Abstract
Knowledge of the variability of soil water content (SWC) in space and time plays a key role in hydrological and climatic modelling. However, limited attention has been given to arid regions. The focus of this study was to investigate the spatio-temporal variability of surface soil (0–6 cm) water content and to identify its controlling factors in a region of the Gobi Desert (40 km2). The standard deviation of SWC decreased logarithmically as mean water content decreased, and the coefficient of variation of SWC exhibited a convex upward pattern. The spatial variability of SWC also increased with the size of the investigated area. The spatial dependence of SWC changed over time, with stronger patterns of spatial organization in drier and wetter conditions of soil wetness and stochastic patterns in moderate soil water conditions. The dominant factors regulating the variability of SWC changed from combinations of soil and topographical properties (bulk density, clay content and relative elevation) in wet conditions to combinations of soil and vegetation properties (bulk density, clay content and shrub coverage) in dry conditions. This study has important implications for the assessment of soil quality and the sustainability of land management in arid regions.
Résumé
La connaissance de la variabilité de la teneur en eau du sol (TES) dans l’espace et le temps joue un rôle clé dans la modélisation hydrologique et climatique. Cependant, les régions arides ont fait l’objet d’une attention limitée. L’objectif de cette étude était d’étudier la variabilité spatio-temporelle de la teneur en eau superficielle du sol (0–6 cm) et d’en identifier les facteurs de contrôle dans une région du désert de Gobi (40 km2). Les résultats montrent que l’écart-type de la TES diminue de manière logarithmique avec la diminution de la teneur en eau moyenne, et que le coefficient de variation de la teneur en eau moyenne affiche un profil convexe vers le haut. La variabilité spatiale de la TES augmente aussi avec la taille de la zone étudiée. La dépendance spatiale de la TES change au fil du temps, avec des structures d’organisation spatiale plus fortes dans les conditions d’humidité du sol plus sèches et plus humides, et des structures aléatoires dans les conditions intermédiaires. Les facteurs dominants qui régissent la variabilité de la TES vont de combinaisons de propriétés pédologiques et topographiques (densité apparente, teneur en argile et élévation relative) dans les conditions humides, à des combinaisons de propriétés de sol et de végétation (densité apparente, teneur en argile et couverture arbustive) dans les conditions sèches. Cette étude a des implications importantes pour l’évaluation de la qualité des sols et la durabilité de la gestion du territoire dans les régions arides.
1 INTRODUCTION
The water content in surface soils plays a critical role in hydrology, agrology and ecology, and strongly controls various natural processes, such as energy fluxes of the land surface, soil degradation, land–atmosphere interactions, and conservation and restoration of vegetation (Charpentier and Groffman Citation1992, Wang et al. Citation2012, Jia and Shao Citation2013). Characterizing the spatio-temporal variability of soil water content (SWC) is a key challenge for improving hydrological and climatic modelling and prediction (Ampofo Citation2006), and has thus become increasingly important in recent years, from the scale of hillslopes (Famiglietti et al. Citation1998, Zhu and Shao et al. Citation2008, Penna et al. Citation2009) and small watersheds using ground-based measurements (Hébrard et al. Citation2006, Hu et al. Citation2010, Brocca et al. Citation2012, Takagi and Lin Citation2012) to national and global scales using remote sensing (IGPO Citation1995, Lu and Shi Citation2012). Remote sensing is promising for the description and measurement of surface SWC on large scales, but its usefulness and interpretation require further testing due to its relatively low spatial and temporal resolution (Brocca et al. Citation2007). Data obtained from accurate on-site measurements should thus continue to be important when studying the variability of SWC (Owe et al. Citation1982). SWC in arid regions is the key resource limiting the development of vegetation and is the main constraint to permanently controlling desertification (Berndtsson et al. Citation1996). Soil water also determines the organization and functioning of the ecosystems in arid regions. Despite its importance to vegetation, soil water has received little attention in arid environments, especially in the characterization of its variability in natural deserts, due to the difficulty of data collection and the associated costs.
The spatio-temporal variability of SWC has been characterized by testing its link to field (spatial) means of soil wetness. Knowledge of this relationship is useful for determining the minimum amount of sampling needed to effectively estimate field means of water content with an acceptable error, estimating the error if the amount of sampling has been predefined, and verifying and validating remotely sensed effects of SWC (Brocca et al. Citation2007). Whether the variability of SWC has a positive or negative relationship with field means of water content, however, has not been conclusively determined (Reynolds Citation1970, Owe et al. Citation1982, Zhang et al. Citation2011).
The variability and distribution pattern of surface SWC in time and space can be controlled by many factors, such as soil properties, topographical characteristics and vegetation (Qiu et al. Citation2001, Hu et al. Citation2008, Zhao et al. Citation2011, Takagi and Lin Citation2012). However, identifying the relative significance of these individual factors is challenging because of their mutual or multiple influences on SWC. Many studies have indicated that the spatial pattern of SWC, especially in arid regions, cannot be accurately predicted by any single factor. In addition, the dominant factors may change with the chosen sites and depend on the conditions of soil wetness resulting from seasonal wetting and drying (Grayson et al. Citation1997, Famiglietti et al. Citation1998, Western et al. Citation1999, Pan and Wang Citation2009). For example, Famiglietti et al. (Citation1998) explored the possible factors influencing the spatial pattern of SWC along the profile of a hillslope. In wetter conditions, the heterogeneity of soil properties (porosity, hydraulic conductivity) significantly contributed to the spatial variability of SWC, but in drier conditions, topographical characteristics (relative elevation and aspect) largely determined the variability of SWC. Western et al. (Citation1999) showed that the indices associated with the lateral flow of water and those related to evapotranspiration were the most important predictors of distributional patterns of SWC in wet and dry conditions, respectively. Hawley et al. (Citation1983) demonstrated that the variability of SWC was mainly driven by topography, but the explanatory power tended to decrease with seasonal variations of the vegetation cover. Gómez-Plaza et al. (Citation2001) found that some local controls (e.g. soil texture and slope angle) were the main regulators of the distribution of SWC in non-vegetated areas, whereas the controls operating on SWC in vegetated areas were those linked to the properties of the vegetation. The factors controlling the dynamics of SWC remain quite uncertain, and much more research is needed in a variety of environments and on a variety of scales.
Soil water content and its complex spatial distribution are very important in arid regions. We therefore studied the spatio-temporal variability of surface soil (0–6 cm) water content in a typical fenced region of the Gobi Desert where precipitation is the only source of soil moisture. The main goals of the study were: (a) to quantify the spatio-temporal variability of the surface SWC, (b) to understand the relative roles of elevation, soil properties and vegetation in controlling the variability of SWC under different conditions of soil wetness, (c) to determine the amount of sampling needed to estimate field means of water content at a predefined level of statistical significance, and (d) to determine the relationship between the variability of SWC and the size of the sampling area using a re-sampling method.
2 MATERIALS AND METHODS
2.1 Description of the study area
The study was conducted in a region of the Gobi Desert (39°24′–39°28′N; 100°08′–100°11′E) in the central reaches of the Heihe River basin in Gansu Province, northern China (), ()). The elevation ranges from 1390 to 1470 m a.s.l., and the water table is at a depth of nearly 11–13 m (Yu et al. Citation2012). The area is characterized by low and highly variable rainfall, and has an arid desert climate. The average annual rainfall and air temperature are 117 mm year-1 and 7.6°C, respectively. Rainfall in brief summer showers contributes 65% of the total annual precipitation. The mean annual pan-evaporation is approximately 2390 mm year-1; this is 20 times greater than the annual precipitation. The zonal soil is classified as grey-brown desert soil and is derived from diluvial-alluvial materials. The soil profile is sandy in texture, low in nutrients and has a small amount of gravels in the surface and subsoil horizons. The above-ground plant cover is discontinuous and can be described as patches of sub-shrubs surrounded by bare areas. The study area has been fenced and is protected from grazing for the purpose of revegetation and reclamation. The dominant plant species are Nitraria sphaerocarpa Maxim. and Reaumuria soongorica (Pall.) Maxim., and the accompanying plant species are mainly Kalidium gracile Fenzl, Allium mongolicum Rgl., Bassia dasyphylla (Fisch. and Mey.) Kuntze, Halogeton arachnoideus Moq., Suaeda microphylla (Mey.) Pall., Caragana brachypoda Pojark., Salsola ruthenica Iljin, Asterothamnus centrali-asiaticus and Sympegma regelii Bunge.
2.2 Soil sampling and data collection
A regular sampling grid (500 m × 500 m) of 187 sampling points was established over a 40-km2 (5 km × 8 km) experimental area ()). Each sampling point was positioned with a portable Garmin GPS receiver (with a resolution of 3 m), and each point was marked by a wooden stick. The volumetric water content of the surface soil (0–6 cm) was measured with FDR probes (Type ML2x, Delta-T Devices) approximately every two weeks from 15 April to 15 October 2012. A total of 13 measurement campaigns were conducted. Measurements were taken in the same sequence to reduce the possible influence of sampling time. The probes were handled carefully to avoid touching the gravel, which can influence the measurements. To further reduce the possible influence of micro-scale variability, we averaged three measurements for each point and date. The reliability of the Theta Probe was enhanced by calibrating it with a site-specific gravimetric assessment. Calibration was performed at three times of year (May, August and October) to allow for variations in soil wetness. The calibrations had a root mean squared error of approximately 1.09%, indicating adequacy.
A quantitative survey of the vegetation was performed in mid-August 2012, the period of vigorous growth of the vegetation. Quadrats of 20 m × 20 m for shrubs and 2 m × 2 m for herbaceous species were established at each sampling point to investigate species richness and the coverage of herbs and shrubs. Species richness was defined as all occurrences of species in all quadrats. The coverage of the herbs and shrubs was calculated as a percentage of the area.
At each sampling point, disturbed soil samples of the surface layer (0–5 cm) were collected with a soil auger from five randomly selected positions. These samples were then pooled to produce one representative sample. The samples were air-dried, weighed and then sieved to 2 mm to separate the gravel (>2 mm) from the fine soil (<2 mm). The former was reweighed to determine the gravel content. The soil component was separated into two parts. One was analysed for particle size by laser diffraction with a Mastersizer 2000 (Malvern). The other was crushed and passed through a 0.25-mm sieve for determining the amount of soil organic carbon (SOC) with the dichromate-oxidation method (Nelson and Sommer Citation1982). Undisturbed soil core samples (100 cm3) were collected from the surface layer (0.5–5.5 cm) at each point for measurement of saturated hydraulic conductivity (KS), using the constant hydraulic head method (Klute and Dirksen Citation1986), and of soil bulk density (BD), based on the volume of each original soil core and the total weight of the soil after oven drying at 105°C for 48 h.
2.3 Statistical methods
To clarify the descriptions, we will first define the terms used herein:
‘micro-site’ is the ground location at which a measurement was taken (three measurements were done at each point);
‘point’ is the mean location of a group of micro-sites; and
‘sampling day’ represents an individual day on which several measurements were taken.
A descriptive statistical analysis was first applied to determine the overall trends and variation of the SWC. Let be the SWC at micro-site i and point j on sampling day k. The spatial mean at point j on sampling day k is then calculated as:
The coefficient of variation of each sampling day in space, CVk, and each sampling point in time, CVj, can consequently be defined as:
For each sampling day, the coefficient of variation of the sampling point, CVkpoint, was calculated as the mean of the coefficients of variation determined for each point as:
With an understanding of the spatial standard deviation (σk), we can quantify the minimum amount of sampling needed for estimating the field mean water content at a given level of confidence, which can be described by:
One-sample Kolmogorov-Smirnov (K-S) tests were used to examine the normality of the data. Semivariograms were then constructed to evaluate the spatial dependence of SWC (David Citation1977). Correlation analysis was conducted for each day to evaluate the possible relationships between SWC and relative elevation, soil properties and vegetation. Stepwise linear regression analysis was subsequently carried out to detect the relative significance of these factors influencing the dynamics of the SWC. All descriptive statistical analyses used the program SPSS 16.0. The geostatistical analysis was performed with GS+ software (version 7.0, Gamma Design Software).
3 RESULTS AND DISCUSSION
3.1 Temporal dynamics of SWC
The mean SWC had apparent seasonal changes and was tightly linked to rainfall (). The mean water content was higher in the rainy season (summer) than in dry seasons (spring and autumn) but never rose above 10%. Rain is the main source of soil moisture in this area because of the deep water tables. Because this area is characterized by low precipitation and high evapotranspiration, the soil is subjected to drought most of the time.
The mean SWC generally increased after a significant rainfall and decayed rapidly during dry periods. Relatively larger rainfalls always corresponded to higher SWCs, but this relationship was largely due to the timing of the sampling relative to the periods of rain. The higher SWCs occurred immediately after a rainfall event (e.g. 31 July and 15 August), with a lag following the rain. As seen in , the two most significant rainfall events occurred on 5 June (15.8 mm) and 27 June (14.4 mm), but did not give rise to the highest SWCs, which were on 18 June and 30 June, long after the rains. Rain wets the soil easily, and the water content accordingly increases quickly. The amount of water in the surface soil then decreases rapidly due to drainage to the sublayer and due to high evapotranspiration. The water content of the surface soil is to a very large extent influenced by the timing, amount and intensity of the rain. The influence of prior conditions of SWC, though, should also be considered. A wetter antecedent condition can produce greater water content when comparing rainfall events with the same amount and intensity (Pan et al. Citation2008). This phenomenon was not very evident in this study because of the low frequency of data collection.
3.2 Statistical analysis of SWC
3.2.1 Statistical parameters
The variability of SWC in space and time was surveyed by calculating the temporal mean of CV of the spatial SWC, CVk, and the spatial mean of CV of the time series of SWC, CVi. The value of CVk ranged from 26.57% to 63.78% (), with a mean of 46.09%, indicating that SWC in this region was moderately variable. Based on the results of Hills and Reynolds (Citation1969), CVk should remain above 5% for any specific field. In our study, all values were larger than 25%, which may have been due to the low amount of SWC in this area. In contrast, the mean CVi was 60.42% (ranging from 40.61% to 97.9%), which was obviously larger than the mean CVk, suggesting that the temporal variability of SWC should receive more attention than spatial variability in studies of SWC (Brocca et al. Citation2010). Coarse spatial sampling at a high frequency may thus be a good option for modelling hydrological processes in this region.
Statistical parameters were tested for the field mean water content, , to determine if the variability of surface soil water was related to the conditions of wetness. ) shows increasing standard deviation (σk) with increasing mean water content. This result is in accordance with previous studies by Famiglietti et al. (Citation1998), Western et al. (Citation1998) and Li et al. (Citation2013), but is contrary to the studies of Meyles et al. (Citation2003) and Brocca et al. (Citation2007), who stated that the relationship between σk and mean water content is linked to climatic conditions; in humid regions, σk is larger when the soil is dry, while in semi-arid regions, σk increases as the soil becomes wetter. Vereecken et al. (Citation2007) demonstrated that the change of σk with soil wetness was largely determined by the ability of soil to retain water and by the associated spatial variability.
An inherent characteristic of σk, though, is its strong dependence on , which usually restricts its application when comparing datasets with widely varying mean water contents. To overcome this difficulty, the relative variance, i.e. CVk, can be used, which can remove the influence of the different means. The CVk exhibited a general convex upward relationship with , with its highest value at a moderate level of wetness, as shown in ). The variability of SWC is mainly regulated by the wilting point in extremely dry conditions, the hydraulic conductivity in mid-range conditions and the soil porosity in wet conditions (Lawrence and Hornberger Citation2007). In moderate conditions, small patches of quickly drying soil (generally characterized by coarse texture) may coexist with patches that remain wet (generally characterized by fine texture), leading to heterogeneous conditions of wetness (Hills and Reynolds Citation1969). This situation, though, is not common, as demonstrated by the above studies that indicated a universal negative relationship between CVk and over all conditions of wetness. However, this negative relationship may be due to the limited periods of measurement or to the low frequencies of measurement in these studies, which would limit the data to an incomplete range of soil wetness. The relatively wetter climates of their study areas could further restrict the range of SWCs.
The relationship between σk and can be well described logarithmically as: , which can quantify the amount of sampling needed to determine at given confidence levels and relative errors. The sampling number as a function of for predefined confidence levels of 95% and 90% and relative errors of 5, 10 and 15% is shown in . The required amount of sampling generally exhibited higher values for the 95% than for the 90% confidence level and declined slightly as the allowable relative error increased from 5% to 15%. In accordance with other studies, these results indicated that many more samples are needed in relatively drier conditions for estimating in the study area. The maximum number of samples needed coincided with a water content of approximately 4.5%. The number of sampling points in our study (187) was sufficient to meet the requirement of a relative error of ±10% at a 95% confidence level, which had a maximum sampling number of approximately 150.
3.2.2 Variability of SWC and size of area
Estimating the variability of SWC over a wide variety of scales has significant implications for developing an effective scheme of data acquisition and monitoring. For each sampling date, the coefficient of variation of the sampling point (three measurements at each point), CVkpoint, was calculated as the mean of the coefficients determined for each point and was compared with CVk of the entire area. The value of CVkpoint ranged from 7.34% to 27.99% and averaged 17.12%, which was considerably lower than CVk. This comparison suggests an increasing trend of variability as the size of the investigated area increases. As displayed in , the coefficient of variation for the entire study area had a significant linear relationship with the coefficient at the point scale (i.e. CVk and CVkpoint), and the determining coefficient was as high as 0.78, implying that the change of spatial variability at the point scale follows a trend similar to that of the entire area: when the latter is high, so is the former.
To further detect the tendency of the variability of SWC to change with the size of the study area, a series of sampling-point allocations of different sizes were re-sampled using all sampling points (n = 187) in the study area (5 km × 8 km). For convenience, the east−west and north−south spans of the re-sampling area were assigned integral multiples of 1 km to generate five re-sampling options for the west−east span and eight options for the south−north span. The random combination of these two sets of options thus yielded 40 potential re-sampling methods. Some of these methods, though, were the same within the study area, so a final total of 24 types of re-sampling areas were obtained (see ). The variability of SWC for an area was first computed by averaging the CVk values of all possible re-sampling scenarios with the same area for each sampling date, and the resulting data were then averaged over the observation period. The averaged spatial CVk clearly increased with the size of area and could be well parameterized as a power function of the size ()). The factors influencing the variability were scale dependent. As the size increases, the causes of variation, such as vegetation, topographical parameters and parental material, may become increasingly complex and heterogeneous, leading to greater variability. The fractal power parameter of the function was 0.0838, nearly half the value obtained by Brocca et al. (Citation2012), who discussed the changes of spatial CVk with sizes of between 1 m2 and 250 km2. The different increases of spatial CVk with size may be attributed to the characteristics of the different regions or to the different methods used to obtain the results. The conclusions of Brocca et al. (Citation2012) were mainly derived from the information reported in previously published studies. The spatial CVk of the different sizes was determined for different times, so the final results are not likely to be very accurate due to the changes in spatial CVk over time.
We also discovered that the spatial CVk had a convex upward pattern with the field mean water content independent of the point scale or the other size scales, confirming that the relationships between the two statistical parameters were characterized by similar behaviours at different scales. The variability of the spatial CVk over time, however, differed at different scales. As seen in , the coefficient of variability of the CVk over time significantly decreased with increasing size. This relationship may be due to the dependence on scale of the factors influencing the variability of surface water content. How they change over time is also dependent on scale.
3.3 Geostatistical analysis of SWC
The coefficients of skewness and kurtosis, together with the non-parametric K-S test of the observations (), indicated that the SWCs for all dates were not normally distributed, so we logarithmically transformed the data. The transformed data for all dates passed the K-S test at the 0.05 significance level and consequently could be used for geostatistical analyses. The semivariograms of the SWCs were optimally fitted by exponential models; the structural parameters are given in . All models were satisfactory, with R2 varying from 0.572 to 0.917, suggesting that the theoretical models well represented the space structure feature of the surface SWC in the study area.
The nugget effect, C0, representing the variation at zero distance, ranged from 0.007 to 0.0648, perhaps due to an undetected sampling error, a finer-scale variability or an inherent random heterogeneity. The sill values, C0 + C, representing the total variation, ranged from 0.105 to 0.375. The differences in C0 and C0 + C over time markedly influenced the degree of soil-water heterogeneity under different water conditions. Both C0 and C0 + C had convex upward relationships with the field mean water content, (), ()). Soil water after a significant rainfall can move easily on the soil surface, so the distribution of SWC is relatively uniform over the study area. However, irregular upward evapotranspiration and downward infiltration during periods of drying become increasingly significant forces for the distribution of soil water, increasing its spatial heterogeneity. When the SWC falls below the level at which C0 or C0 + C is maximal, evapotranspiration is then water limited, and the distribution pattern is mainly controlled by the wilting coefficient. The wilting coefficient is distributed quite uniformly across the study area and thus leads to a reduction in spatial heterogeneity. Pan et al. (Citation2007) and Gao et al. (Citation2011) also demonstrated that the spatial heterogeneity of SWC peaked in moderate conditions of wetness and then decreased independent of further soil drying or wetting. The degree of spatial heterogeneity of SWC is determined by the percentage of total variation explained by the systemic variation, i.e. C/(C0 + C). This ratio varied with soil wetness, with values ranging between 0.731 and 0.933 ()), indicating that SWC was strongly to moderately spatially dependent. A strong spatial dependence is generally caused by intrinsic factors such as soil parental material and topographical parameters.
The range value, A, also changed significantly with mean SWC ()). These changes were likely due to the control of the distribution of soil moisture by different hydrological processes under different conditions of wetness. The sampling conducted immediately following a heavy rainfall event (e.g. 31 July and 15 August) presented relatively high range values (7185 and 8646 m) (). These results agree with the findings of Grayson et al. (Citation1997) and Brocca et al. (Citation2007) that SWC under wet conditions can exhibit much better distributions of spatial continuity. The high range under wet conditions may be attributed to spatially uncorrelated sampling errors and can play a relatively more significant role in these wet conditions (Brocca et al. Citation2012). As the soil dried, even with sporadic light rain during the drying period, the surface SWC tended to be much more independent and was characterized by a stochastic pattern of water content ()). For example, on 30 June, three days after the last large rainfall event (14.4 mm on 27 June), the range decreased to 4368 m. On 18 June, 13 days after the last heavy rainfall (15.8 mm on 5 June, but with light intervening rain), the range was only 2085 m (). When the soil dries enough for the wilting point to control the distribution of water content, however, the range can again increase. As displayed in , the range increased from 2874 to 9342 m between 3 September and 16 October.
3.4 Factors regulating the spatial pattern of SWC
Correlation analysis was conducted for all dates to evaluate the possible relationships between SWC and potentially associated properties (relative elevation, soil properties and vegetation). The coefficients are presented in and is the correlation matrix of these associated properties. Significant correlations were observed between some of these properties.
3.4.1 Relative elevation
Relative elevation is an important topographical property that always varies jointly with some other soil and topographical properties (e.g. the specific contributing area, clay content, etc.), which can significantly regulate the pattern of surface soil water by its effect on the lateral redistribution of soil water (e.g. infiltration and runoff). Relative elevation was negatively correlated with surface SWC in this study (), in agreement with other studies (Qiu et al. Citation2001, Hébrard et al. Citation2006) that highlighted a declining trend of SWC with increasing relative elevation. The magnitude of the negative correlation generally increased with an increase in mean water content (). According to Pan et al. (Citation2008), rain can increase the relative significance of elevation on the spatial pattern of surface SWC, because elevation becomes the dominant factor governing the redistribution of soil water immediately after a rainfall event. The effect of relative elevation, though, can quickly be replaced by strong evapotranspiration soon after a rainfall event. The degree of association between SWC and relative elevation in drier conditions when redistribution is minor or absent may thus decrease (Grayson et al. Citation1997). However, relative elevation also correlated strongly with soil and vegetation properties (), so the increasingly weaker correlations as soil dries are also likely to be the consequence of the joint contributions of soil and vegetation properties.
3.4.2 Soil properties
Soil BD, soil texture, gravel content, SOC concentration and KS have been considered as the main soil properties regulating the distribution of SWC. shows remarkable correlations between water content and most of the soil properties. These correlations confirm the importance of the spatial heterogeneity of soil properties to the variability of SWC.
In our study, BD and soil texture were the most crucial properties regulating the distribution of surface SWC. The SWC was highly inversely correlated with BD and sand content but was positively correlated with silt and clay contents for all dates (). The BD strongly reflects soil structure and the distribution of large pores and thus plays a major role in soil saturation and hydraulic conductivity. Soil texture can influence hydraulic properties and the ability of soil to retain water and can hence have a particularly significant effect on the distribution pattern of SWC. These results are supported by other studies (Zhao et al. Citation2010, Gao and Shao Citation2012), which also found strong correlations between SWC and soil texture and BD. Increased SOC content generally leads to a better-aggregated soil structure and a lower BD, and therefore encourages an enhanced capacity of soil to retain water, which can considerably reduce the loss of water (mainly by evapotranspiration and deep percolation).
Compared with the soil properties mentioned above, gravel content and KS had much weaker effects on the distribution of SWC. The relationship between KS and SWC did not even reach a significant level on some drier days (). Variations in stone content and KS in this area may thus at best have a much lower influence on the distribution of SWC. Correlations with soil texture were largely unaffected by soil wetness, which were nearly constant over the complete range of soil wetness (). This result agrees with that of Gómez-Plaza et al. (Citation2001), who found that the role of soil texture in the distribution of SWC did not vary with seasonal variations of soil wetness condition. The strong correlation with SWC thus suggests that soil texture is important in both wet and dry conditions. In contrast, the correlations between SWC and other soil properties (BD, gravel content, KS and SOC concentration) were affected by soil wetness. The correlations with BD, gravel content and KS were stronger in wet conditions, and the correlation with SOC concentration was stronger in dry conditions. These results suggest that the relative significance of these soil properties in regulating the distribution of SWC may vary with wetness. The BD can be reasonably expected to influence the distribution of SWC immediately after a significant rainfall via its relationship with soil porosity and hydraulic conductivity. Moreover, the presence of gravel on the soil surface can further promote to some degree the percolation of soil water to lower layers (Williams et al. Citation2003). As soil dries, evaporation and plant transpiration gradually become the dominant hydraulic processes, and SOC and clay content may then become more important in the distribution of SWC because of their higher capacities to retain water.
3.4.3 Vegetation properties
Vegetation regulates the pattern of SWC mainly by intercepting rain, shading the soil surface, extracting soil water for transpiration, increasing the infiltration of soil water and reducing soil temperatures. Also, plant litter and root residues accumulate on the soil surface and thereby prolong the periods of increased levels of soil water. Species composition affects interspecific competition for the limited water resources and may thus play a significant role in the evaporation and storage of SWC.
Surprisingly, we found no obvious dependence of SWC on the coverages of shrubs and herbs (low correlation coefficients, as shown in ), except on a few days. Other studies have indicated that the dynamics of SWC should be strongly influenced by variations in vegetation coverage (Reynolds Citation1970, Hawley et al. Citation1983, Bhark and Small Citation2003). The relatively weak dependence of SWC on vegetation coverage in the present study might be ascribed to the very limited amount of data collected (one time in mid-August). The correlations based on these limited data cannot represent the entire season. As shown in , SWC was significantly correlated with herb coverage on 15 August, near the date when vegetation data were collected. On the other hand, SWC was significantly and negatively correlated with species richness (). The presence of more species will increase the intensity of interspecific competition for the scarce water and will consequently accelerate the loss of soil water.
The correlation coefficients for the vegetation properties displayed a clearly linear function for field mean water content (), likely due to the differential regulation of soil-water distribution by vegetation at different stages of growth. The effect of transpiration was illustrated by the negative correlation between SWC and shrub coverage under dry conditions. In contrast, shrub coverage was positively correlated with soil water under wet conditions, perhaps caused by the interception of rainfalls. This would happen, but is not likely if just a small rainfall event occurs, because the high temperature and absence of wind would enable rapid evaporation and prevent the rain from reaching the soil surface (Bhark and Small Citation2003). The herbs in this region are mainly annuals, such as B. dasyphylla, H. arachnoideus and S. ruthenica, which prefer locations with high water contents. Herb coverage was thus positively correlated with field mean water content (). Interspecific competition also increases with higher SWC, as shown by the strong correlation between species richness and SWC ().
3.5 Stepwise linear regression analysis
The distribution of SWC was not regulated by a single factor but was governed in more complicated and comprehensive ways because many of the controlling factors were interdependent. Stepwise linear regression analysis was performed to find the factors that could best predict the dynamics of SWC. To investigate how combinations of factors changed with the seasonal variations of soil wetness, we separated the measurement campaigns into three groups based on the conditions of wetness: wet ( > 5%), medium ( = 3–5%) and dry ( < 3%).
shows that the factors regulating the distribution of SWC varied with wetness, although BD was the most dominant factor in all three conditions. Under wet conditions, relative elevation and clay content were also important factors influencing the pattern of SWC. Under dry conditions, clay content and shrub coverage became more important through the ability to retain water and to transpire, respectively. Stone content and KS also reached significant levels, but played minor roles in the variability of SWC compared to other soil properties. These results have led us to the conclusion that the major factors controlling the dynamics of SWC changed from combinations of topographical and soil properties under wet conditions to combinations of soil and vegetation properties under dry conditions.
4 CONCLUSIONS
This study surveyed the variability of surface soil (0–6 cm) water content in space and time, and identified the factors influencing the distribution of SWC in a 40-km2 area of the Gobi Desert. The spatial variability of SWC increased during drying after a wet period, peaked at specific SWCs and then decreased during further drying. The minimum amount of sampling needed coincided with a water content of approximately 4.5%. A re-sampling method further indicated that the spatial variability of SWC increased with the expansion of area and could be well parameterized as a power function of the size of the sampling area. Geostatistical analysis indicated that SWC exhibited a variable spatial dependence with time. It had a stronger spatial pattern of organization following significant rainfalls and in extremely dry conditions but had a stochastic pattern under moderate conditions of wetness. Correlation analysis implied that the dynamics of SWC in this area were linked to relative elevation, soil properties and vegetation. However, the dominant factors influencing the dynamics of SWC largely depended on the average state of SWC. Stepwise regression analysis further indicated that SWC was strongly associated with topographical and soil properties (BD, relative elevation and clay content) under wet conditions, but soil and vegetation properties (BD, clay content and shrub coverage) controlled the variability of SWC under dry conditions. This study is applicable to the description of soil-water distribution in arid regions, which may have important implications for sampling design, hydrological modelling, environmental protection and sustainability of land use.
Funding
This work was supported by grants from the Major Programme of the National Natural Science Foundation of China (no. 91025018).
REFERENCES
- Ampofo, E.A., 2006. Soil moisture dynamics in coastal savanna soils in the tropics under different soil management practices. Hydrological Sciences Journal, 51 (6), 1194–1202. doi:10.1623/hysj.51.6.1194
- Berndtsson, R. et al., 1996. Soil water and temperature patterns in an arid desert dune sand. Journal of Hydrology, 185, 221–240.doi:10.1016/0022-1694(95)02987-7
- Bhark, E.W. and Small, E.E., 2003. Association between plant canopies and the spatial patterns of infiltration in shrubland and grassland of the Chihuahuan Desert, New Mexico. Ecosystems, 6, 185–196. doi:10.1007/s10021-002-0210-9
- Brocca, L., et al., 2007. Soil moisture spatial variability in experimental areas of central Italy. Journal of Hydrology, 333, 356–373. doi:10.1016/j.jhydrol.2006.09.004
- Brocca, L., et al., 2010. Spatial-temporal variability of soil moisture and its estimation across scales. Water Resources Research, 46, W02516. doi:10.1029/2009WR008016
- Brocca, L., et al., 2012. Catchment scale soil moisture spatial-temporal variability. Journal of Hydrology, 422–423, 63–75. doi:10.1016/j.jhydrol.2011.12.039
- Charpentier, M.A. and Groffman, P.M., 1992. Soil moisture variability within remote sensing pixels. Journal of Geophysical Research, 97, 18987–18995. doi:10.1029/92JD00882
- David, M., 1977. Geostatistical ore reserve estimation. Amsterdam, The Netherlands: Elsevier.
- Famiglietti, J.S., Rudnicki, J.W., and Rodell, M., 1998. Variability in surface moisture content along a hillslope transect: Rattlesnake Hill, Texas. Journal of Hydrology, 210, 259–281. doi:10.1016/S0022-1694(98)00187-5
- Gao, X.D., et al., 2011. Soil moisture variability along transects over a well-developed gully in the Loess Plateau, China. Journal of Hydrology, 87, 357–367.
- Gao, L. and Shao, M., 2012. Temporal stability of shallow soil water content for three adjacent transects on a hillslope. Agricultural Water Management, 110, 41–54. doi:10.1016/j.agwat.2012.03.012
- Gómez-Plaza, A., et al., 2001. Factors regulating spatial distribution of soil water content in small semiarid catchments. Journal of Hydrology, 253, 211–226. doi:10.1016/S0022-1694(01)00483-8
- Grayson, R.B., et al., 1997. Preferred states in spatial soil moisture patterns: Local and nonlocal controls. Water Resources Research, 33, 2897–2908. doi:10.1029/97WR02174
- Hawley, M.E., Jackson, T.J., and McCuen, R.H., 1983. Surface soil moisture variation on small agricultural watersheds. Journal of Hydrology, 62, 179–200. doi:10.1016/0022-1694(83)90102-6
- Hébrard, O., et al., 2006. Spatio-temporal distribution of soil surface moisture in a heterogeneously farmed Mediterranean catchment. Journal of Hydrology, 329, 110–121. doi:10.1016/j.jhydrol.2006.02.012
- Hills, R.C. and Reynolds, S.G., 1969. Illustrations of soil moisture variability in selected areas and plots of different sizes. Journal of Hydrology, 8, 27–47. doi:10.1016/0022-1694(69)90029-8
- Hu, W., et al., 2008. Soil water content temporal-spatial variability of the surface layer of a Loess Plateau hillside in China. Scientia Agricola, 65, 277.
- Hu, W., et al., 2010. Watershed scale temporal stability of soil water content. Geoderma, 158, 181–198. doi:10.1016/j.geoderma.2010.04.030
- International GEWEX Project Office, 1995. Global Soil Wetness Project, 47 pp.
- Jia, Y.H. and Shao, M.A., 2013. Temporal stability of soil water storage under four types of revegetation on the northern Loess Plateau of China. Agricultural Water Management, 117, 33–42. doi:10.1016/j.agwat.2012.10.013
- Klute, A. and Dirksen, C., 1986. Hydraulic conductivity of saturated soils. In: A. Klute, ed. Methods of Soil Analysis. Madison, WI: ASA and SSSA, 694–700.
- Lawrence, J.E. and Hornberger, G.M., 2007. Soil moisture variability across climate zones. Geophysical Research Letters, 34, L20402. doi:10.1029/2007GL031382
- Li, H.D., et al., 2013. Spatio-temporal variability of soil moisture and its effect on vegetation in a desertified aeolian riparian ecotone on the Tibetan Plateau, China. Journal of Hydrology, 479, 215–225. doi:10.1016/j.jhydrol.2012.12.002
- Lu, H. and Shi, J.C., 2012. Reconstruction and analysis of temporal and spatial variations in surface soil moisture in China using remote sensing. Chinese Science Bulletin, 57, 2824–2834. doi:10.1007/s11434-012-5011-8
- Meyles, E., et al., 2003. Runoff generation in relation to soil moisture patterns in a small Dartmoor catchment, Southwest England. Hydrological Processes, 17, 251–264. doi:10.1002/hyp.1122
- Nelson, D.W., and Sommers, L.E., 1982. Total carbon, organic carbon and organic matter. In: A.L. Page, R.H. Miller, and D.R. Keeney, eds. Methods of soil analysis, Part 2, Chemical and Microbiological Properties. Madison, WI: ASA and SSSA, 534–580.
- Owe, M., Jones, E.B., and Schmugge, T.J., 1982. Soil moisture variation patterns observed in Hand County, South Dakota. Journal of the American Water Resources Association, 18, 949–954. doi:10.1111/j.1752-1688.1982.tb00100.x
- Pan, Y.X., et al., 2007. Variability characteristic of surface soil moisture content in sand areas covered by different vegetation types. Journal of Soil and Water Conservation, 21, 106–109. (in Chinese with English abstract).
- Pan, Y.X., et al., 2008. Spatial variability of surface soil moisture content in a re-vegetated desert area in Shapotou, Northern China. Journal of Arid Environments, 72, 1675–1683. doi:10.1016/j.jaridenv.2008.03.010
- Pan, Y.X. and Wang, X.P., 2009. Factors controlling the spatial variability of surface soil moisture within revegetated-stabilized desert ecosystems of the Tengger Desert, Northern China. Hydrological Processes, 23, 1591–1601. doi:10.1002/hyp.7287
- Penna, D., et al., 2009. Hillslope scale soil moisture variability in a steep alpine terrain. Journal of Hydrology, 364, 311–327. doi:10.1016/j.jhydrol.2008.11.009
- Qiu, Y., et al., 2001. Spatial variability of soil moisture content and its relation to environmental indices in a semi-arid gully catchment of the Loess Plateau, China. Journal of Arid Environments, 49, 723–750. doi:10.1006/jare.2001.0828
- Reynolds, S.G., 1970. The gravimetric method of soil moisture determination Part III An examination of factors influencing soil moisture variability. Journal of Hydrology, 11, 288–300. doi:10.1016/0022-1694(70)90068-5
- Takagi, K. and Lin, H.S., 2012. Changing controls of soil moisture spatial organization in the Shale Hills Catchment. Geoderma, 173–174, 289–302. doi:10.1016/j.geoderma.2011.11.003
- Vereecken, H., et al., 2007. Explaining soil moisture variability as a function of mean soil moisture: A stochastic unsaturated flow perspective. Geophysical Research Letters, 34, L22402. doi:10.1029/2007GL031813
- Wang, Y.Q., et al., 2012. Regional spatial pattern of deep soil water content and its influencing factors. Hydrological Sciences Journal, 57 (2), 265–281. doi:10.1080/02626667.2011.644243
- Western, A.W., et al., 1999. Observed spatial organization of soil moisture and its relation to terrain indices. Water Resources Research, 35, 797–810. doi:10.1029/1998WR900065
- Western, A.W., Blöschl, G., and Grayson, R.B., 1998. Geostatistical characterisation of soil moisture patterns in the Tarrawarra catchment. Journal of Hydrology, 205, 20–37. doi:10.1016/S0022-1694(97)00142-X
- Williams, A.G., et al., 2003. Soil moisture variability and land use in a seasonally arid environment. Hydrological Processes, 17, 225–235. doi:10.1002/hyp.1120
- Yu, S.W., et al., 2012. Determination of water sources of Gobi plants by D and 18O state isotopes in middle reaches of the Heihe River. Journal of Desert Research, 32, 717–723. (in Chinese).
- Zhang, J.G., et al., 2011. Spatial variability and patterns of surface soil moisture in a field plot of karst area in southwest China. Plant, Soil and Environment, 57, 409–417.
- Zhao, Y., et al., 2010. Controls of surface soil moisture spatial patterns and their temporal stability in a semi-arid steppe. Hydrological Processes, 24, 2507–2519.
- Zhao, Y., et al., 2011. Factors controlling the spatial patterns of soil moisture in a grazed semi-arid steppe investigated by multivariate geostatistics. Ecohydrology, 4, 36–48. doi:10.1002/eco.121
- Zhu, Y.J. and Shao, M.A., 2008. Variability and pattern of surface moisture on a small-scale hillslope in Liudaogou catchment on the northern Loess Plateau of China. Geoderma, 147, 185–191. doi:10.1016/j.geoderma.2008.08.012