Watershed scale temporal stability of soil water content
Introduction
Soil water content is a key variable controlling water and energy fluxes in soils (Vereecken et al., 2007). It acts as one crucial factor for vegetation production, mainly in semi-arid environments. It varies not only in space but also in time, however, a large body of data has proven to follow the so called concept of temporal stability (Grayson and Western, 1998, Comegna and Basile, 1994, Cassel et al., 2000, Mohanty and Skaggs, 2001, Thierfelder et al., 2003, Martínez-Fernández and Ceballos, 2003, Martínez-Fernández and Ceballos, 2005, De Lannoy et al., 2006, Teuling et al., 2006, Lin, 2006, Brocca et al., 2009, Brocca et al., 2010), which was described as the time invariant association between a spatial location and classical statistical parameters by Vachaud et al. (1985).
Temporal stability of soil water content has been mostly observed during the soil moisture sampling campaigns aimed at the validation of remotely sensed mean soil moisture estimation (Jacobs et al., 2004, Cosh et al., 2006, Starks et al., 2006, Cosh et al., 2008, Vivoni et al., 2008). Therefore, most efforts on temporal stability analysis of soil water content have focused the surface soil layer and very few reports refer to the whole soil profile, exception made for the very latest contributions of Martínez-Fernández and Ceballos, 2005, Pachepsky et al., 2005, Starks et al., 2006, De Lannoy et al., 2006, and Guber et al. (2008). Temporal stability has been widely recognized over different areas worldwide, e.g. Ontario-Canada (da Silva et al., 2001), Spain (Martínez-Fernández and Ceballos, 2003), USA-Oklahoma (Cosh et al., 2004), Italy (Brocca et al., 2009). With exception to the recent work by Hu et al. (2009), this concept has not yet been widely applied to measurements taken on the Loess Plateau in China, where the water content of the different soil depths is the most crucial factor for vegetation restoration. In fact, the study of temporal stability of soil water contents for different soil depths on the sloping land of this and other similar watersheds can be very important for their monitoring, mainly related to soil water management during the process of vegetation restoration.
With the aim of identifying locations of temporal stability of soil water content, many studies have been conducted to look for the contributing factors of this temporal stability. Grayson and Western, 1998, Vivoni et al., 2008 believed that the best locations to represent the mean soil water content of a catchment should be the locations which capture the average characteristics of that catchment, e.g., near to mid slopes or mid aspects (Grayson and Western, 1998) or mid elevation (Vivoni et al., 2008). Gómez-Plaza et al. (2000) observed that at the transect scale, when the factors affecting soil water content were limited to topographical position or local topography, spatial patterns presented time stability, which was in agreement with the findings of Thierfelder et al., 2003, Brocca et al., 2009. On the other hand, Tallon and Si (2003) found poor relationships between temporal stability locations and soil and topographic properties. Jacobs et al. (2004) observed that sampling locations with moderate to moderately high clay content tended to have a more pronounced time stability, whereas Mohanty and Skaggs (2001) believed that sandy loam soils were more time stable than silt loams. Therefore, no consistent conclusions have been drawn on contributing factors to temporal stability. With the main aim of exploring the effects of calibration procedures on the temporal stability of neutron probe soil water storage, Hu et al. (2009) found that soil particle size and organic matter content were the main factors influencing temporal stability, but their conclusions are made based on soil water storage data from only 12 sampling locations. Thus, besides the inherent differences of contributing factors to temporal stability, the subjective choice of factors and the low number of samples may also explain the differences of contributing factors.
The spatial pattern of soil water content has been well recognized (Western et al., 1998a, Western et al., 1998b, Bárdossy and Lehmann, 1998, Wendroth et al., 1999, Walker et al., 2001, Wang et al., 2001, Western et al., 2004, Brocca et al., 2007, Hu et al., 2008), and several fields were proven to have soil water contents spatially variable in an organized way. In addition, the spatial patterns of temporal stability indicators were also discussed by some researchers (Petrone et al., 2004, Zhou et al., 2007, Brocca et al., 2009, Williams et al., 2009). To date, however, no report referred to the geostatistical analysis of the spatial pattern of temporal stability indicators. Sufficient sampling numbers required for the geostatistical techniques may be one of the limits for the analysis of their spatial patterns. In fact, spatial pattern analysis of temporal stability by geostatistical techniques can be very important for the identification of stable locations for mean soil moisture estimation in a watershed. If the temporal stability has a sound spatial structure, then the locations close to the observed stable locations should also tend to be stable. In this way, stable locations would not only be limited to a given sampling location, but the stability could be extended to a broader area, which would greatly facilitate the identification of stable locations.
The standard deviation of relative difference (SDRD) was widely used to identify the temporally stable sampling locations (Grayson and Western, 1998, Gómez-Plaza et al., 2000, Cosh et al., 2006, Cosh et al., 2008, Brocca et al., 2009). In these situations, a temporal stable location is characterized by a low value of SDRD. Other indices that judge the temporal stability can also be found. Jacobs et al. (2004) introduced the root mean square error (RMSE) of the relative differences, which is computed from the mean relative difference and associated variance to identify the best representative locations. According to Jacobs et al. (2004), the location with the most pronounced time stability is identified as the one with the lowest RMSE. Guber et al. (2008) employed a new temporal stability index Tik, which was computed as the width of the 90% empirical tolerance interval of empirical probability distribution functions of relative soil water content. In this case, lower values of Tik correspond to more pronounced temporal stability. Furthermore, Guber et al. (2008) used the root-mean-squared differences, Dik, to select the locations that appeared to be the best for estimating the average water contents at different depths. With these indices, a temporal stable location can be identified for the mean soil water content estimation for a scale of interest. However, they cannot be directly related to the estimation error with the exception of Dik developed by Guber et al. (2008). From the point of view of the mean soil moisture estimation when considering the constant relative differences of stable locations as suggested by Grayson and Western, 1998, Hu, 2009 and Hu et al. (2010) developed an index of mean absolute bias error (MABE) to identify the temporal stable locations. The lower MABE refers to a stronger stability. In this sense, the MABE was not only an index for the identification of stable locations, but also a direct reflection of the error of the mean soil water content estimation for a given period of time.
This study explored the temporal stability of soil water content for various layers (0–0.8 m) measured over one year in a small watershed on the Loess Plateau in China. Two indices, SDRD and MABE were compared to judge the temporal stability. Specific concerns were: 1) the relationship of temporal stability with soil depth; 2) the effects of soil texture and land use on temporal stability; and 3) the spatial pattern of temporal stability.
Section snippets
Watershed description
A small watershed called LaoYeManQu (LYMQ, about 20 ha) was selected inside the larger Liudaogou watershed (110°21′ to 110°23′ E and 38°46′ to 38°51′ N), Shenmu County, Shaanxi Province, China (Fig. 1). This area is particular for its special environmental conditions, where severe soil and wind erosions concur. The Liudaogou watershed is characterized by a large number of deep gullies and undulating loess slopes. Climatic conditions are characterized by a mean annual temperature of 8.4 °C, and a
Soil water content dynamics
From June 11 to August 22 in 2007, a large fluctuation of soil water contents expressed as the mean volumetric soil water contents among all the locations was recognized for the soil depths of 0.1 m and 0.2 m (Fig. 3) as influenced by precipitation measured by a weather station about 50 m away from the northwest of the watershed, while for the deeper soil depths, the averaged soil water content showed only a slight decrease under the combined effects of precipitation and soil water uptake by
Conclusions
- (1)
Temporal stability reflected by the standard deviation of relative differences (SDRD) was generally more pronounced for the dry areas, however when characterized by the mean absolute bias error (MABE) tended to be more pronounced for the wet areas. Therefore, from the point of view of mean soil water content estimation, considering the constant offset of a temporally stable location as suggested by Grayson and Western (1998), it is considered advisable to choose the location with the minimum
Acknowledgements
The study was financially supported by the National Key Basic Research Project (2007CB106803), the “Innovative Team” program of the Ministry of Education (IRT0749), the “Innovative Team” program of the Chinese Academy of Sciences, the Knowledge Innovation Project (KZCX2-XB2-13 & KSCX2-YW-N-003) of Chinese Academy of Sciences, and the Open Fund of State Key Laboratory of Soil Erosion and Dryland Farming on the Loess Plateau (10501-269). We thank the two anonymous reviewers for their constructive
References (52)
- et al.
Spatial distribution of soil moisture in a small catchment. Part 1: geostatistical analysis
Journal of Hydrology
(1998) - et al.
Soil moisture temporal stability over experimental areas in Central Italy
Geoderma
(2009) - et al.
Soil moisture spatial variability in experimental areas of central Italy
Journal of Hydrology
(2007) - et al.
Temporal stability of spatial patterns of soil water storage in a cultivated Vesuvian soil
Geoderma
(1994) - et al.
Watershed scale temporal and spatial stability of soil moisture and its role in validating satellite estimates
Remote sensing of Environment
(2004) - et al.
Temporal persistence and stability of surface soil moisture in a semi-arid watershed
Remote Sensing of Environment
(2008) - et al.
Temporal stability of surface soil moisture in the Little Washita River watershed and its applications in satellite soil moisture product validation
Journal of Hydrology
(2006) - et al.
Spatial and temporal characteristics of soil moisture in an intensively monitored agricultural field (OPE3)
Journal of Hydrology
(2006) - et al.
Geostatistical analysis of soil moisture variability in grassland
Journal of Arid Environments
(2004) Geostatistics in soil science: state-of-the-art and perspectives
Geoderma
(1999)
Towards areal estimation of soil water content from point measurements: time and space stability of mean response
Journal of Hydrology
Temporal stability in soil water content patterns across agricultural fields
Catena
Time stability of soil water storage measured by neutron probe and the effects of calibration procedures in a small watershed
Catena
SMEX02: field scale variability, time stability and similarity of soil moisture
Remote Sensing of Environment
Evaluating impact of irrigation water quality on a calcareous clay soil using principal component analysis
Geoderma
Mean soil moisture estimation using temporal stability analysis
Journal of Hydrology
Spatio-temporal evolution and time-stable characteristics of soil moisture within remote sensing footprints with varying soil, slope, and vegetation
Advances in Water Resources
Gstat: a program for geostatistical modeling, prediction and simulation
Computers & Geosciences
Spatial variability of soil moisture content and its relation to environmental in indices in a semi-arid gully catchment of the Loess Plateau, China
Journal of Arid Environments
Temporal stability of soil moisture profile
Journal of Hydrology
Inferring the location of catchment characteristic soil moisture monitoring sites. Covariance structures in the temporal domain
Journal of Hydrology
Comparison of ground-based and remotely-sensed surface soil moisture estimates over complex terrain during SMEX04
Remote Sensing of Environment
Spatial variability of soil total nitrogen and soil total phosphorus under different land uses in a small watershed on the Loess Plateau, China
Geoderma
Spatio-temporal patterns and covariance structures of soil water status in two northeast German field sites
Journal of Hydrology
Geostatistical characterisation of soil moisture patterns in the Tarrawarra catchment
Journal of Hydrology
Spatial correlation of soil moisture in small catchments and its relationship to dominant spatial hydrological processes
Journal of Hydrology
Cited by (200)
Estimation of soil water storage according to its multi-scale correlations with environmental factors
2024, Soil and Tillage ResearchImproved downscaling of microwave-based surface soil moisture over a typical subtropical monsoon region
2023, Journal of HydrologyEstimation of root zone soil moisture at point scale based on soil water measurements from cosmic-ray neutron sensing in a karst catchment
2023, Agricultural Water ManagementVariation of soil water stability and dominant influencing factors after sparse forestation in the loess alpine hills of northwest China
2023, Journal of Hydrology: Regional StudiesScale-specific controls of soil water storage along a transect in a semiarid catchment
2023, European Journal of Agronomy