陆地-大气相互作用是气候系统圈层相互作用中的重要环节，陆面-大气之间的热量、水汽交换对天气和气候的发展演变起到至关重要的作用。感热通量、潜热通量是陆气之间水、热交换过程的关键物理参量，准确估计地表热交换系数是提高感热通量和潜热通量模拟精度的关键。目前，WRF模式中对地表热交换系数的计算，有多种参数化方法，比较各种参数化方案的优劣对提高区域天气气候模拟具有重要意义。同时，在高分辨率数值模拟中，复杂的非均匀下垫面条件也影响地表热通量的模拟效果。因此，基于WRF的高分辨率数值模拟，本文评估了多种地表热交换系数参数化方案在不同植被类型地区的模拟偏差，并针对模拟不足提出了改进参数化方案的思路。

本研究针对中国东部地区，基于WRF模式，对多种地表热交换系数参数化方案进行了检验，细致评估了多种地表热交换系数参数化方案在高低不同植被地区的误差情况。评估的多种地表热交换系数参数化方案包括：Chen等人在1997年提出的参数化方案（Chen97方案）、2009年Chen等人在Chen97方案的基础上的改进方案（Chen09方案）以及Jiménez等人在2012年提出的方案（J12方案）。本研究在Chen等人方案的基础上，引入了新的组合参数化方案（Chen_combine方案），主要关注了新方案与Chen09方案在高植被地区的模拟对比情况。主要工作及其结果如下：

（1） 本研究设计了七组模拟试验方案，包括J12、Chen09方案以及Chen97方案中不同的C_zil（C_zil=0.01、0.05、0.1、0.3和0.5）参数取值的试验分别在中国东部高低植被地区通过偏差和平均绝对误差的方法进行模拟评估，发现在低植被地区，Chen09方案对于多种变量（2 m温度、感热、潜热和地表感热交换系数C_h）的模拟与观测较为接近。在高植被地区，J12和Chen97（C_zil=0.3）方案更接近实际观测。在潜热的模拟当中，J12方案具有一定的优势。

（2） 根据各方案在高低植被地区对2 m温度的模拟效果，本研究设计了一种陆气耦合系数组合计算方案（Chen_combine），即：在植被高度大于等于5 m时采用Chen97（C_zil=0.3）方案；在植被高度小于5 m时采用Chen09方案。本研究对该方案进行了模拟试验，并与Chen09方案在高植被地区进行对比。结果发现，Chen_combine方案在高植被地区的2 m温度模拟效果明显优于Chen09方案，日均平均绝对误差降至1.62 ℃，相较于Chen09方案的1.74 ℃，表现出了明显的优势。此外，Chen_combine方案在高植被地区对地表潜热、感热和地表感热交换系数C_h的模拟效果也明显改善，与实际观测更为接近。Chen_combine方案表明，合理修正高植被地区的陆气耦合强度可以改善Chen09方案在高植被地区模拟的准确性和可靠性。

Land surface-atmosphere interaction is an important link in the interaction of the climate system. The exchange of heat and water vapor between the land surface and the atmosphere plays a crucial role in the development and evolution of weather and climate. Sensible heat flux and latent heat flux are key physical parameters in the water and heat exchange process between land and air. Accurate estimation of surface heat exchange coefficient is the key to improving the simulation accuracy of sensible heat flux and latent heat flux. Currently, there are multiple parameterization methods for calculating surface heat exchange coefficients in WRF models. Comparing the advantages and disadvantages of various parameterization schemes is of great significance for improving regional weather and climate simulations. At the same time, in high-resolution numerical simulations, complex non-uniform underlying surface conditions also affect the simulation effect of surface heat flux. Therefore, based on the high-resolution numerical simulation of WRF, this paper evaluates the simulation deviations of various surface heat exchange coefficient parameterization schemes in areas with different vegetation types, and proposes ideas for improving parameterization schemes to address simulation deficiencies.

Focusing on eastern China, this study tested various surface heat exchange coefficient parameterization schemes based on the WRF model, and carefully evaluated the errors of various surface heat exchange coefficient parameterization schemes in areas with different levels of vegetation. Various surface heat exchange coefficient parameterization schemes evaluated include: the parameterization scheme proposed by Chen et al. in 1997(Chen97 scheme), the improved scheme based on the Chen97 scheme by Chen et al. in 2009 (Chen09 scheme), and Jiménez et al. The plan proposed in 2012 (J12 plan). This study introduced a new combined parameterization scheme (Chen_combine scheme) based on the scheme of Chen et al., and mainly focused on the simulation comparison between the new scheme and the Chen09 scheme in high vegetation areas. The main conclusions obtained are as follows:

(1) This study designed seven groups of simulation test plans, including tests with different C_zil (C_zil =0.01, 0.05, 0.1, 0.3 and 0.5) parameter values in the J12, Chen09 and Chen97 plans, respectively, in the high and low vegetation areas of eastern China through deviation and The mean absolute error method was used for simulation evaluation, and it was found that in low vegetation areas, the Chen09 scheme's simulation of multiple variables (2 m temperature, sensible heat, latent heat and surface sensible heat exchange coefficient C_h) was relatively close to observations. In high vegetation areas, the J12 and Chen97 (C_zil=0.3) schemes are closer to actual observations. In the simulation of latent heat, the J12 solution has certain advantages.

(2) This study designed a land-atmosphere coupling coefficient combination calculation scheme (Chen_combine) based on the simulation effect of each scheme on 2 m temperature in high and low vegetation areas, that is: Chen97 (C_zil=0.3) is used when the vegetation height is greater than or equal to 5 m, use the Chen09 scheme when the vegetation height is less than 5 m. This study conducted a simulation test on this plan, and paid special attention to its simulation comparison with the Chen09 plan in high vegetation areas. The results showed that the Chen_combine scheme's 2 m temperature simulation effect in high vegetation areas was significantly better than that of the Chen09 scheme, with the average daily average absolute error reduced to 1.62°C, which showed obvious advantages compared to the 1.74°C of the Chen09 scheme. In addition, the simulation effect of the Chen_combine scheme on surface latent heat, sensible heat and surface sensible heat exchange coefficient C_h in high vegetation areas is also significantly improved, and is closer to actual observations. The Chen_combine experiment shows that reasonable correction of the land-atmosphere coupling strength in high vegetation areas can improve the accuracy and reliability of the Chen09 scheme in simulations in high vegetation areas.