Monitoring the effects of land use and cover type changes on soil moisture using remote-sensing data: A case study in China's Yongding River basin
Introduction
Soil moisture is an important factor that determines energy exchanges between the land and the atmosphere, and is also an important factor monitored drought by means of remote sensing. The spatial and temporal distribution of soil moisture and changes in this distribution strongly influence the surface heat balance, evapotranspiration, and soil temperature. Soil moisture is also a key factor that determines the spatial and temporal dynamics of terrestrial ecosystems and is therefore an important parameter in ecological models (Carlson et al., 1994, Zhan et al., 2006, Mao et al., 2007). The soil moisture content also determines the water available for evaporation from land surfaces and thereby influences the demand for water by crops (i.e., the need for supplemental irrigation), thereby controlling crop growth and yield. Reliable soil moisture data are thus an important tool for predicting crop yield (Foody, 1991, Chen et al., 2007.
Acquisition of data on land surface energy exchanges is an important part of monitoring changes in regional resources and environments, and because soil moisture plays such an important role in these exchanges, large-scale monitoring of soil moisture levels can play an important role in agricultural research and environmental evaluations. Retrieval of soil moisture values at regional or global scales is therefore an important tool in studies of land surface processes (Moran et al., 1994, Liu et al., 2003, Xiao & Sha, 2007).
Unfortunately, monitoring soil moisture values over large areas and long time periods by means of field data is time-consuming and expensive. For this reason, the use of remote sensing to retrieve soil moisture data is an important research goal, despite the difficulty of obtaining reliable estimates from these measurements. Because soil moisture data is so important for the management of agriculture and water resources in China, which is currently suffering from severe water shortages in many areas, its estimation is a high-priority area of research (Liu et al., 2003, Xiao & Sha, 2007). Because changes in land use and cover type are important anthropogenic factors that influence the spatial distribution of soil moisture, understanding the spatial and temporal relationships between these changes and soil moisture will provide important data that is required to support the efficient use of the available soil moisture and sustainable use of China's land resource.
The radiation or reflection characteristics of a surface differ at different wavelengths. Because of the absorption and reflection of light by chlorophyll in leaves and the subsequent reflection and dispersion of the absorbed light within the internal structure of the leaves, vegetation has unique spectral characteristics: it absorbs visible light between 380 and 500 nm, 600 and 760 nm, and reflects green light. In the near-infrared wavelengths, the spectrum mainly features reflection (Pratt & Ellyett, 1979, Liu et al., 2002).
Similarly, soil moisture influences the spectral characteristics of soil, which also depend on the solid components of the soil. Soil reflectance decreases with increasing soil moisture (Moran et al., 1994, Zhan et al., 2006). The soil moisture status also influences vegetation growth, and thereby changes the spectral characteristics of the vegetation. On this basis, a vegetation index can be developed that indicates the growth status of vegetation based on its spectral characteristics, and different vegetation indices can be calculated based on remote-sensing data to indicate the growth and level of moisture stress of vegetation, thereby allowing indirect estimates of soil moisture content even when the soil surface is not visible. Examples include the crop water stress index (Jackson & Pinter, 1981, Liu et al., 2003) and the vegetation anomaly index (Xiao et al., 1994).
In the thermal infrared wavelengths, surface objects differ in their temperature and emissivity. In addition to determining heat exchange with the atmosphere, the temperature of vegetation and soils is mainly determined by the thermal inertia of these materials (i.e., their ability to absorb light energy without changing temperature), and their reflectance and emission of different wavelengths of light. Therefore, soil moisture can be estimated by the relationship between canopy and soil surface temperatures and soil moisture based on a combination of a soil moisture-balance equation and energy-balance theory. The thermal inertia method is in common use (e.g., Xue & Cracknell, 1995, Cheng et al., 2006). But this kind of estimation method seems difficult to apply to large-scale soil moisture monitoring. For areas covered by vegetation, land surface temperature monitored by remote sensing actually represents vegetation canopy temperature. When meeting water shortage, vegetation would close parts of stomata to reduce transpiration in order to avoid excessive water lose, which causes increase of canopy temperature. As a result, vegetation canopy temperature could be regarded as an indicator to characterize vegetation water status. Land surface temperature could reflect soil moisture of bare soil surface and soil in depth of 2 cm, as well as soil moisture in roots (Richardson & Wiegand, 1977, Gitelson, 2004). On the other side, remote-sensed normalized difference vegetation index (NDVI) is a common index to reveal vegetation growth. Water shortage would affect chlorophyll content, especially when wilting occurs, which causes obviously decrease of leaf area index (Goetz, 1997). There are limitations when simply using only vegetation index or canopy temperature to monitor soil moisture. Introducing the combination of vegetation index (NDVI, for example) and land surface temperature to establish an integrated model to monitor soil moisture is applicable to large-scale soil moisture monitoring. And the most commonly used one is TVDI, which is showed in a two-dimension space composed of NDVI and land surface temperature (Sandholt et al., 2002).
In studying the influence of land use or cover type changes on soil moisture, most researchers have only used field-measured point data to study soil moisture contents before and after the change, and have seldom discussed the spatial and temporal patterns (Carlson et al., 1994, Moran et al., 1994, Cheng et al., 2006). This is caused by the difficulty and expense of obtaining many samples over a large area or long period of time. As a result, most research is based on a limited number of observations obtained over a limited time period, and thus cannot adequately reflect the spatial and temporal diversity of the changes. However, remote sensing can solve this problem by providing coverage of large areas and long time periods for a reasonable cost, and if a reliable model of the relationship between spectral data and soil moisture content can be developed, remote sensing can allow effective monitoring of the effects of land use and cover type change on soil moisture.
In this paper, we retrieve land surface temperature in July of Yongding River basin in 1987, 2000 and 2005 using remote-sensing images. TVDI is calculated according to Ts/TVDI feature space. Then soil moisture at depths of 0 to 10 cm and 10 to 20 cm are estimated based on the linear relationship between soil moisture and TVDI. Further, effects of land use/cover change on soil moisture are studied on the basis of land use/cover classification. The structuring of the paper is as follows:
- 1
Introduction
- 2
Area descriptions, data and methods
- 2.1
Study area
- 2.2
Data
- 2.3
Soil moisture estimation model
- 2.1
- 3
Soil moisture estimation
- 3.1
Land surface temperature
- 3.2
TVDI
- 3.3
Spatial distribution of soil moisture
- 3.1
- 4
Influence of land use and cover type changes on soil moisture
- 4.1
Soil moisture levels under different land use and cover types
- 4.2
Influence of land use and cover type changes on soil moisture
- 4.1
- 5
Discussions and conclusions.
Section snippets
Study area
The Yongding River basin is located between 111°35′E and 117°12′E and between 41°30′N and 38°35′N (Fig. 1). Total area of the basin is 44,600 km2. The river flows across Shanxi province and Inner Mongolia, and through Hebei, Beijing, and Tianjin provinces. The upper reaches of the Yongding River include the cities of Datong and Shuozhou in Shanxi province and Zhangjiakou in Hebei province. Datong and Shuozhou experience a temperate continental monsoon climate. This area is influenced by the
Land surface temperature
To calibrate the estimated Ts values, we measured surface temperatures under different vegetation types using a geothermometer. In July 2005, we obtained these measurements simultaneously with the remote-sensing data, from 10:00 to 11:00 each day, at intervals of 5 min. When measured in the field, the probe of a geothermometer is inserted vertically into soil about 10 cm. Data were recorded when soil temperature data shown on the monitors keep stable. For one plot three geothermometers are used
Soil moisture levels under different land use and cover types
Table 6 summarizes the soil moisture contents to a depth of 10 cm (mean ± standard deviation) under different land use and cover types during the study period. The surface soil moisture content was highest in July 2000 for all types, and was lowest in 1987 for all types except forest. Forests had the highest soil moisture content in all years, indicating a lower level of evaporation from the soil and generally good soil moisture status in the forests. Grasslands had higher soil moisture than
Discussions and conclusions
In this study, we determined land surface temperatures (Ts) in the Yongding River basin in July of 1987, 2000, and 2005 using thermal infrared data from Landsat TM and ETM+ images. Verification of the estimated temperatures using field-measured temperatures indicated that the results were sufficiently reliable to use in subsequent analyses.
We then determined the relationships between NDVI and Ts, and used the results to calculate TVDI values based on the maximum and minimum temperatures and the
Acknowledgments
This work was supported by the National High Technology Research and Development Program (“863”) of China (No: 2006AA120108), the National Natural Science Foundation of China (No.30970513), and the Open Project Program of State Key Laboratory of Earth Surface Processes and Resource Ecology of China (No: 2009-KF-15).
References (28)
- et al.
Remote estimation of soil moisture availability and fractional vegetation cover for agricultural fields
Agricultural and Forest Meteorology
(1990) Wide dynamic range vegetation index for remote quantification of biophysical characteristics of vegetation
Journal of Plant Physiology
(2004)- et al.
Estimating crop water deficit using the relation between surface-air temperature and spectral vegetation index
Remote Sensing of Environment
(1994) - et al.
A simple interpretation of the surface temperature/vegetation index space for assessment of surface moisture status
Remote Sensing of Environment
(2002) - et al.
Monitoring the effects of land use and cover changes on net primary production: a case study in China's Yongding River basin
Forest Ecology and Management
(2009) - et al.
A method to make use of thermal infrared temperature and NDVI measurements to infer surface soil water content and fractional vegetation cover
Remote Sensing Reviews
(1994) Image-based atmospheric corrections—revisited and improved
Photogrammetric Engineering and Remote Sensing
(1996)- et al.
The study of estimating soil moisture using ERS wind scatterometer
Journal of Remote Sensing in China
(2007) - et al.
The soil moisture detection for different vegetation coverage based on the MODIS data
Journal of Remote Sensing in China
(2006) - et al.
Comparison of impervious surface area and normalized difference vegetation index as indicators of surface urban heat island effects in Landsat imagery
Remote Sensing of Environment
(2007)
Soil moisture content ground data for remote sensing investigation of agriculture region
International Journal of Remote Sensing
Sources of variation in radiometric surface temperature over a tallgrass prairie
Remote Sensing of Environment
A verification of the ‘triangle’ method for obtaining surface soil water content and energy fluxes from remote measurements of the normalized difference vegetation index (NDVI) and surface radiant temperature[J]
International Journal of Remote Sensing
Multi-sensor analysis of NDVI, surface temperature and biophysical variable at a mixed grassland site
International Journal of Remote Sensing
Cited by (51)
Land use and cover change (LUCC) impacts on Earth's eco-environments: Research progress and prospects
2023, Advances in Space ResearchCitation Excerpt :The ground observation and satellite-based methods mainly include correlation analysis and experimental watershed approaches. The former described the degree of change in LUCC and hydrological factors; however, they did not reveal the degree of influence that was exerted by climate change and LUCC (Wang et al., 2010). The experimental watershed approach selects two or more areas with similar eco-environmental factors and analyses the variability in the environment under different land cover types, which provides a practical method for evaluating LUCC impacts (Venkatesh et al., 2011, Wang et al., 2013).
Drought monitoring of the maize planting areas in Northeast and North China Plain
2021, Agricultural Water ManagementSoil moisture mapping for different land-use patterns of lower Bhavani river basin using vegetative index and land surface temperature
2024, Environment, Development and SustainabilityIrrigation Scheduling for Small-Scale Crops Based on Crop Water Content Patterns Derived from UAV Multispectral Imagery
2023, Sustainability (Switzerland)