【作者】 杨启东；

【导师】 左洪超；

【作者基本信息】 兰州大学 ， 大气物理学与大气环境， 2012， 博士

【摘要】 干旱半干旱区占全球陆地面积近40%,其地表类型复杂多样,涵盖了草地,农田,沙漠,戈壁以及冻土等多种类型。这些复杂下垫面的陆面过程特征随地域和季节的变化较大,并且明显受到自然变迁及人类生活生产活动所引起的土地利用和地表覆盖方式改变的影响,因而其陆面过程在全球气候系统中占有十分重要的地位。现有的大部分陆面过程模式对干旱半干旱区典型下垫面的陆面过程模拟仍然存在较大偏差,并且陆面过程观测实验中存在的能量不闭合等问题更是增加了陆面过程模拟研究的复杂性。因此如何改进陆面过程参数化方案,发展更适用于干旱半干旱区的陆面过程模式是陆面过程研究中的重要任务。为了提高陆面模式对干旱半干旱区典型下垫面的模拟性能,本文从以下三个方面进行了研究。首先发展了一个陆面过程模式TBLSHAW,并利用兰州大学半干旱气候与环境观测站(SACOL)观测资料进行验证；其次针对陆面观测实验中普遍存在的近地层能量不闭合问题,利用陆面过程模式TBLSHAW和SACOL站近地层观测资料,研究了半干旱区近地层能量闭合度对陆面过程模式的影响；最后,鉴于干旱区地表大量覆盖地膜抑制蒸发的事实,利用西北干旱区开展的“古浪非均匀下垫面近地层观测实验(GHUSLE)"中地膜覆盖农田下垫面的观测资料,建立了相应的参数化方案,发展了TBLSHAW_MULCH模式,并对干旱地区地膜覆盖农田下垫面的陆面过程特征进行了模拟研究。主要结论如下：(1)基于陆气间物质和能量平衡原理,在SHAW模式和CoLM模式的基础上发展了TBLSHAW模式。该模式吸收了CoLM模式中的双大叶模型用于计算植被与大气间的物质和能量平衡,详细考虑了植被层辐射传输,降水截留,蒸发蒸腾以及光合作用等物理过程；利用通用的土壤水热耦合传输模型计算土壤温度和湿度,并包含了土壤冻融,蒸发及降水渗透等过程；结合由SACOL站长期观测的土壤温湿及近地层资料发展而来的适用于半干旱区的部分陆面参数化方案,最后采用SHAW模式的动力框架,形成了一个土壤-植被-大气相互作用的陆面过程模式。(2)利用SACOL站观测资料,通过模拟黄土高原半干旱区典型干季(裸土)和湿季(草地)陆面过程特征以及土壤冻融过程验证了TBLSHAW模式的模拟性能,通过对比分析观测资料和不同模式的模拟结果,发现TBLSHAW模式改善了该地区土壤水热传输过程的模拟能力。TBLSHAW模式能较好的模拟黄土高原半干旱区干湿两季的净短波和净长波辐射以及土壤温度和湿度,并能合理的反映黄土高原地区的能量分配状况,模式模拟值与相应的观测值偏差较小,模式效率较高；TBLSHAW模式能较好的模拟土壤冻融过程中土壤温度和湿度的变化,模拟的土壤温度和湿度与相应的观测值偏差较小。(3)针对干旱半干旱区(甚至大部分地区)观测的近地层能量不闭合问题,利用2007年春季SACOL站的近地层观测资料和TBLSHAW模式,通过数值模拟和对比分析方法研究了半干旱区近地层能量闭合度对陆面过程模式的影响。结果表明：在半干旱区近地层能量不闭合将通过近地层湍流参数化方案传导到陆面过程模式中,当湍流通量低于可利用能量时,能量不闭合将使得陆面过程模式模拟的地表长波辐射和土壤温度明显较观测值偏高；而将实际观测中能量不闭合部分以某种形式分配到观测的湍流通量中,即修正观测的湍流通量使之达到近地层能量平衡,并依据修正的湍流通量发展湍流参数化方案,在不改变任何地表土壤物理生化属性的情况下,陆面过程模式能较好地模拟地表长波辐射和土壤温度。(4)以GHUSLE实验中地膜覆盖农田下垫面观测为基础,发展了一个地膜层子模型,并建立了相应的陆面参数化方案,发展了TBLSHAW_MULCH模式,利用该模式模拟了地膜覆盖农田下垫面的陆面过程特征；同时利用相同的大气强迫数据运行TBLSHAW模式作为数值对比试验,研究了地膜覆盖对农田陆面过程特征的影响。结果表明：TBLSHAW_MULCH模式能够较好的模拟地膜覆盖农田下垫面的各项陆面过程特征,模式效率较高,模拟值与观测值的偏差较小。通过与TBLSHAW模式模拟的陆面过程特征对比发现,地膜覆盖显著地影响了陆面过程,因此在模拟干旱区地膜覆盖下垫面的陆面过程时应该考虑地膜的作用。更多还原

【Abstract】 The arid and semi-arid regions are nearly40%of the global land area, its surface cover are diverse and complex, including various landscapes such as grassland, farmland, desert, Gobi, and permafrost. For these complicated underlying surface, their characteristics of land surface process varies with area and season obviously, and is significantly affected by natural changes and the changes of land use and land cover which is caused by the human activities, and thus their land surface process plays a very important role in the global climate system. But for the most of present land surface process models, there is still existence of a large deviation in land surface process simulation over the typical underlying surface in the arid and semi-arid areas. Furthermore, there is the energy imbalance in the surface layer in observation experiment; this makes the land surface process simulation more difficult. How to improve the land surface parameterization schemes, and develop land surface process model more applicable to the arid and semi-arid regions is an important task in the study of land surface processes.In order to improve the performance of land surface model simulation over the typical underlying surface in the arid and semi-arid areas, the land surface model was studied from three aspects in this paper. Firstly, a land surface process model TBLSHAW was developed and verified with the observation data in the Semi-Arid Climate and Environment Observatory of Lanzhou University (SACOL). Secondly, focusing on the influence of the energy imbalance, which was commonly existed in the land surface observation experiment, the impact of energy imbalance upon land surface process simulation in semi-arid region were studied by using the SACOL surface layer observation data and land surface model TBLSHAW. Finally, in view of the fact that the large number of arid areas surface was covered by plastic mulch to inhibit evaporation, a new model TBLSHAW_MULCH was developed by improving the parameterization scheme with the observation at a cropland underlying surface covered by plastic mulch in the Gulang Heterogeneous Underlying Surface Layer Experiment (GHUSLE) in the arid region to simulate the land surface process of the cropland underlying covered by plastic mulch in arid region. We draw the following conclusions:(1) Based on the energy and material balance principle between land surface and atmosphere, a land surface model TBLSHAW was developed by means of SHAW and CoLM models. The TBLSHAW model, which was a land surface model including soil-vegetation-atmosphere interactions, absorbed the two-big-leaf model to calculate the material and energy balance between the vegetation and the atmosphere and to consider the vegetation canopy radiation transfer, precipitation interception, evaporation, transpiration, photosynthesis and other physical processes in detail; It used the general soil heat and water coupling transfer model to calculate soil moisture and temperature, and included the process of soil freezing and thawing, evaporation and precipitation infiltration; It combined part of the land surface parameterization schemes which was developed using the long-term observations of soil temperature and moisture and surface layer at SACOL; and it employed SHAW model framework.(2) Using observational data in SACOL, the performance of the TBLSHAW model was verified by simulating the land surface process of the typical dry (bare soil), wet season (grass land), freezing and thawing. Comparing among the observation and the simulations by TBLSHAW and different land surface process models, the results indicated that TBLSHAW model improved the simulation of soil water and heat transfer process in SCAOL. The TBLSHAW model could simulate the net shortwave, net longwave, soil temperature and moisture well, and could reasonably reproduce the energy distribution of the Loess Plateau both in dry and wet seasons. The simulation was close to the observation, and the model was efficient; The TBLSHAW model could also well simulate soil freezing and thawing process, and the bias between simulated soil temperature and the observation was small, so as the soil moisture.(3) In view of the fact that the observed surface layer energy was not closed in the arid and semi-arid regions (even the most of the global land surface), the TBLSHAW model was employed to study the impacts of energy closure ratio upon land surface simulations in semi-arid region through numerical simulation and comparative analysis method with the observation at SACOL in the spring,2007. The results showed that the observed energy imbalance could conducted into land surface process model by turbulent parameterization scheme in semi-arid area, when the sum of turbulent fluxes were lower than the available energy, the energy imbalance would make the land surface process model significantly overestimated the surface longwave radiation and soil temperature; when the actual observed energy imbalance part distributed into observed turbulent fluxes in some way, that meant correcting the observed turbulent fluxes so as to make the surface energy balance, and the turbulent fluxes parameterization scheme was developed based on the modified turbulent heat flux, then the land surface process model could well simulate the surface longwave radiation and soil temperature without changing any physical and biochemical properties of soil.(4) Based on the observation at a cropland underlying surface covered by plastic mulch in the GHUSLE, a new model TBLSHAW_MULCH was developed by coupling a plastic mulch sub-model to simulate the land surface process of the cropland underlying covered by plastic mulch in arid region; A comparative run the TBLSHAW model was performed to investigate the effect of the plastic mulch. The results showed that TBLSHAW_MULCH model could well simulate the land surface process characteristics of the cropland underlying surface covered by plastic mulch. And the model was efficient and the deviations between simulated and observed values were small. Compared the simulation of the TBLSHAW model with that of the TBLSHAW_MULCH model, it was suggested that the plastic mulch obviously affected the land surface process, and consequently the role of the plastic mulch should be considered in the land surface model.更多还原