Within a river catchment, the relationship between pollutant load migration and its related factors is nonlinear generally. When neural network models are used to identify the nonlinear relationship, data scarcity and random weight initialization might result in overfitting and instability. In this paper, we propose an averaged weight initialization neural network (AWINN) to realize the multi-index integrated prediction of a pollutant load under data scarcity. The results show that (1) compared with the particle swarm optimization neural network (PSONN) and AdaboostR models that prevent overfitting, AWINN improved simulation accuracy significantly. The R² in test sets of different pollutant load models reached 0.51–0.80. (2) AWINN is effective in overcoming instability. With more hidden layers, the stability of the models’ outputs was stronger. (3) Sobol sensitivity analysis explained that the main influencing factors of the whole process were the flows of the catchment inlet and outlet, and main factors changed across seasons. The algorithm proposed in this paper can realize stably integrated prediction of pollutant load in the catchment under data scarcity and help to understand the mechanism that influences pollutant load migration.

Agricultural production depends on local agroclimatic conditions to a great extent, affected by ENSO and other ocean-atmospheric climate modes. This paper analyzed the spatio-temporal distributions of climate elements in the Jianghan Plain (JHP), Central China, and explored the impacts from teleconnection patterns, aimed at providing references for dealing with climate change and guiding agricultural activities. Both linear and multifactorial regression models were constructed based on the frequentist quantile regression and Bayesian quantile regression method, with the daily meteorological data sets of 17 national stations in the plain and teleconnection climate characteristic indices. The results showed that precipitation in JHP had stronger spatial variability than evapotranspiration. El Niño probably induced less precipitation in summer while the weakening Arctic Oscillation might lead to more summertime precipitation. The Nash-Sutcliffe efficiency (NSE) of the multifactorial and linear regression model at the median level were 0.42–0.56 and 0.12–0.18, respectively. The mean relative error (MRE) ranged −2.95–−0.26% and −7.83–0.94%, respectively, indicating the much better fitting accuracy of the multiple climatic factors model. Meanwhile it confirmed that the agricultural climate in JHP was under the influence from multiple teleconnection patterns.