The cause-effect associations between geographical phenomena are an important focus in ecological research. Recent studies in structural equation modeling (SEM) demonstrated the potential for analyzing such associations. We applied the variance-based partial least squares SEM (PLS-SEM) and geographically-weighted regression (GWR) modeling to assess the human-climate impact on grassland productivity represented by above-ground biomass (AGB). The human and climate factors and their interaction were taken to explain the AGB variance by a PLS-SEM developed for the grassland ecosystem in Inner Mongolia, China. Results indicated that 65.5% of the AGB variance could be explained by the human and climate factors and their interaction. The case study showed that the human and climate factors imposed a significant and negative impact on the AGB and that their interaction alleviated to some extent the threat from the intensified human-climate pressure. The alleviation may be attributable to vegetation adaptation to high human-climate stresses, to human adaptation to climate conditions or/and to recent vegetation restoration programs in the highly degraded areas. Furthermore, the AGB response to the human and climate factors modeled by GWR exhibited significant spatial variations. This study demonstrated that the combination of PLS-SEM and GWR model is feasible to investigate the cause-effect relation in socio-ecological systems.

Monitoring of rangeland forage production at specified spatial and temporal scales is necessary for grazing management and also for implementation of rehabilitation projects in rangelands. This study focused on the capability of a generalized regression neural network (GRNN) model combined with GIS techniques to explore the impact of climate change on rangeland forage production. Specifically, a dataset of 115 monitored records of forage production were collected from 16 rangeland sites during the period 1998-2007 in Isfahan Province, Central Iran. Neural network models were designed using the monitored forage production values and available environmental data (including climate and topography data), and the performance of each network model was assessed using the mean estimation error (MEE), model efficiency factor (MEF), and correlation coefficient (r). The best neural network model was then selected and further applied to predict the forage production of rangelands in the future (in 2030 and 2080) under A1B climate change scenario using Hadley Centre coupled model. The present and future forage production maps were also produced. Rangeland forage production exhibited strong correlations with environmental factors, such as slope, elevation, aspect and annual temperature. The present forage production in the study area varied from 25.6 to 574.1 kg/hm². Under climate change scenario, the annual temperature was predicted to increase and the annual precipitation was predicted to decrease. The prediction maps of forage production in the future indicated that the area with low level of forage production (0-100 kg/hm²) will increase while the areas with moderate, moderately high and high levels of forage production (≥100 kg/hm²) will decrease both in 2030 and in 2080, which may be attributable to the increasing annual temperature and decreasing annual precipitation. It was predicted that forage production of rangelands will decrease in the next couple of decades, especially in the western and southern parts of Isfahan Province. These changes are more pronounced in elevations between 2200 and 2900 m. Therefore, rangeland managers have to cope with these changes by holistic management approaches through mitigation and human adaptations.

Water and nitrogen (N) are generally two of the most important factors in determining the crop productivity. Proper water and N managements are prerequisites for agriculture sustainable development in arid areas. Field experiments were conducted to study the responses of water productivity for crop yield (WP_Y-ET) and final biomass (WP_B-ET) of film-mulched hybrid maize seed production to different irrigation and N treatments in the Hexi Corridor, Northwest China during April to September in 2013 and also during April to September in 2014. Three irrigation levels (70%-65%, 60%-55%, and 50%-45% of the field capacity) combined with three N rates (500, 400, and 300 kg N/hm²) were tested in 2013. The N treatments were adjusted to 500, 300, and 100 kg N/hm² in 2014. Results showed that the responses of WP_Y-ET and WP_B-ET to different irrigation amounts were different. WP_Y-ET was significantly reduced by lowering irrigation amounts while WP_B-ET stayed relatively insensitive to irrigation amounts. However, WP_Y-ET and WP_B-ET behaved consistently when subjected to different N treatments. There was a slight effect of reducing N input from 500 to 300 kg/hm² on the WP_Y-ET and WP_B-ET, however, when reducing N input to 100 kg/hm², the values of WP_Y-ET and WP_B-ET were significantly reduced. Water is the primary factor and N is the secondary factor in determining both yield (Y) and final biomass (B). Partial factor productivity from applied N (PFP_N) was the maximum under the higher irrigation level and in lower N rate (100-300 kg N/hm²) in both years (2013 and 2014). Lowering the irrigation amount significantly reduced evapotranspiration (ET), but ET did not vary with different N rates (100-500 kg N/hm²). Both Y and B had robust linear relationships with ET, but the correlation between B and ET (R²=0.8588) was much better than that between Y and ET (R²=0.6062). When ET increased, WP_Y-ET linearly increased and WP_B-ET decreased. Taking the indices of Y, B, WP_Y-ET, WP_B-ET and PFP_N into account, a higher irrigation level (70%-65% of the field capacity) and a lower N rate (100-300 kg N/hm²) are recommended to be a proper irrigation and N application strategy for plastic film-mulched hybrid maize seed production in arid Northwest China.

Plant capacity for water storage leads to time lags between basal stem sap flow and transpiration in various woody plants. Internal water storage depends on the sizes of woody plants. However, the changes and its influencing factors in time lags of basal stem flow during the development of herbaceous plants including crops remain unclear. A field experiment was conducted in an arid region of Northwest China to examine the time lag characteristics of sap flow in seed-maize and to calibrate the transpiration modeling. Cross-correlation analysis was used to estimate the time lags between stem sap flow and meteorological driving factors including solar radiation (R_s) and vapor pressure deficit of the air (VPD_air). Results indicate that the changes in seed-maize stem sap flow consistently lagged behind the changes in R_s and preceded the changes in VPD_air both on hourly and daily scales, suggesting that light-mediated stomatal closures drove sap flow responses. The time lag in the maize’s sap flow differed significantly during different growth stages and the difference was potentially due to developmental changes in capacitance tissue and/or xylem during ontogenesis. The time lags between stem sap flow and Rs in both female plants and male plants corresponded to plant use of stored water and were independent of total plant water use. Time lags of sap flow were always longer in male plants than in female plants. Theoretically, dry soil may decrease the speed by which sap flow adjusts ahead of shifts in VPD_air in comparison with wet soil and also increase the speed by which sap flow adjusts to R_s. However, sap flow lags that were associated with Rs before irrigation and after irrigation in female plants did not shift. Time series analysis method provided better results for simulating seed-maize sap flow with advantages of allowing for fewer variables to be included. This approach would be helpful in improving the accuracy of estimation for canopy transpiration and conductance using meteorological measurements.

Greenhousing is a technique to bridge season gap in vegetable production and has been widely used worldwide. Calculation of water requirement of crops grown in greenhouse and determination of their irrigation schedules in arid and semi-arid regions are essential for greenhouse maintenance and have thus attracted increased attention over the past decades. The most common method used in the literature to estimate crop evapotranspiration (ET) is the Penman-Monteith (PM) formula. When applied to greenhouse, however, it often uses canopy resistance instead of surface resistance. It is understood that the surface resistance in greenhouse is the result of a combined effect of canopy restriction and soil-surface restriction to water vapor flow, and the relative dominance of one restriction over another depends on crop canopy. In this paper, we developed a surface resistance model in a way similar to two parallel resistances in an electrical circuit to account for both restrictions. Also, considering that wind speed in greenhouse is normally rather small, we compared three methods available in the literature to calculate the aerodynamic resistance, which are the r_a1 method proposed by Perrier (1975a, b), the r_a2 method proposed by Thom and Oliver (1977), and the r_a3 method proposed by Zhang and Lemeu (1992). We validated the model against ET of tomatoes in a greenhouse measured from sap flow system combined with micro-lysimeter in 2015 and with weighing lysimeter in 2016. The results showed that the proposed surface resistance model improved the accuracy of the PM model, especially when the leaf area index was low and the greenhouse was being irrigated. We also found that the aerodynamic resistance calculated from the r_a1 and r_a3 methods is applicable to the greenhouse although the latter is slightly more accurate than the former. The proposed surface resistance model, together with the r_a3 method for aerodynamic resistance, offers an improved approach to estimate ET in greenhouse using the PM formula.

Dew is an important water source for plants in arid and semi-arid regions. However, information on dew is scarce in such regions. In this study, we explored dew formation, amount, and duration of rain-fed jujube (Zizyphus jujube Mill) trees in a semi-arid loess hilly region of China (i.e., Mizhi County). The data included dew intensity and duration, relative humidity, temperature, and wind speed measured from 26 July to 23 October, 2012 and from 24 June to 17 October, 2013 using a micro-climate system (including dielectric leaf wetness sensors, VP-3 Relative Humidity/Temperature Sensor, High Resolution Rain Gauge, and Davis Cup Anemometer). The results show that atmospheric conditions of relative humidity of >78% and dew point temperature of 1°C-3°C are significantly favorable to dew formation. Compared with the rainfall, dew was characterized by high frequency, strong stability, and long duration. Furthermore, heavy dew accounted for a large proportion of the total amount. The empirical models (i.e., relative humidity model (RH model) and dew point depression model (DPD model)) for daily dew duration estimation performed well at 15-min intervals, with low errors ranging between 1.29 and 1.60 h, respectively. But it should be noted that the models should be calibrated firstly by determining the optimal thresholds of relatively humidity for RH model and dew point depression for DPD model. For rain-fed jujube trees in the semi-arid loess hilly regions of China, the optimal threshold of relative humidity was 78%, and the optimal upper and lower thresholds of dew point depression were 1°C and 5°C, respectively. The study further demonstrates that dew is an important water resource that cannot be ignored for rain-fed jujube trees and may affect water balance at regional scales.