Ecohydrology in water-limited environment using quantitative remote sensing - the Heihe River basin (China) case

Water-limited environments exist on all continents of the globe and they cover more than
30% of the Earth’s land surface. The eco-environments of these regions tend to be fragile and
they are changing in a dramatic way through processes like land desertification, shrinking of
oases, groundwater depletion, and soil erosion. These are either human induced or results of a
changing climate. Implications of these changes for both the regional hydrologic cycle and the
vegetation have been documented. Since these changes occur over a wide range of scales in
space and time, remote sensing methods are needed to monitor the land surface characteristics,
to observe changes in vegetation and hydrological states, and to compare these with
predictions from hydrological models. It is widely accepted that remote sensing methods offer
the ability to acquire spatially continuous measurements over large areas. Remote sensing can
also help to visualize complex processes because the spatial data can be captured regularly
over time.
China is one of several countries with large arid and semi-arid areas. The Heihe River basin,
situated in the arid inland of northwestern China, is one of the areas severely affected by ecoenvironmental
degradation and recovery. The problem of the degraded environment is due to
overexploitation of surface and ground water leading to shrinking of oases, including the
decline and death of natural vegetation, and the lowering of the groundwater table. Exhaustive
(over-)use of water resources is the main cause of land degradation in the lower reaches of the
basin, called the Ejina oasis. The whole Heihe River basin is therefore selected as study area
in this thesis to analyze the long-term eco-environmental changes. What happens in this river
basin is likely to have a growing influence on regional hydrological cycles, even affecting
human life. Effective management of eco-environmental problems in this critical zone of
water-limited conditions will provide scientific evidence for protecting and improving the
eco-environment in these Chinese northwestern arid regions, eventually resulting in land
improvement.
Studies on quantifying the relationship between the vegetation and the water resources are a
critical step in developing an ecohydrological approach to resources management in order to
minimize environmental degradation. Remote sensing measurements can help us to better
understand the effects of changes in water management on hydrological processes and their
subsequent feedback to the eco-environment at the regional scale. Remote sensing methods
can also provide information to quantify heterogeneity and change at a large scale. Therefore,
the main objective of this thesis is to develop a methodology for the quantitative assessment
of eco-environmental changes at a large scale in arid regions by integrating remote sensing
methods in ecohydrological approaches.
Chapter 1 outlines the significance of quantitative assessment of eco-environmental
changes using remote sensing methods and applying them for ecohydrology in northwestern
China, resulting in the specific research objectives of this thesis.
Chapter 2 quantifies both the vertical and horizontal distribution of vegetation in the Qilian
Mountains area, representing the upper reaches of the Heihe River basin, based on MODIS
NDVI images from the year 2000 - 2006. Our analysis reveals that elevation and aspect are
two important impact factors for the vertical distribution of vegetation in a mountainous area.
The NDVI increases with the elevation and reaches a maximum value at a certain elevation
threshold, and then decreases as the elevation increases beyond this threshold. The optimal
vegetation growth is on the shady side of the mountains because of less evapotranspiration.
The best combination of temperature and precipitation is assessed providing good conditions
for vegetation growth.
Chapter 3 presents an efficient method to estimate the regional annual evapotranspiration
(ET) based on the SEBS algorithm (Surface Energy Balance System) in the Zhangye basin,
representing the middle reaches of the Heihe River basin. The method proposed is a
combination of the daily SEBS results and data collected by meteorological stations. The
result shows that the annual ET increased gradually during the period 1990-2004 and the main
impact factor on the long-term increase of annual ET was the vegetation change. The
accuracy of the ET result is validated using a water balance for the whole watershed and the
validation reveals that the SEBS algorithm can be used to effectively estimate annual ET in
the Zhangye basin.
Chapter 4 establishes the quantitative relationship between the runoff of the Heihe River
and the long-term vegetation change of the Ejina oasis, located in the lower reaches of the
Heihe River. In this part, two time periods are distinguished corresponding to before and after
the implementation of a new water allocation scheme in the Heihe River basin. The GIMMS
NDVI and MODIS NDVI data sets are used to quantify the long-term change of the oasis
vegetation in the first period 1989-2002 and the second period 2000-2006, respectively. The
vegetation change shows a decreasing trend from 1989 to 2002 and an increasing trend
between 2000 and 2006. Good relation between the runoff of the river and the vegetation
growth are found at both stages and the time lag of the observed hysteresis effect of the runoff
of the river on the oasis vegetation is one year. In addition, the yearly smallest water amount
which sustains the demand of the eco-environment of the Ejina area is estimated to be 4×108
m3 based on MODIS images.
Chapter 5 explores a method to quantify the effect of the groundwater depth on the
vegetation growth in the year 2000 in the oasis area by combining MODIS NDVI with
groundwater observation data. The result demonstrates that the groundwater depth suitable for
vegetation growth in this region ranges from 2.8 to 5 m, depending on species composition.
Hardly any vegetation growth occurs when the groundwater depth is below 5 m because the
rooting depth of the occurring species is limited and cannot maintain adequate water supplies
to their canopies when the water depth is below 5 m. The situation changes after
implementation of the new water allocation scheme since 2000. The mean NDVI increased
and the annual conversion of bare land into vegetated land is about 38 km2 per year during the
period 2000 – 2008. It reflects a potential recovery of the eco-environment of the Ejina area.
Chapter 6 comprises the main conclusions and the outlook for possible improvements in
future research. The main contribution of this study is the successful integration of remote
sensing with ecohydrology in quantifying the relationship between water resources and
vegetation occurrence at large scale. It provides a methodology to evaluate the long-term
vegetation change and the water resources impact using remote sensing data in water-limited
areas. The approach of vegetation dynamics, runoff and groundwater impacts presented in this
thesis serves as a sound foundation for predicting the effects of future environmental changes.