Plants Physiological Processes And Estimation And Spatial-temporal Variation Of Evapotranspiration Over Loess Plateau

Loess Plateau is a region with unique geographical and ecological environment. Land surface physiological processes and water transfer processes in this region are important scientific problems that are urgent to be solved. The physiological features of vegetation in response to micro- meteorological conditions and furrow-ridge and film- mulching treatments were analyzed. The performance of eight evapotranspiration model which are commonly used worldwide was evaluated in the Loess Plateau. Based on this, a new approach was developed for distributed estimation of evapotranspiration over Loess Plateau. Using this method, evapotranspiration over Loess Plateau in recent 30 years was calculated, and the spatial-temporal variation of evapotranspiration over Loess Plateau and its influencing factors were analyzed. Besides, an observational analyze was made of the role of vegetation in controlling evapotranspiration. The main conclusions are summarized as follows:(1) A field experiment was conducted to analyze dynamic changes of photosynthetic characteristics of spring wheat flag leaf at heading and filling stage, and the responses of photosynthetic characteristics to air temperature and humidity, as well as their response threshold. At the same time, photosynthesis stomatal and non-stomatal limitations were discussed under natural conditions. It is found that net photosynthetic rate was negatively related to air temperature at heading state, while net photosynthetic rate was quadratic correlated to air temperature at grain filling stage, with the threshold temperature of 23℃. Transpiration rate was was quadratic correlated to air temperature, with threshold temperature being 21.7℃ and 23.4℃ for heading and filling stage, respectively. Stomatal conductance was also quadratic correlated to air temperature, with threshold temperature being 21.0℃ and 24.0℃ for heading and filling stage, respectively. Net photosynthesis rate, transpiration rate, and stomatal conductance were positively correlated to air humidity at both stages. Net photosynthesis rate were highly correlated to stomatal conductance at both stages. Stomatal limitation of photosynthesis was apparent at heading and filling stages. Because of the decreasing sensitivity of stomata to vapor pressure deficit and " midday depression" strategy, stomatal limitation is decreasing from heading stage to filling stage, which contributes to higher photosynthetic efficiency in flag leaves of spring wheat in semi-arid areas and thus guarantees good harvest.(2) Experiment was conducted to explore the responses of potato leaf and soil water potential to furrow-ridge and film- mulching treatments. The results showed that treatments of furrow and ridges with film mulched had a significant impact on leaf and soil water potential. In different soil layers, diurnal variation pattern of soil water potential was different at different growth stages. In 0-20 cm soil layer, the diurnal variation pattern of soil potential of the treatment of soil ridge(NM) at the stage of anthesis, the diurnal variation pattern of soil potential of the treatment of soil ridge with plastic-covered(M) at tuber expansion stage, and the diurnal variation pattern of soil potential of the treatments of plastic-covered ridge(M) and double-ridge with film covered(DR) at maturity were going down at first, and then going up at late afternoon, while the diurnal variation of soil potential of other treatments were going down gradually. In 20-40 cm soil layer, the diurnal variations patterns of soil potential of all treatments were going down gradually. Diurnal variation patterns of leaf water potential had two different styles: double trough type and single trough type. All treatments had a same pattern, but the magnitudes of leaf potential were much different. At the stages of anthesis and tuber expansion, diurnal variation patterns of leaf water potential were double trough type and the troughs occurred at about 13:00 and 17:00. At maturity stage, diurnal variation pattern of leaf potential was single trough type and the trough occurred at about 17:00. With the progress of growth, soil and leaf water potential firstly decreased and then increased. The 20-40 cm soil layer was the most important soil layer to leaf water potential. In this soil layer, NM has kept soil under the soil ridge having the best moisture status, resulting in leaf water potential of such a treatment being the highest. For the treatment of M and DR, large transpiration rate accelerated water transfer rate in the soil-plant-atmosphere continuum, causing a relatively low leaf and soil water potential.(3) The performance of eight commonly used ET models was evaluated in Loess Plateau. The order of overall performance from best to worst is as follows: the revised Priestley-Taylor model(PT-JPL), the modified PT-JPL model(M1-PT-JPL), the Community Land Model(CLM), the modified PT-JPL model(M2-PT-JPL), the revised Penman-Monteith model(RS-PM), an empirical model(Wang), the advection-aridity model(AA) and the energy balance model(SEBS). The performance of all of the models is comparably poor in winter and summer, except for the PT-JPL model, and relatively good in spring and autumn. Due to the vegetation control on ET, the Wang, RS-PM, PT-JPL, M1-PT-JPL and M2-PT-JPL models perform better for cropland, whereas the AA model, SEBS model, and CLM perform better for grassland. The PT-JPL and Wang models perform better in semi-arid region than in semi- humid region, whereas the opposite is true for CLM, SEBS and RS-PM. The AA, M1-PT-JPL, M2-PT-JPL models perform similarly in semi-arid and semi- humid regions. When considering the inter-annual variability in precipitation, the Wang model has relatively good performance under only some annual precipitation conditions; the performance of the PT-JPL and AA models is reduced under conditions of high precipitation; the two modified PT-JPL models inherited the steady performance of the PT-JPL model and improved the performance under conditions of high annual precipitation by the modification of the soil moisture constraint. RS-PM is more appropriate for humid conditions. CLM and PT-JPL models could be effectively applied to all precipitation conditions because of their good performance across a wide annual precipitation range.(4) Since C LM has proven to be suitable to simulate ET in Loess Plateau, we further compared ET simulated by C LM and observations and found that, CLM had much uncertainty in the semi-arid region of Loess Plateau and the errors were positively correlated with NDVI. Therefore, a new method for ET estima tion over the Loess Plateau is proposed combining remotely sensed NDVI and CLM simulated ET. The ET in the semi-arid areas estimated by the new method had a better correlation with observations than the simulated ET by the CLM, with correlation coefficient increased by 0.07 at SACOL. The root mean square error is decreased by 6.20 mm, and the relative error is reduced by 13.9% at SACOL site.(5) Using ET observed by eddy covariance system, other meteorological data, and physiological and ecological data, the inter-annual variability of ET and its control factors were analyzed in the semi-arid and sub-humid Loess Plateau. ET showed large inter-annual variability in the Loess Plateau, with an averaged variation rate being 19.5% for the four selected sites. Environmental variables, including reference ET(ET0), precipitation(P), soil volumetric water content(VWC), Normalized Difference Vegetation Index(N DVI), and surface conductance(gs) also showed notable inter-annual variability. P was the original forcing causing the inter-annual variability of ET. NDVI and gs played an important role in regulating ET in the semi-arid regions, which reduced the direct correlation between ET and VWC. In the semi-arid region, different ecosystems differed in biological processes regulating ET under water stress condition: for grassland, gs responded to soil volumetric water content and surface air humidity directly and controlled ET via regulating transpiration; for cropland, N DVI responded to VWC and e directly and determined gs, and then regulated ET. This suggests that using NDVI to characterize physiological constraints of ET is feasible for semi-arid cropland but not suitable for semi-arid native grassland when designing evapotranspiration estimation algorithms. Parameterization of gs using air humidity could help better estimation of ET in the semi-arid grassland. In the semi- humid areas, although soil volumetric water content altered NDVI, the ecological and physiological controls on ET were much weaker.(6) ET was estimated using the newly established method over the Loess Plateau from 1982-2013, and an analysis was made on the temporal-spatial variation of ET over the Loess Plateau. ET decreases gradually from southeast to northwest over the Loess Plateau, with a range from 50 mm to 500 mm and a mean value of 299.5 mm. High ET areas are located at the southern Gansu province and the intersection of the Yellow River, Wei River and Luo River. The domain-average ET was decreasing in recent 31 years, with the decrease rate being-1.65 mm year-1, indicating that the regional water cycle is becoming weaker. ET showed a slightly decrease in the 80 s and 90 s and a much larger decrease in the 00 s of the 21 th century. ET fluctuated much more strongly in the 90 s than 80 s and 00 s, which can cause disturbance in the water cycle and drought. Expect in autumn, ET decreased in all seasons. Particularly, ET in summer comprised about half of the annual ET and had a sharpest trend, which dominated the declining inter-annual trend. ET had a slight decrease but large fluctuated strongly in spring. ET in winter was very small, so was the decline trend. ET decreased more sharply in semi-arid region than semi- humid region. The declining trend of ET was attributed to declining precipitation and air humidity over the Loess Plateau during the study period. The climate tendency rate of local ET was highly related to local mean annual precipitation: in areas with precipitation less than 400 mm, climate tendency rate was negative; in areas with precipitat ion ranging from 400 to 600 mm, negative and positive rate existed; in areas with precipitation larger than 600 mm, climate tendency rate is positive. This implies that there existed two feedbacks in the Loess Plateau: ET was decreasing in areas with less annual precipitation and increasing in areas with large annual precipitation, leading to arid areas becoming more arid and humid areas becoming more humid.(7) Using C LM and PT-JPL, ET was estimated over global typical semi-arid regions from 1982 to 2011. Global typical semi-arid regions all showed a decline ET trend in recent 30 years. Among them, North Africa had the fastest decline rate up to-8.6 mm year-1, while South Africa had the slowest decline rate of about-0.7 mm year-1. Among the semi-arid regions with the same latitude, Loess Plateau had the maximum decrease rate of about-3.7 mm year-1. North America, South America, North Africa, and Australia showed a clear decrease trend in both winter-half and summer- half year, while Loess Plateau, East As ia, Central Asia and South African only showed decrease trend in summer-half year. Among the three basic factors controlling ET, atmospheric demand was increasing in all semi-arid regions, while radiation and rainfall was declining in all semi-arid regions in recent 30 years. By conducting factorial experiment, RH was found to be the dominant factor determining the decrease in ET, and its contribution to the decline of ET surpassed sum of all other factors.