Impacts of uncertainty in land surface information on simulated surface temperature and precipitation over China

Abstract

This study mainly focuses on the effects of uncertainty in land surface information on mesoscale numerical simulation. The Weather Research and Forecasting (WRF) model was used to simulate meteorological fields over China at a spatial resolution of 10 km during 2006. Near‐surface temperature and precipitation values obtained from WRF were evaluated using site observations. The effects of accurate and updated land surface information, including Shuttle Radar Topography Mission (SRTM) data, Moderate resolution imaging spectroradiometer (MODIS) land use data, vegetation fraction derived from MODIS normalized difference vegetation index (NDVI) and Harmonized World Soil Database (HWSD) data (LAST simulation), on WRF's performance were investigated by comparison with a simulation using the default land surface information (BASE simulation). WRF reproduced the temporal and spatial variations of near‐surface temperature and precipitation over China accurately, although its performance varied significantly by season and region. WRF underestimated near‐surface temperatures in most areas of the Yunnan–Guizhou Plateau, the Tibetan Plateau, the Northeast Plain, and the southeastern coastal regions, but temperatures were overestimated in most areas of the North China Plain, the Loess Plateau, Sichuan Basin, and western Xinjiang. WRF overestimated (underestimated) precipitation in most humid (arid) areas. A positive (negative) bias in simulated precipitation is found in summer (winter). With updated land surface information, WRF's performance in terms of both daily average values and extremes improved, and the root mean squared error values for daily mean temperature and daily accumulated precipitation decreased by 7 and 2.3%. These improvements are significant for temperature, but not significant for precipitation. The uncertainty in land surface information has a greater influence on temperature than on precipitation. These findings are very important for weather forecasting and studies involving climatological analyses covering East Asia.

Impact of updating land surface data on micrometeorological weather simulations from the WRF model

“…The modified run thus attempted to minimize the cold bias, and the new land boundary led to a reduction in RMSE from 2.28 ºC in the control run to 1.34 ºC at this station. Our results are in line with those reported in previous studies (Pienda et al, 2004;He et al, 2017). Significant changes in land cover are observed at Bhatinda due to urbanization.…”

“…Although the rainfall simulation from modified runs followed a similar pattern as that of observed values by reducing false alarms, the rainfall amount is largely overpredicted for this event. Rainfall is a result of combined processes of physics and dynamics at a certain location, and land surface changes alone are insufficient to capture the entire variability (He et al, 2017).…”

“…Kurkowski et al (2003) observed improvement in surface temperature forecasts after using updated fractional vegetation cover dataset from the Advanced Very High-Resolution Radiometer (AVHRR) in the Eta model. He et al (2017) updated all the land surface information fields like elevation, land-use data, vegetation fraction data and soil type data in the Weather Research and Forecasting (WRF) model and tested its capability of simulating near-surface temperature and precipitation. They concluded that the modified land surface representation had significant impacts on temperature and precipitation simulations reducing the root mean squared error values by 7 and 2.3%, respectively.…”

Effect of soil texture and hydraulic parameters on WRF simulations in summer in east China

“…For ST HWSD , loam, loamy sand, and sandy loam are the main soil types in the study area, accounting for 77.5, 10.9 and 6.3%, respectively (Figure (d)). ST HWSD has been described in our previous study (He et al , ). Large differences exist between ST FAO and ST HWSD , especially in the west and the north of the study domain.…”

Updated emission inventories of power plants in simulating air quality during haze periods over East China

“…A better simulation performance is reflected by higher R, smaller absolute value of MFB, and lower MFE, respectively tagged with upward arrows in Table 3. Additionally, significance of the improvements between the statistical indices was checked by using the method of bootstrap confidence interval (DiCiccio and Efron, 1996;He et al, 2017). In response to the introduction of UEIPP, chemical simulation showed a comprehensive improvement in MOD2 (see the upward arrows in Table 3).…”