Xiaohu Wen

Support-Vector-Machine-Based Models for Modeling Daily Reference Evapotranspiration With Limited Climatic Data in Extreme Arid Regions

Evapotranspiration is a major factor that controls hydrological process and its accurate estimation provides valuable information for water resources planning and management, particularly in extremely arid regions. The objective of this research was to evaluate the use of a support vector machine (SVM) to model daily reference evapotranspiration (ET0) using limited climatic data. For the SVM, four combinations of maximum air temperature (Tmax), minimum air temperature (Tmin), wind speed (U2) and daily solar radiation (Rs) in the extremely arid region of Ejina basin, China, were used as inputs with Tmax and Tmin as the base data set. The results of SVM models were evaluated by comparing the output with the ET0 calculated using Penman–Monteith FAO 56 equation (PMF-56). We found that the ET0 estimated using SVM with limited climatic data was in good agreement with those obtained using the conventional PMF-56 equation employing the full complement of meteorological data. In particular, three climatic parameters, Tmax, Tmin, and Rs were enough to predict the daily ET0 satisfactorily. Moreover, the performance of SVM method was also compared with that of artificial neural network (ANN) and three empirical models including Priestley-Taylor, Hargreaves, and Ritchie. The results showed that the performance of SVM method was the best among these models. This offers significant potential for more accurate estimation of the ET0 with scarce data in extreme arid regions.

Identifying separate impacts of climate and land use/cover change on hydrological processes in upper stream of Heihe River, Northwest China

&NA; Climate change and land use/cover change (LUCC) are two factors that produce major impacts on hydrological processes. Understanding and quantifying their respective influence is of great importance for water resources management and socioeconomic activities as well as policy and planning for sustainable development. In this study, the Soil and Water Assessment Tool (SWAT) was calibrated and validated in upper stream of the Heihe River in Northwest China. The reliability of the SWAT model was corroborated in terms of the Nash‐Sutcliffe efficiency (NSE), the correlation coefficient (R), and the relative bias error (BIAS). The findings proposed a new method employing statistical separation procedures using a physically based modeling system for identifying the individual impacts of climate change and LUCC on hydrology processes, in particular on the aspects of runoff and evapotranspiration (ET). The results confirmed that SWAT was a powerful and accurate model for diagnosis of a key challenge facing the Heihe River Basin. The model assessment metrics, NSE, R, and BIAS, in the data were 0.91%, 0.95%, and 1.14%, respectively, for the calibration period and 0.90%, 0.96%, and −0.15%, respectively, for the validation period. An assessment of climate change possibility showed that precipitation, runoff, and air temperature exhibited upward trends with a rate of 15.7 mm, 6.1 mm, and 0.38 °C per decade for the 1980 to 2010 period, respectively. Evaluation of LUCC showed that the changes in growth of vegetation, including forestland, grassland, and the shrub area have increased gradually while the barren area has decreased. The integrated effects of LUCC and climate change increased runoff and ET values by 3.2% and 6.6% of the total runoff and ET, respectively. Climate change outweighed the impact of LUCC, thus showing respective increases in runoff and ET of about 107.3% and 81.2% of the total changes. The LUCC influence appeared to be modest by comparison and showed about −7.3% and 18.8% changes relative to the totals, respectively. The increase in runoff caused by climate change factors is more than the offsetting decreases resulting from LUCC. The outcomes of this study show that the climate factors accounted for the notable effects more significantly than LUCC on hydrological processes in the upper stream of the Heihe River.

Wavelet and adaptive neuro-fuzzy inference system conjunction model for groundwater level predicting in a coastal aquifer

Accurately predicting groundwater level (GWL) fluctuations is one of the most important issues for managing groundwater resources. In this study, the feasibility of predicting weekly GWL fluctuations in a coastal aquifer using the wavelet-adaptive neuro-fuzzy inference system (WANFIS) was investigated. WANFIS was a conjunction model that combined discrete wavelet transform and adaptive neuro-fuzzy inference system (ANFIS). GWL data of two wells located in the coastal aquifer of eastern Laizhou bay, China, were used to establish WANFIS model. The performances of WANFIS model, along with ANFIS model, were assessed in terms of the following statistical indices, such as coefficient of correlation (R), root mean square error, and mean absolute relative error. Compared with the best ANFIS models, the best WANFIS model gave a better prediction. Moreover, it was found that wavelet transform positively affected the ANFIS’s predicting ability. In addition, the WANFIS model was also found to be superior to the best ANN model. This study indicated that WANFIS model was preferable and could be applied successfully due to its high accuracy and reliability for predicting GWL.

Risk assessment and source identification of coastal groundwater nitrate in northern China using dual nitrate isotopes combined with Bayesian mixing model

ABSTRACT Due to the intensive and complicated human activities, the identification of nitrate pollution source of coastal aquifer is usually a challenge. This study firstly adopted stable isotope technique and stable isotope analysis in R (SIAR) model to identify the nitrate sources and contribution proportions of different sources in typical coastal groundwater of northern China. The results showed that about 91.5% of the groundwater samples illustrated significantly high nitrate concentrations exceeding the maximum WHO drinking water standard (50 mg/l), reflecting the high risk of groundwater nitrate pollution in the coastal area. A total of 57 sampling sites were classified into three groups according to hierarchical cluster analysis (HCA). The δ15N-NO3− and δ18O-NO3− values of groundwater samples from Group C (including nine samples) were much higher than those from Group A (including 40 samples) and Group B (including 8 samples). SIAR results showed that NH4+ fertilizer was the dominant nitrate source for groundwater samples of Groups A and B while manure and sewage (M&S) served as dominant source for Group C. This study provided essential information on the high risk and pollution sources of coastal groundwater nitrate of northern China.

Separation of the Climatic and Land Cover Impacts on the Flow Regime Changes in Two Watersheds of Northeastern Tibetan Plateau

Assessment of the effects of climate change and land use/cover change (LUCC) on the flow regimes in watershed regions is a fundamental research need in terms of the sustainable water resources management and ecosocial developments. In this study, a statistical and modeling integrated method utilizing the Soil and Water Assessment Tool (SWAT) has been adopted in two watersheds of northeastern Tibetan Plateau to separate the individual impacts of climate and LUCC on the flow regime metrics. The integrated effects of both LUCC and climate change have led to an increase in the annual streamflow in the Yingluoxia catchment (YLC) region and a decline in the Minxian catchment (MXC) region by 3.2% and 4.3% of their total streamflow, respectively. Climate change has shown an increase in streamflow in YLC and a decline in MXC region, occupying 107.3% and 93.75% of the total streamflow changes, respectively, a reflection of climatic latitude effect on streamflow. It is thus construed that the climatic factors contribute to more significant influence than LUCC on the magnitude, variability, duration, and component of the flow regimes, implying that the climate certainly dominates the flow regime changes in northeastern Tibetan Plateau.

Application of multivariate recursive nesting bias correction, multiscale wavelet entropy and AI-based models to improve future precipitation projection in upstream of the Heihe River, Northwest China

Accurate projection of future precipitation is a major challenge due to the uncertainties arising from the atmospheric predictors and the inherent biases that exist in the global circulation models. In this study, we employed multivariate recursive nesting bias correction (MRNBC) and multiscale wavelet entropy (MWE) to reduce the bias and improve the projection of future (i.e., 2006–2100) precipitation with artificial intelligence (AI)-based data-driven models. Application of the developed method and the subsequent analyses are performed based on representative concentration pathway (RCP) scenarios: RCP4.5 and RCP8.5 of eight Coupled Model Intercomparison Project Phase-5 (CMIP5) Earth system models for the upstream of the Heihe River. The results confirmed the MRNBC and MWE were important statistical approaches prudent in simulation performance improvement and projection uncertainty reduction. The AI-based methods were superior to linear regression method in precipitation projection. The selected CMIP5 outputs showed agreement in the projection of future precipitation under two scenarios. The future precipitation under RCP8.5 exhibited a significantly increasing trend in relative to RCP4.5. In the future, the precipitation will experience an increase by 15–19% from 2020 to 2050 and by 21–33% from 2060 to 2090.