论文部分内容阅读
近年来,青藏高原多年冻土区生态环境呈现出逐年恶化趋势,从而对多年冻土活动层水热过程造成显著影响.此外,如何构建更加有效、针对寒区的陆面过程模式成为寒区研究的重点、热点之一.作为一种有效的参数估计方法,Bayes参数估计算法具有准确估计陆面过程模式参数的能力.因此,基于2005—2008年观测数据,利用CoupModel模型对青藏高原风火山流域土壤水热运移过程进行模拟;同时,利用Bayes参数估计方法估计部分水热运移参数.结果显示:模型对土壤温度(ST)的模拟效果较好,NSE系数均在0.90以上;也能够较好模拟浅层(0～40cm)土壤水分,NSE值均达到0.80以上,而对40cm以下土壤水分的模拟结果较差.模型也能够较准确模拟活动层土壤的冻结-融化过程.模型对温度水分极值和40cm深度以下水分的模拟存在一些偏差.值得一提的是,基于重要性采样MCMC方案的Bayes参数估计算法能够有效估计水热运移参数,模型模拟结果得到极大的改进.Bayes算法能够广泛解决陆面过程模式参数估计问题.
In recent years, the ecological environment in the permafrost regions of Qinghai-Tibet Plateau shows a trend of worsening year by year, which has a significant impact on the hydrothermal processes in the permafrost active layer.In addition, how to construct a more effective and cold-adapted land surface process model to become a cold region As an effective parameter estimation method, Bayesian parameter estimation algorithm has the ability to accurately estimate the parameters of the land surface process.Therefore, based on the observed data from 2005 to 2008, the Windmount Volcanic Watershed Simultaneously, Bayesian parameter estimation method was used to estimate some hydrothermal transport parameters. The results showed that the simulation results of soil temperature (ST) were better and the NSE coefficients were both above 0.90. The soil water and NSE values of the well (0 ~ 40cm) in the simulated shallow layer (0-40cm) all reach above 0.80, while the simulation results of the soil moisture below 40cm are poor.The model can also accurately simulate the freeze-thaw process of the soil in the active layer.The model is sensitive to temperature and moisture There are some deviations between the extreme values and the simulation of the moisture below 40 cm depth.It is worth mentioning that the Bayesian parameter estimation algorithm based on importance sampling MCMC scheme can Effective estimation of hydrothermal transport parameters, model simulation results have been greatly improved.Bayes algorithm can solve a wide range of land surface process model parameter estimation.