balance concept of the Xin’anjiang model (Yong et al. 2009). The input data of the TOPX model include precipitation, potential evaporation, and the distribution of topographic index. Before coupling, the calibration and validation of the TOPX were performed by simulating two long-term rainfall runoff processes covering the periods of 2001–05 and 2014–15, and seven short-term flood events that happened during the long-term rainfall runoff processes. The final main parameters of the TOPX are listed in Table 2. The details of calibration and validation, and the related topographic index in the WJB watershed can be found in a prior study (Yi et al. 2018a).

Table 2.

Main parameters in the hydrological model TOPX.

Table 2.

3) 4D-Var data assimilation and coupling experiments

The commonly used incremental 4D-Var formulation (Barker et al. 2012; Courtier et al. 1994; Lorenc 2003; Yi et al. 2018b) was utilized in this study. The incremental approach was designed to determine an analysis increment that minimized the cost function, which was defined as a function of the analysis increment instead of the analysis itself. The background error covariance matrices in August 2015 required in the 4D-Var assimilation was generated with one-month ensemble forecasts every 12 h using the NMC method (Parrish and Derber 1992). For the 4D-Var assimilation, the IMERG was aggregated to 6 h as to be assimilated in a time window of 6 h. Furthermore, except for the initial conditions, the setting of the WRF 4D-Var system is almost the same as that set in the WRF Model (Table 1).

Although the WRF contains its own land surface model such as Noah, RUC, CLM, and others (Pei et al. 2014; Smirnova et al. 2016; Srivastava et al. 2015), these land surface models can only simulate runoff in vertical direction. While the TOPX model can numerically describe the runoff in both vertical and lateral directions, while the lateral flow is especially important when simulating the water cycle in watershed at large scale. Therefore, we used the coupled model WRF–TOPX to simulate the rainfall runoff process in the WJB watershed. One-way coupling was applied in this study. The outputs of the WRF were directly collected to drive the TOPX.

To investigate the impact of 4D-Var data assimilation on the performance of the coupled model WRF–TOPX, five numerical experiments were designed (Table 3). The CMWR_PreEvp experiment is a kind of traditional rainfall runoff simulation only by the TOPX. It served as a control experiment and was driven by in situ gauged precipitation and evaporation from the CMWR. Furthermore, to assess the impact of the applied 4D-Var and different connecting element on the performance of the coupled model, we set four coupling experiments that were different in whether they applied the 4D-Var in the WRF Model and whether they used the connecting element of evaporation except for the precipitation. Meanwhile, to overcome the spatial-scale mismatch between the WRF and the TOPX, the spatial resolution of the WRF was set to 1 km through a four-layer nested domain, and the grid resolution in the TOPX set to 1 km as well. Given the temporal resolution of the WRF output and the data for the performance evaluation, the TOPX ran at the time interval of 1 h. Limited by the input data, the CMWR_PreEvp experiment was performed at daily scale.

Table 3.

Design of the land–atmosphere coupling experiments. The CMWR_PreEvp denotes the hydrological modeling only applying the TOPX model driven by the in situ gauged precipitation and evaporation (from the CMWR); it serves as a benchmark for the assessment of coupling experiments. Experiments 2–5 were of land–atmosphere coupling; they are labeled as A_B, where A denotes the applied atmosphere model (the WRF Model or the WRF 4D-Var system), and B denotes the connected element (only precipitation or both precipitation and evaporation).

Table 3.

4) Evaluations of the hydrological variables

Before the five experiments were conducted, the connecting elements of precipitation and evaporation predicted by the WRF Model and the WRF 4D-Var system were evaluated against the in situ observations. The hourly precipitation or evaporation values were extracted from the grid points nearest to the precipitation or evaporation stations, then accumulated to daily values for comparison with the daily station observations. In this evaluation, we used the error scores of the mean error (ME), relative error (RE), RMSE, and CC (Table 4), which describe the errors, deviations, and correlation between the simulations and the observations.

Table 4.

Statistical metrics applied in the evaluations. The terms Ps,i and Po,i denote the simulated and observed values, respectively, of the i grid, and Ps,i` and Po,i` are their respective means. The terms Qo,i and Qs,i are the observed and simulated discharges at time step i, and Q¯o is the average of the observed discharge during the study period.

Table 4.

The soil moisture and watershed discharges simulated by the control experiment and four coupling experiments were evaluated as well. The unit of soil moisture set by the TOPX is the depth of the water column (mm), which is different from water volume content (mm3 mm−3) used in the SMAP. Moreover, the soil depth cannot be concretely defined in the TOPX; it applies parameters such as maximum storage capacities of the three soil layers and impact of vegetation root to control the variation of soil moisture. Given the different units of soil moisture in the TOPX and the SMAP and the unspecified soil depth in the TOPX, we indirectly evaluated the soil moisture simulated by the coupled experiments with the index of CC, which is frequently used in the evaluation of soil moisture (Cui et al. 2018). The Nash–Sutcliffe (NS) coefficient was used to evaluate the simulated discharge at the outlet of the WJB watershed.

3. Results and discussions

a. Evaluations of the simulated precipitation and evaporation

Before coupling, the accuracies of the two connecting elements, i.e., the precipitation and evaporation simulated by the WRF Model or the WRF 4D-Var system, were evaluated. Their daily means in the WJB watershed are calculated and illustrated in Fig. 3. It is found that both the precipitation and potential evaporation are overestimated by the WRF Model. During the continuous 2-day raining, the evaporation decreased as the precipitation continued. This may be related to the reduction of temperature during rainfalls. The precipitation simulated by the WRF Model and the WRF 4D-Var system had been evaluated in detail at the hourly, 12-hourly, and daily scales with station observations and merged CMORPH data in a prior study of Yi et al. (2018a), more detailed information can be found in it. According to that investigation, it can be concluded that, after the use of the 4D-Var data assimilation with GPM, the ME and RE values turned from positive to negative. This implied a change of the simulated precipitation from overestimation to underestimation. The larger absolute values of the ME and RE of the WRF 4D-Var simulations indicated larger errors and deviations. However, the RMSE of the WRF 4D-Var system decreased from 13.411 to 11.946 (Table 5). This reduction indicated decreases in errors of the extremely high and low precipitation and enhancement in the accuracy of the WRF 4D-Var simulated precipitation. Meanwhile, the CC value was improved from 0.343 to 0.444 (Table 5). This indicated a better correlation between the simulations and observations. In sum, the 4D-Var assimilation with GPM could play a positive effect on the precipitation simulation of the WRF Model.

Fig. 3.
Fig. 3.

Daily watershed means of precipitation and potential evaporation from the WRF Model, the WRF 4D-Var system, and the CMWR in situ stations.

Citation: Journal of Hydrometeorology 22, 3; 10.1175/JHM-D-20-0161.1

Table 5.

Statistics of the precipitation simulated by the WRF and WRF 4D-Var against the daily CMWR in situ observations.

Table 5.

The daily values of the grids nearest to the evaporation stations were extracted out and evaluated with the evaporation station measurements. The ME, RE, RMSE, and CC of the evaporation simulations are listed in Table 6. The ME of the evaporation predicted by the WRF Model and the WRF 4D-Var system were 2.817 and 2.709 mm, respectively. The RE, RMSE, and CC values of the WRF 4D-Var simulations were all slightly improved than the simulations of the WRF Model. After using the 4D-Var assimilation, the determination coefficient (R2) between the WRF prediction and the CMWR observation increased from 0.385 to 0.556 (Fig. 4). These indicated that the 4D-Var data assimilation with IMERG could improve the accuracy of the WRF-predicted potential evaporation as well. However, looking into the RE changes of the simulated precipitation and potential evaporation, we found that before and after the data assimilation, the RE of the former stayed at 3.7% and −43.6%, respectively (Table 5), and the RE of the latter stayed at 96.3% and 92.6%, respectively (Table 6). The smaller change of the latter RE implied a weaker impact of the applied 4D-Var data assimilation on the simulated potential evaporation than those on the precipitation. This slight influence is also indirectly reflected in Fig. 4c, since the determination coefficient between the WRF-simulated and the WRF_4D-Var-simulated evaporation is still relatively high. The reason for this weaker impact on evaporation mostly laid in the observation operator of the data assimilation; we used direct satellite-estimated precipitation but not evaporation or other basic satellite radiance, which has closer relationship with evaporation than precipitation.

Table 6.

Statistics of the potential evaporation simulated by the WRF Model and the WRF 4D-Var system comparing to the CMWR in situ observations.

Table 6.
Fig. 4.
Fig. 4.

Linear fittings among the potential evaporation (mm day−1) observed by the CMWR stations, the WRF Model, and the WRF 4D-Var system.

Citation: Journal of Hydrometeorology 22, 3; 10.1175/JHM-D-20-0161.1

b. Evaluation of the simulated soil moisture

To keep consistency with the spatiotemporal resolutions of the referenced data, the soil moisture simulated by the five experiments was extracted from the three soil layers of the TOPX model and vertically averaged, then upscaled from 1 to 9 km with a kriging interpolation method, and the hourly results of the other four coupling experiments were aggregated to mean daily values. The distributions of the mean daily soil moistures are depicted in Fig. 5. It is obvious that the simulated soil moisture from the five experiments all reflect the main trend with the SMAP (Fig. 5); the northwest has lower soil moisture while the southeast has higher soil moisture. This is closely influenced by the precipitation distribution.

Fig. 5.
Fig. 5.

The 9-km grid data of the average daily soil moisture in the Wangjiaba watershed extracted from (a) the SMAP and the experiments of (b) CMWR_PreEvp, (c) WRF_Pre, (d) WRF_PreEvp, (e) 4D-Var_Pre, and (f) 4D-Var_PreEvp during the study period.

Citation: Journal of Hydrometeorology 22, 3; 10.1175/JHM-D-20-0161.1

According to the water balance rule, the amounts of soil moisture mainly depend on the initial soil moisture and the net precipitation, which is determined by precipitation and evaporation. Therefore, the mean daily soil moisture of the WRF_Pre was higher than that of the CMWR_Pre and the WRF_PreEvp as the WRF-predicted precipitation and potential evaporation were overestimated, and the mean daily soil moistures simulated by the 4D-Var_PreEvp were lower than those simulated by the 4D-Var_Pre experiments, as the potential evaporation predicted by the WRF Model was still overestimated after using 4D-Var assimilation. This can be also found in the simulated watershed mean daily soil moistures (Fig. 6). As the precipitation occurred, the watershed mean of the daily soil moisture started to increase after the precipitation on 18 August, reached the peak values one or two days later, and then began to fall as the rain stopped and evaporation from bare soils took place.

Fig. 6.
Fig. 6.

The Wangjiaba watershed average of the daily soil moisture obtained from the SMAP (mm3 mm−3) and the different land–atmosphere modeling experiments (mm).

Citation: Journal of Hydrometeorology 22, 3; 10.1175/JHM-D-20-0161.1

For the evaluation at field scale, the 412 grids in the WJB watershed with resolution of 9 km were compared to the corresponding grid values of the daily SMAP. There were 96.1%, 78.4%, 94.7%, 93.2%, and 94.9% of the total grids passed the statistical assessment at the level of 0.05 for the five experiments, respectively. The daily simulated soil moisture of the CMWR_PreEvp showed the best agreement with the SMAP data (Fig. 7a); its mean CC value reached 0.820, while the mean CC value of the WRF_Pre showed the lowest 0.650, due to the overestimation of precipitation in the WRF_Pre (Fig. 7b). The mean values of CC for the WRF_PreEvp, 4D-Var_Pre, and 4D-Var_PreEvp were 0.783, 0.781, and 0.802, respectively. The results showed that the coupled models with the 4D-Var assimilation generally produced better matches with the SMAP.

Fig. 7.
Fig. 7.

The 9-km grid data of the Pearson’s correlation coefficient (CC) between the daily soil moisture from the SMAP and the land–atmosphere coupling experiments of (a) CMWR_PreEvp, (b) WRF_Pre, (c) 4D-Var_Pre, (d) WRF_PreEvp, and (e) 4D-Var_PreEvp during the study period in the Wangjiaba watershed. The black filled circles denote the CC value that is statistically significant at the level of 0.05.

Citation: Journal of Hydrometeorology 22, 3; 10.1175/JHM-D-20-0161.1

c. Evaluation of the simulated discharge

The discharges at the WJB watershed outlet simulated by the five experiments were compared to the in situ gauged discharges at hourly and daily scales. It is clear in Fig. 8 that the WJB watershed outlet experienced the flood peak (773.7 m3 s−1) one day after the precipitation on 18 August, the peak simulated by the four coupling experiments were one day later. Nevertheless, the simulated discharges all reflected the similar fluctuation tendency with the in situ observations.

Fig. 8.
Fig. 8.

The simulated and observed discharges at the Wangjiaba watershed outlet: (a) hourly and (b) daily.

Citation: Journal of Hydrometeorology 22, 3; 10.1175/JHM-D-20-0161.1

As listed in Table 7, the CC values of the simulated discharges with the 4D-Var were less than those without data assimilation. This might be caused by the increase of the RE of the 4D-Var simulated precipitation, which indicated larger deviations of the simulated precipitation. Although the evaluated indices showed different changes, the NS coefficient has higher weight in the assessment of flood simulation. The NS values of the 4D-Var_Pre and the 4D-Var_PreEvp were improved after applying the 4D-Var. The WRF_Pre and WRF_PreEvp showed higher positive RE values than the CMWR_PreEvp because of the overestimation of precipitation in the WRF. The join of simulated evaporation as connecting element increased the uncertainty of the final simulation through its impact on net precipitation. The overestimation of evaporation in the WRF reduced the overestimated magnitude of the net precipitation in the WRF_PreEvp, and resulted in a better NS than the WRF_Pre. However, since the WRF 4D-Var predicted much lower precipitation and still relatively higher evaporation than the observations, the net precipitation in the 4D-Var_PreEvp was much smaller and rendered a lower NS than that in the 4D-Var_Pre, and also resulted in a higher absolute RE and lower NS than that in the WRF_PreEvp as well. It can be concluded that whether applying the connecting element of evaporation during the land–atmosphere coupling depends on whether its application could render a better simulation of net precipitation, this is also the reason why the NS and RE of the WRF_PreEvp are much closer to their perfect values than the WRF_Pre, while the NS and RE of the 4D-Var_PreEvp are further from their perfect values than the 4D-Var_Pre.

Table 7.

Statistical results of the simulated discharges at the Wangjiaba watershed outlet.

Table 7.

The hourly and daily NS values of the 4D-Var_Pre were 0.547 and 0.560, respectively. The daily NS of the CMWR_PreEvp (0.673) was still higher than any other four coupled models. Nevertheless, the CMWR_PreEvp experiment can only represent the daily discharges limited by the temporal resolution of the available CMWR data. On the whole, the 4D-Var data assimilation with the IMERG data could generally improve the discharge simulation.

4. Conclusions

Differing in whether they applied 4D-Var assimilation with the IMERG and the connecting elements for land–atmosphere coupling, four coupling experiments based on the WRF and TOPX were applied to simulate a flood event that happened in the WJB watershed at 1-km and hourly scales. Before coupling, the precipitation and potential evaporation predicted by the WRF 4D-Var system were evaluated. It was found that the WRF Model overestimated both the precipitation and potential evaporation. Although the observation operator in the 4D-Var assimilation is precipitation, the accuracy of the WRF-predicted evaporation could be improved as well. The determination coefficient between the simulated potential evaporation and the CMWR observation increased from 0.385 to 0.556 after using the 4D-Var assimilation.

The spatiotemporal variation of the soil moisture in the WJB watershed simulated by the WRF–TOPX was largely determined by the distribution of the predicted precipitation. Compared to the daily SMAP soil moisture, the CC values of the WRF_PreEvp was 0.019 higher than the 4D-Var_PreEvp experiment. The applied 4D-Var assimilation could improve the discharge simulation by the coupled land–atmosphere model as well. Among the four coupling models, the 4D-Var_Pre gave the best NS coefficients of 0.547 and 0.560 at the hourly and daily scales, respectively, due to having more exact net precipitation. The NS of the four coupled models ranged from 0.464 to 0.547 at the hourly scale and from 0.499 to 0.560 at daily scale. The performances of the coupled land–atmosphere models with 4D-Var assimilation were better than the coupled models without 4D-Var assimilation.

The 4D-Var assimilation with the IMERG could improve the performance of the coupled land–atmosphere model WRF–TOPX in simulating both soil moisture and discharge, as it played positive impacts on decreasing the overestimations of the WRF-predicted precipitation and potential evaporation. The four coupled models showed no better performances than the TOPX simulation only driven by traditional in situ observations, but they had much higher temporal resolutions and forecasting abilities that the TOPX model cannot provide; this makes the land–atmosphere model much more preponderant. The application of the 4D-Var assimilation with the IMERG can provide a new way to improve the performance of the coupled land–atmosphere model. The biases between the observed and simulated values may result from the hypothesis in the atmospheric and hydrological models, biases in the satellite observations, and computational errors, which need more investigations in future related studies.

Acknowledgments

This research is funded by the National Natural Science Foundation of China (4197060711), Zhejiang Provincial Natural Science Foundation of China (LQ21D010005), China Postdoctoral Science Foundation (2020M671815), and the postdoctoral fellowship supported by the Hangzhou, Zhejiang province in China. We are grateful to the NCAR Command Language (NCL) email list (ncl-talk@ucar.edu), which helped us with meteorological data processing by the NCL. Appreciations should also be given to the CMWR and the CMA for providing the in-situ observed hydrological data.

REFERENCES

  • Arnault, J., S. Wagner, T. Rummler, B. Fersch, J. Bliefernicht, S. Andresen, and H. Kunstmann, 2016: Role of runoff–infiltration partitioning and resolved overland flow on land–atmosphere feedbacks: A case study with the WRF-Hydro coupled modeling system for West Africa. J. Hydrometeor., 17, 14891516, https://doi.org/10.1175/JHM-D-15-0089.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bannister, R. N., 2017: A review of operational methods of variational and ensemble-variational data assimilation. Quart. J. Roy. Meteor. Soc., 143, 607633, https://doi.org/10.1002/qj.2982.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Barker, D., and et al. , 2012: The Weather Research and Forecasting Model’s Community Variational/Ensemble Data Assimilation System: WRFDA. Bull. Amer. Meteor. Soc., 93, 831843, https://doi.org/10.1175/BAMS-D-11-00167.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bosilovich, M. G., and W. Y. Sun, 1999: Numerical simulation of the 1993 Midwestern flood: Land–atmosphere interactions. J. Climate, 12, 14901505, https://doi.org/10.1175/1520-0442(1999)012<1490:NSOTMF>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bouttier, F., and G. Kelly, 2001: Observing-system experiments in the ECMWF 4D-Var data assimilation system. Quart. J. Roy. Meteor. Soc., 127, 14691488, https://doi.org/10.1002/qj.49712757419.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bukovsky, M. S., and D. J. Karoly, 2009: Precipitation simulations using WRF as a nested regional climate model. J. Appl. Meteor. Climatol., 48, 21522159, https://doi.org/10.1175/2009JAMC2186.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Chen, F., and J. Dudhia, 2001: Coupling an advanced land surface-hydrology model with the Penn State–NCAR MM5 modeling system. Part I: Model implementation and sensitivity. Mon. Wea. Rev., 129, 569585, https://doi.org/10.1175/1520-0493(2001)129<0569:CAALSH>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Courtier, P., J. N. Thépaut, and A. Hollingsworth, 1994: A strategy for operational implementation of 4D-Var, using an incremental approach. Quart. J. Roy. Meteor. Soc., 120, 13671387, https://doi.org/10.1002/qj.49712051912.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Cui, C. Y., and et al. , 2018: Soil moisture mapping from satellites: An intercomparison of SMAP, SMOS, FY3B, AMSR2, and ESA CCI over two dense network regions at different spatial scales. Remote Sens., 10, 33, https://doi.org/10.3390/rs10010033.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dagon, K., and D. P. Schrag, 2016: Exploring the effects of solar radiation management on water cycling in a coupled land–atmosphere model. J. Climate, 29, 26352650, https://doi.org/10.1175/JCLI-D-15-0472.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • DelSole, T., M. Zhao, P. A. Dirmeyer, and B. P. Kirtman, 2008: Empirical correction of a coupled land–atmosphere model. Mon. Wea. Rev., 136, 40634076, https://doi.org/10.1175/2008MWR2344.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dirmeyer, P. A., and M. Zhao, 2004: Flux replacement as a method to diagnose coupled land–atmosphere model feedback. J. Hydrometeor., 5, 10341048, https://doi.org/10.1175/JHM-384.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dudhia, J., 1989: Numerical study of convection observed during the winter monsoon experiment using mesoscale two-dimensional model. J. Atmos. Sci., 46, 30773107, https://doi.org/10.1175/1520-0469(1989)046<3077:NSOCOD>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ferguson, C. R., E. F. Wood, and R. K. Vinukollu, 2012: A global intercomparison of modeled and observed land–atmosphere coupling. J. Hydrometeor., 13, 749784, https://doi.org/10.1175/JHM-D-11-0119.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Gao, J. Q., M. Qiao, X. Qiu, Y. Zeng, H. Hua, and M. Adamu, 2018: Estimation of actual evapotranspiration distribution in the Huaihe River upstream basin based on the generalized complementary principle. Adv. Meteor., 2018, 2158168, https://doi.org/10.1155/2018/2158168.

    • Search Google Scholar
    • Export Citation
  • Grell, G. A., and D. Devenyi, 2002: A generalized approach to parameterizing convection combining ensemble and data assimilation techniques. Geophys. Res. Lett., 29, 1693, https://doi.org/10.1029/2002GL015311.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hong, S.-Y., Y. Noh, and J. Dudhia, 2006: A new vertical diffusion package with an explicit treatment of entrainment processes. Mon. Wea. Rev., 134, 23182341, https://doi.org/10.1175/MWR3199.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Jasper, K., J. Gurtz, and H. Lang, 2002: Erratum to “Advanced flood forecasting in Alpine watersheds by coupling meteorological observations and forecasts with a distributed hydrological model.” J. Hydrol., 269, 253254, https://doi.org/10.1016/S0022-1694(02)00345-1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kim, Y., and G. L. Wang, 2007: Impact of initial soil moisture anomalies on subsequent precipitation over North America in the coupled land–atmosphere model CAM3–CLM3. J. Hydrometeor., 8, 513533, https://doi.org/10.1175/JHM611.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kleczek, M. A., G. J. Steeneveld, and A. A. M. Holtslag, 2014: Evaluation of the weather research and forecasting mesoscale model for GABLS3: Impact of boundary-layer schemes, boundary conditions and spin-up. Bound.-Layer Meteor., 152, 213243, https://doi.org/10.1007/s10546-014-9925-3.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Li, Q. L., W. C. Zhang, L. Yi, J. P. Liu, and H. Chen, 2018: Accuracy evaluation and comparison of GPM and TRMM precipitation product over Mainland China. Adv. Water Sci., 29, 303313, https://doi.org/10.14042/j.cnki.32.1309.2018.03.001.

    • Search Google Scholar
    • Export Citation
  • Li, Y. P., Z. Li, Z. Zhang, L. Chen, S. Kurkute, L. Scaff, and X. Pan, 2019: High-resolution regional climate modeling and projection over western Canada using a weather research forecasting model with a pseudo-global warming approach. Hydrol. Earth Syst. Sci., 23, 46354659, https://doi.org/10.5194/hess-23-4635-2019.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lim, K. S. S., and S. Y. Hong, 2010: Development of an effective double-moment cloud microphysics scheme with prognostic cloud condensation nuclei (CCN) for weather and climate models. Mon. Wea. Rev., 138, 15871612, https://doi.org/10.1175/2009MWR2968.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lin, L. F., A. M. Ebtehaj, R. L. Bras, A. N. Flores, and J. F. Wang, 2015: Dynamical precipitation downscaling for hydrologic applications using WRF 4D-var data assimilation: Implications for GPM era. J. Hydrometeor., 16, 811829, https://doi.org/10.1175/JHM-D-14-0042.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Liu, S., X. Mo, W. Zhao, V. Naeimi, D. Dai, C. Shu, and L. Mao, 2009: Temporal variation of soil moisture over the Wuding River basin assessed with an eco-hydrological model, in-situ observations and remote sensing. Hydrol. Earth Syst. Sci., 13, 13751398, https://doi.org/10.5194/hess-13-1375-2009.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Liu, Y. B., G. P. Wu, and C. Q. Ke, 2016: Hydrological Remote Sensing. Science Press, 96–108.

  • Liu, Z., 2016: Comparison of Integrated Multisatellite Retrievals for GPM (IMERG) and TRMM Multisatellite Precipitation Analysis (TMPA) monthly precipitation products: Initial results. J. Hydrometeor., 17, 777790, https://doi.org/10.1175/JHM-D-15-0068.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lorenc, A. C., 2003: Modelling of error covariances by 4D-Var data assimilation. Quart. J. Roy. Meteor. Soc., 129, 31673182, https://doi.org/10.1256/qj.02.131.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Maggioni, V., and C. Massari, 2018: On the performance of satellite precipitation products in riverine flood modeling: A review. J. Hydrol., 558, 214224, https://doi.org/10.1016/j.jhydrol.2018.01.039.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Mahfouf, J. F., P. Bauer, and V. Marecal, 2005: The assimilation of SSM/I and TMI rainfall rates in the ECMWF 4D-Var system. Quart. J. Roy. Meteor. Soc., 131, 437458, https://doi.org/10.1256/qj.04.17.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Mlawer, E. J., S. J. Taubman, P. D. Brown, M. J. Iacono, and S. A. Clough, 1997: Radiative transfer for inhomogeneous atmospheres: RRTM, a validated correlated-k model for the longwave. J. Geophys. Res., 102, 16 66316 682, https://doi.org/10.1029/97JD00237.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Pan, X. D., X. Li, G. D. Cheng, and Y. Hong, 2017: Effects of 4D-Var data assimilation using remote sensing precipitation products in a WRF Model over the complex terrain of an arid region river basin. Remote Sens., 9, 963, https://doi.org/10.3390/rs9090963.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Parrish, D. F., and J. C. Derber, 1992: The National Meteorological Center spectral statistical interpolation analysis system. Mon. Wea. Rev., 120, 17471763, https://doi.org/10.1175/1520-0493(1992)120<1747:TNMCSS>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Pei, L. S., N. Moore, S. Zhong, L. Luo, D. W. Hyndam, W. E. Heilman, and Z. Gao, 2014: WRF Model sensitivity to land surface model and cumulus parameterization under short-term climate extremes over the southern Great Plains of the United States. J. Climate, 27, 77037724, https://doi.org/10.1175/JCLI-D-14-00015.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Pennelly, C., G. Reuter, and T. Flesch, 2014: Verification of the WRF Model for simulating heavy precipitation in Alberta. Atmos. Res., 135–136, 172192, https://doi.org/10.1016/j.atmosres.2013.09.004.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Reichle, R. H., and et al. , 2017: Assessment of the SMAP Level-4 surface and root-zone soil moisture product using in situ measurements. J. Hydrometeor., 18, 26212645, https://doi.org/10.1175/JHM-D-17-0063.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Rogelis, M. C., and M. Werner, 2018: Streamflow forecasts from WRF precipitation for flood early warning in mountain tropical areas. Hydrol. Earth Syst. Sci., 22, 853870, https://doi.org/10.5194/hess-22-853-2018.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Santanello, J. A., S. V. Kumar, C. D. Peters-Lidard, K. Harrison, and S. J. Zhou, 2013: Impact of land model calibration on coupled land–atmosphere prediction. J. Hydrometeor., 14, 13731400, https://doi.org/10.1175/JHM-D-12-0127.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Santi, E., S. Paloscia, S. Pettinato, C. Notarnicola, L. Pasolli, and A. Pistocchi, 2013: Comparison between SAR soil moisture estimates and hydrological model simulations over the Scrivia test site. Remote Sens., 5, 49614976, https://doi.org/10.3390/rs5104961.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Sharifi, E., R. Steinacker, and B. Saghafian, 2016: Assessment of GPM-IMERG and other precipitation products against gauge data under different topographic and Climatic conditions in Iran: Preliminary results. Remote Sens., 8, 135, https://doi.org/10.3390/rs8020135.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Shrestha, D. L., D. E. Robertson, Q. J. Wang, T. C. Pagano, and H. A. P. Hapuarachchi, 2013: Evaluation of numerical weather prediction model precipitation forecasts for short-term streamflow forecasting purpose. Hydrol. Earth Syst. Sci., 17, 19131931, https://doi.org/10.5194/hess-17-1913-2013.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Smirnova, T. G., J. M. Brown, S. G. Benjamin, and J. S. Kenyon, 2016: Modifications to the Rapid Update Cycle Land Surface Model (RUC LSM) available in the Weather Research and Forecasting (WRF) Model. Mon. Wea. Rev., 144, 18511865, https://doi.org/10.1175/MWR-D-15-0198.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Srinivas, D., and D. V. B. Rao, 2014: Implications of vortex initialization and model spin-up in tropical cyclone prediction using Advanced Research Weather Research and Forecasting Model. Nat. Hazards, 73, 10431062, https://doi.org/10.1007/s11069-014-1125-4.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Srivastava, P. K., D. Han, M. A. Rico-Ramirez, P. O’Neill, T. Islam, M. Gupta, and Q. Dai, 2015: Performance evaluation of WRF-Noah Land surface model estimated soil moisture for hydrological application: Synergistic evaluation using SMOS retrieved soil moisture. J. Hydrol., 529, 200212, https://doi.org/10.1016/j.jhydrol.2015.07.041.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Sun, R. C., H. L. Yuan, and Y. Z. Yang, 2018: Using multiple satellite-gauge merged precipitation products ensemble for hydrologic uncertainty analysis over the Huaihe River basin. J. Hydrol., 566, 406420, https://doi.org/10.1016/j.jhydrol.2018.09.024.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Tian, J. Y., J. Liu, J. Wang, C. Li, F. Lu, and Z. Chu, 2017a: A spatio-temporal evaluation of the WRF physical parameterisations for numerical rainfall simulation in semi-humid and semi-arid catchments of Northern China. Atmos. Res., 191, 141155, https://doi.org/10.1016/j.atmosres.2017.03.012.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Tian, J. Y., J. Liu, D. Yan, C. Li, Z. Chu, and F. Yu, 2017b: An assimilation test of Doppler radar reflectivity and radial velocity from different height layers in improving the WRF rainfall forecasts. Atmos. Res., 198, 132144, https://doi.org/10.1016/j.atmosres.2017.08.004.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Veerse, F., and J. N. Thepaut, 1998: Multiple-truncation incremental approach for four-dimensional variational data assimilation. Quart. J. Roy. Meteor. Soc., 124, 18891908, https://doi.org/10.1002/qj.49712455006.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Verbunt, M., M. Zappa, J. Gurtz, and P. Kaufmann, 2006: Verification of a coupled hydrometeorological modelling approach for alpine tributaries in the Rhine basin. J. Hydrol., 324, 224238, https://doi.org/10.1016/j.jhydrol.2005.09.036.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wang, Z. L., R. D. Zhong, C. G. Lai, and J. C. Chen, 2017: Evaluation of the GPM IMERG satellite-based precipitation products and the hydrological utility. Atmos. Res., 196, 151163, https://doi.org/10.1016/j.atmosres.2017.06.020.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wu, L. Y., and J. Y. Zhang, 2013: Role of land–atmosphere coupling in summer droughts and floods over eastern China for the 1998 and 1999 cases. Chin. Sci. Bull., 58, 39783985, https://doi.org/10.1007/s11434-013-5855-6.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Xia, J., D. X. She, Y. Y. Zhang, and H. Du, 2012: Spatio-temporal trend and statistical distribution of extreme precipitation events in Huaihe River Basin during 1960–2009. J. Geogr. Sci., 22, 195208, https://doi.org/10.1007/s11442-012-0921-6.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Xiong, Z., S. Y. Wang, Z. M. Zeng, and C. B. Fu, 2003: Analysis of simulated heavy rain over the Yangtze River valley during 11–30 June 1998 using RIEMS. Adv. Atmos. Sci., 20, 815824, https://doi.org/10.1007/BF02915407.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yang, W., Y. Andréasson, L. P. Graham, J. Olsson, J. Rosberg, and F. Wetterhall, 2010: Distribution-based scaling to improve usability of regional climate model projections for hydrological climate change impacts studies. Hydrol. Res., 41, 211229, https://doi.org/10.2166/nh.2010.004.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yi, L., W. C. Zhang, and X. Y. Li, 2018a: Assessing hydrological modelling driven by different precipitation datasets via the SMAP soil moisture product and gauged streamflow data. Remote Sens., 10, 1872, https://doi.org/10.3390/rs10121872.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yi, L., W. C. Zhang, and K. Wang, 2018b: Evaluation of heavy precipitation simulated by the WRF Model using 4D-Var data assimilation with TRMM 3B42 and GPM IMERG over the Huaihe River Basin, China. Remote Sens., 10, 646, https://doi.org/10.3390/rs10040646.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yong, B., L.-L. Ren, X. Chen, Y. Zhang, W.-C. Zhang, C.-B. Fu, and G.-Y. Niu, 2009: Development of a large-scale hydrological model TOPX and its coupling with regional integrated environment modeling system RIEMS. Chin. J. Geophys., 8, 19541965, https://doi.org/10.1002/cjg2.1399.

    • Search Google Scholar
    • Export Citation
  • Yuan, Z., Z. Y. Yang, X. D. Zheng, and Y. Yuan, 2012: Spatial and temporal variations of precipitation in Huaihe river basin in recent 50 years. South-to-North Water Diversion Water Sci. Technol., 10, 98103.

    • Search Google Scholar
    • Export Citation
Save