残膜对土壤水分入渗和蒸发的影响及不确定性分析

为探究残膜对土壤水分入渗和蒸发过程的影响规律,通过室内土柱试验,设置6个残膜量水平(0、80、160、320、640和1 280 kg/hm2),研究了残膜对湿润锋运移、土壤水分分布、累积入渗量和累积蒸发量及其不确定性的影响。结果表明:随残膜量增加,湿润锋垂直运移速率和累积入渗量逐渐减小;残膜量80 kg/hm2时,湿润锋运移速率大幅下降;累积蒸发量随残膜量增加而递减而蒸发系数呈递增趋势,土壤保水能力减弱;随残膜量增加,0~10和20~45 cm含水率呈降低趋势,而土壤水分的变异系数呈增加趋势,残膜加剧了土壤水分垂直分布的变异性,残膜量320 kg/hm2的处理会出现表土层"板结"现象;基于Gibbs抽样算法分析表明,Kostiakov入渗模型和Rose蒸发模型各参数的95%后验置信区间上下限的差值和标准差均随残膜量增加而增大,累积入渗量和累积蒸发量的95%后验置信区间面积呈增大趋势,土壤累积入渗量和累积蒸发量的不确定性随残膜增多而增强。该研究可探明残膜污染区的土壤水分运动规律,并为提高Kostiakov模型、Rose模型的模拟效率和模拟精度提供参考。

Effects of residual plastic film mixed in soil on water infiltration, evaporation and its uncertainty analysis

Pollution of residual plastic film, a continuous pollutant and difficult to degrade, is a major limiting factor for sustainable development of agriculture in northwest China. Residual plastic film can destroy homogeneity of soil texture and seriously impede the movement of soil water and solute, and thus greatly enhance the uncertainties in soil water movement. In order to reveal the negative effects of residual plastic film on soil water infiltration, evaporation and their uncertainties, soil column simulation experiments were conducted to observe the processes of soil water infiltration and evaporation with different amounts of residual plastic film in the Key Laboratory of Agricultural Soil and Water Engineering in Arid Area (108°02′E, 34°17′N), at Northwest A&F University, in Yangling, Shaanxi Province of China, from September to October, 2015. Six amounts of residual plastic film were designed including 0, 80, 160, 320, 640, 1 280 kg/hm2 with triplicate. During the experiment, the wetting front, the cumulative infiltration and the cumulative evaporation were investigated. Meanwhile, gravimetric soil moisture was measured after the evaporation process. To mimic the actual distribution characteristic of residual plastic film in field, fragile plastic film and soil samples were mixed evenly with a blender. During infiltration, when the wetting front arrived at 40 cm, irrigation water was cut off. Soil columns were covered with waterproof plastic film to reduce atmospheric evaporation. Soil columns stood for 12 h until the infiltration process completely stopped. In the consequent process of soil water evaporation, the columns were continuously heated by infrared lamps of 275 W from morning till night. Soil columns were weighed with an electronic balance at 9:00 a.m. every day to calculate mass change and standard daily evaporation. Furthermore, an evaporation pan that had the same diameter as soil column was used to measure daily evaporation from free water surface. Uncertainty analyses were conducted for the simulations with Kostiakov model and Rose model. The uncertainties of soil cumulative infiltration and evaporation under different treatments were quantitatively analyzed with the Gibbs sampling algorithm. The results showed that residual plastic film would not only reduce the transport velocity of vertical wetting front, but also decrease soil cumulative infiltration. When the amount of residual plastic film was more than 80 kg/hm2, transport velocity of wetting front decreased dramatically. As the amount of residual plastic film increased, consumed time that wetting front arrived at 40 cm significantly prolonged (F=19.11, P<0.05). On the other hand, cumulative evaporation significantly decreased when the amount of residual plastic film was more than 160 kg/hm2 (F=9.58, P<0.05). However, evaporation coefficient increased with the increment of residual plastic film, which indicated that the water holding capacity of soil decreased. Residual plastic film also changed the distribution of soil moisture. As the amount of residual plastic film increased, gravimetric soil moisture decreased at the 0-10 cm and 20-45 cm depth. At 0-10 cm depth, variable coefficient of soil moisture increased by 4.33%, 127.19%, 330.40%, 539.63% and 640.23% compared with that for no residue treatment, respectively. Moreover, the variable coefficient had a trend of decrease at 20-45 cm depth. This implied that the inhomogeneity of soil water distribution was enhanced due to residual plastic film pollution. Moreover, soil surface hardened when the amount of residual plastic film was more than 320 kg/hm2. As the amount of residual plastic film increased, the 95%confidence interval and standard error of posterior parameters all increased for the Kostiakov infiltration model and Rose evaporation model according to Gibbs sampling algorithm. The areas of posterior distribution intervals of cumulative infiltration and evaporation also increased. In general, residual plastic film could increase the uncertainties in soil water infiltration and evaporation. The understanding of the influences of residual plastic film on the processes of soil water infiltration and evaporation and their uncertainties would provide a good reference for the exploration of soil moisture movement and distribution in fields polluted by residual plastic films, and improve the efficiency and accuracy of simulations with the Kostiakov and Rose models.