The transitional belt with both water and wind erosion is the center of the intensive
erosion and subject to severe vegetation degradation. The wind-water erosion region is
characterized by dramatic change of climate and quite heterogeneous soil substrate,
suffering wind and/or water erosion through the year. Such unique and harsh habitat
quality indicates that the performance of vegetation restoration in eco-environmental serve,
e.g. remedy soil and water erosion, improve soil fertility and ecosystem carbon
sequestration, in the wind-water erosion region should be different from the other parts of
the Loess plateau. In 2007-2009 years, experiments were conducted in Shenmu Erosion
and Environment Research Station of the Institute of Soil and Water Conservation,
CAS&WRR to study the dynamic temporal and spatial distribution of four biogenic
elements (water, carbon, nitrogen and phosphorus), vegetation evapotranspiration, soil
respiration and plant photosynthesis rate, and their responses to four vegetation types
(shrub, grass, fallow and cropland). The main results are:
1. Vegetation types changed the soil moisture distribution and water balance on the
hillslopes. The soil moisture distribution was uniform along the hillslopes for shrub and
grass. However, soil water storage to 4 m depth for fallow and cropland showed a water
lateral movement driven by the topography. The evapotranspiration order of four
vegetation types was shrub > grass > cropland > fallow. At the beginning of the rain season
(June), the evapotranspiration from shrub and grass were quite larger than the precipitation.
The slope scale in length mainly influenced the runoff.
2. The performances in soil quality recovery were quite different among vegetation
types. Compared with the cropland, shrub directly increased soil organic carbon and total nitrogen in the 0-100 cm soil depth, however, grass and fallow could indirectly improve
soil organic carbon and nitrogen by decreasing sediments losses and runoff. The effect of
vegetation types was not significant on soil total phosphorus concentration but pronounced
on soil total phosphorus distribution along the hillslopes. A downward accumulation trend
of soil total phosphorus along the hillslope was observed for fallow and cropland.
Pronounced influences of vegetation types on the temporal and spatial variability of soil
available phosphorus and inorganic nitrogen were detected. Soil available phosphorus of
shrub, grass and fallow was consistently lower than that of cropland. However, the order of
soil inorganic nitrogen between shrub, grass, fallow and cropland varied with season. In
contrast to cropland, soil available phosphorus was higher in August and had no downward
accumulation along the hillslopes for shrub, grass and fallow. During the study period, the
dominant form of soil inorganic nitrogen for shrub and grass changed frequently, however,
the dominant form was NH4
+－N and NO3
－－N for fallow and cropland, respectively. Soil
inorganic nitrogen distributed randomly along the hillslopes for shrub and fallow but had
downward accumulation trend in certain months for grass and cropland. Besides, fertilizer
derived nitrogen leaching was at risk in this semiarid rainfed agriculture ecosystem.
3. Vegetation phonological stage mediated the effects of vegetation types on daytime
and monthly soil respiration. The soil respiration variation amplitude of fallow and
cropland was comparable to that of shrub and grass in July but obviously larger than that of
shrub and grass in August and October. During the study period, monthly soil respiration of
fallow and cropland were comparable and significantly lower than that of shrub and grass,
with an exception in August when soil respiration of cropland was as larger as shrub and
grass. The main controlling factor of monthly soil respiration variation differed among
vegetation types, e.g. soil temperature highly mediated the monthly soil respiration
variation for the three vegetation types but contributed little when went to the cropland.
The biotic factors, root biomass, leaf area index and photosynthesis rate, are mainly
responsible for the monthly differences in soil respiration among vegetation types.
Moreover, the dominant controlling factor varied with vegetation stage.
4. Rock fragments changed the rain pulse characteristics and the level of soil
respiration during the rain season. Rock fragment layer delayed the rain pulse peak of soil
respiration, and the delay duration increased with the increase of rainfall. Average soil respiration during the study period was significantly lower than the control because of the
lower root biomass and different rain pulse properties. Rock fragment coverage showed a
significant effect on soil respiration following rainfall but had no influence on rain pulse
rhythm of soil respiration. Soil respiration of the treatment mixed with rock fragment was
quite lower than that of control. Effect of rock fragment coverage and content on average
soil respiration of the study period was not significant. Additionally, rock fragment did not
change the seasonal variation of soil respiration.
5. Compared with fallow, shrub and grass had better performance in carbon
sequestration. Vegetation restoration improved the carbon sequestration and photosynthetic
C uptake of the vegetation in a small watershed with the area of 20 hm2. Vegetation
restoration in this small watershed annually increased photosynthetic carbon uptake and
carbon sequestration with the values of 25 and 2.2 Mg C, respectively.
The results showed that shrub performed better than grass in improving soil nutrient
content. The large carbon emission caused by soil respiration of shrub and grass was offset
by the corresponding large photosynthetic carbon uptake. Therefore, the small watershed
involved with vegetation restoration program served as carbon sink. Soil water deficit at
the beginning of the rain season aggravated the soil desiccation of planting perennial
vegetations. However, this problem would be alleviated by adopting instructive spatial
distribution pattern of vegetation types. Because urea-derived nitrogen loss by leaching
was at risk in the agroecosystem, optimized fertilizer management should be adopted
instead of the traditional extensive style. The result that rock fragment significantly
changed the rain pulse properties and rain season mean value of soil respiration suggests
that considering soil heterogeneous is important in terrestrial ecosystem carbon emission.
The conclusions of this study can be used to evaluate the eco-environmental performance
of the vegetation restoration program in the wind-water erosion region and be beneficial in
constructing effective management of eco-environment restoration.