a b s t r a c tSimulation of low flow process is critical to water quality, water supply, and aquatic habitat. However, the poor performance of Soil and Water Assessment Tool (SWAT) in dry seasons has impeded its application to watersheds characterized largely by low-flows. Aiming at overcoming this shortage, a seasonal calibration scheme was proposed, in which SWAT was calibrated separately for the dry and wet periods and the "optimal" simulation results of these two periods were combined into a complete runoff series. An extended SWAT model incorporating with the proposed seasonal calibration scheme, named SWAT-SC was constructed and compared with the original SWAT to simulate daily runoff in the Jinjiang watershed dominated by a typical subtropical monsoon climate in southeastern China. The study reveals that when Nash-Sutcliffe efficiency (ENS) of the original SWAT model indicated a satisfied model performance in a wet season or a whole year, it may not guaranty acceptable performance for the dry period. A significant improvement was achieved by using SWAT-SC for simulating runoffs in the dry period, and although not as notably as the dry period, improvements for runoff simulation of the wet and overall periods were observed as well.
We propose a two-stage model selection procedure for the linear mixed-effects models. The procedure consists of two steps: First, penalized restricted log-likelihood is used to select the random effects, and this is done by adopting a Newton-type algorithm. Next, the penalized log-likelihood is used to select the fixed effects via pathwise coordinate optimization to improve the computation efficiency. We prove that our procedure has the oracle properties. Both simulation studies and a real data example are carried out to examine finite sample performance of the proposed fixed and random effects selection procedure. Supplementary materials including R code used in this article and proofs for the theorems are available online.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.