气象因子对太湖地区旱作农田土壤水分动态的影响

【目的】研究太湖地区气象因子对旱作农田土壤水分的影响程度，厘定影响农田土壤水分的主要气象因子，为气候变化背景下农田水分管理提供科学依据。【方法】提取野外试验研究平台的旱作物生长季监测数据（逐日气象资料和土壤水分资料），采用相关分析、逐步回归分析和通径分析等方法进行统计分析，计算直接通径系数和间接通径系数、决策系数。【结果】（1）供试实验基地农田土壤水分与日降水量、日蒸发量、日照时长、平均风速及日最大空气湿度等因子分布呈极显著正相关和负相关（相关系数分别为0.648、-0.566、-0.454、-0.331及0.371），但与日最高气温不显著相关；（2）通径分析表明，气象因子对旱作农田土壤水分的直接影响大小顺序为：日降水量＞平均风速＞日照时长＞日蒸发量＞日平均空气湿度＞日最低气温>日最小空气湿度＞日最高气温＞日最大空气湿度＞日平均气温，但计算的决策系数表明，日降水量对土壤水分的综合决定能力最大；（3）通过逐步回归分析，得到了旱田土壤水分的气象因子多元回归模型：Y=10.174+0.386X4+1.095X7-0.509X8-0.766X9-0.345X10（R2=0.912，P<0.01），达极显著水平。【结论】气象因子对太湖地区旱作农田土壤水分具有显著的控制作用，其中降雨量的影响最为主要。建立的多元回归模型可以用来预估气象因子变化下旱作农田土壤水分变化，但还需要更长时期监测的验证。

Influence of Meteorological Factors on Soil Moisture Dynamics of Upland Soil in Taihu Lake Region

【Objective】The purpose of this study is to integrate and then select the main meteorological factors of soil moisture by analyzing the effects of meteorological factors on soil moisture in dry farmland, and to provide a scientific basis for water management under the climate change conditions. 【Method】 The monitoring data (including the daily meteorological data and the soil moisture data) of drought crop growing season on the field experimental platform were extracted, and analyzed by using the correlation analysis, stepwise regression analysis and path analysis methods. The direct path coefficient, indirect path coefficient and determination coefficient between meteorological factors and soil moisture were calculated. 【Result】 The soil moisture of experimental base was highly positively correlated with daily precipitation and daily maximum air humidity, and was extremely negatively correlated with daily evaporation, sunshine duration and average wind speed (The correlation coefficients were 0.648, 0.371, -0.566, -0.454 and -0.331, respectively), but no significant correlation was observed between the soil moisture and the daily maximum air temperature. The path analysis showed that the decreasing influence sequence of meteorological factors on soil moisture was as follows: daily precipitation, average wind speed, sunshine duration, daily evaporation, daily mean air humidity, daily minimum air temperature, daily minimum air humidity, daily maximum air temperature, daily maximum air humidity and daily mean temperature, and that the daily precipitation had the largest comprehensive influencing ability on soil moisture. The multiple regression function between soil moisture and meteorological factors was Y=10.17+0.386X4 +1.095X7-0.509X8-0.766X9-0.345X10, reached a significant level (R2 = 0.912, P＜0.01). 【Conclusion】The effect of meteorological factors on controlling the soil moisture of upland soil in Taihu Lake plain was obvious, and the daily precipitation was the most important factor. The established multiple regression models could be used to forecast the changes of soil moisture in upland soil when some or all of the meteorological factors vary, but it needs to be further validated with longer-term monitoring data.