| Soil moisture is one of the most important state variables in land surface processes,which controls the exchange of water and energy between the land-atmosphere interface and influences the regional climate.Because soil moisture is a slowly varying variable,researches on soil moisture memory(SMM,also known as long-range correlation)are helpful to understand land-atmosphere interactions and to improve the accuracy of climate system predictions.At present,many studies have estimated SMM based on numerical models.However,owing to the lack of in situ observational data,there were great differences in SMM among various models.To this end,in this study,both in situ observed data and model simulated data were used to analyze SMM characteristics and its influencing factors.Firstly,long-term observed soil moisture data of China,New Zealand and North American were downloaded from the International Soil Moisture Network(ISMN).Then a 1-D vadose zone model(Hydrus-1D)was utilized to simulate soil moisture at selected sites in China(Guangzhou,Zhengzhou,Yinchuan)using a generated soil hydraulic parameter dataset for sandy and clay loam soils.Secondly,by conducting a detrend fluctuation analysis of soil moisture,the characteristics of SMM and the main factors affecting SMM were discussed.Finally,considering the key role of soil moisture in land-atmosphere coupling,the soil moistureprecipitation coupling strength(SM-P)was quantified by feedback parameter based on in situ observational data.Then the correlation between the SMM and SM-P was also investigated.The main conclusions are as follows:(1)Based on in situ observational data and model simulated data,the scale exponents(?)of SM from DFA were obtained.The results showed that all values of ?were greater than 0.5,indicating the existence of the SMM.In China,SM series mainly exhibited persistence characteristics(0.5<?<1).In New Zealand and in North American,SM series mainly showed anti-persistence(1<? <1.5).The results of simulated data showed that SM series was mainly characterized by persistence for sandy,but for clay loam,there were both persistence and anti-persistence.Moreover,SMM increased with the increase of soil depth,aridity index and clay fraction.(2)Based on in situ observational data,the factors influencing SMM were analyzed.The results showed that the main influencing factors varied among regions.In China,the main factor influencing SMM was potential evapotranspiration;in New Zealand,the main factor influencing SMM was precipitation;by comparison,in North American,the main factor influencing SMM was clay fraction at 10 cm depth,and SMM was mainly affected by precipitation at 50 cm and 100 cm depth.This result indicated that SM had significantly different responses to external factors under different climatic conditions.(3)Based on in situ observational data,the SM-P was analyzed.The results revealed that both regions were dominated by positive feedback parameters in North American and New Zealand,which means that SM can promote the formation of precipitation.The SM-P coupling strength in Nebraska(a state in North American)was strong,while it was weak in New Zealand.In addition,it was found that there was a significant negative correlation between SMM and SM-P in New Zealand and North American at surface layer(e.g.,10cm).(4)Based on model simulated data,the main soil hydraulic parameters affecting SMM were studied.The results showed that the main soil hydraulic parameters affecting SMM were 9),,log.The relative contribution of theses parameters varied under different climatic and soil texture conditions.Specifically,9)was the main hydraulic parameter affecting SMM for sandy soils at all regions.For clay loam soils, was the predominated hydraulic parameter affecting SMM in Guangzhou,and and log were the main hydraulic parameters affecting SMM in Zhengzhou and Yinchuan. |
