Seasonal variation in intensive agricultural land use patterns and its spatial scale effects on water nitrogen (N) and phosphorus (P) concentrations in watersheds have always been major concerns. However, the relationships and contributions of the important land use metrics to the pond water N and P concentrations remain uncertain. In this context, the present study aims to investigate the seasonal variation in the N and P concentrations in 28 ponds of a developed agricultural watershed in central China in 2019. Multivariate statistical analyses were applied to reveal the relationships between the multi-scale land use patterns and pond water N and P concentrations. The obtained results showed substantial seasonal variations in total nitrogen (TN), nitrate nitrogen (NO3--N), ammonia nitrogen (NH4+-N), and total phosphorus (TP). The highest average concentrations of TN, NO3--N, TP, and DP in ponds were observed in summer. Redundancy analysis (RDA) showed that the land use patterns predicted better the variations in the pond N and P concentrations at the buffer zone scales than at the catchment scale. Land use patterns showed the greatest effects on the pond N and P concentrations at the 100 m buffer zone scale in spring and summer, and at the 50 m buffer zone scale in fall and winter. In addition, the importance of the land use metrics showed seasonal differences. The most contributions to the pond N and P concentrations were made by the orchard area proportion (POR) in spring and summer and the landscape shape index (LSI) in fall and winter. Stepwise multiple linear regression (SMLR) also indicated spatial consistency in the scale effects on the N and P variables. The results of the this study provide valuable information for effective land use management to control pond water pollution and promote sustainable development in agricultural watersheds.

China contributes the largest share of cropland's greenhouse gas (GHG) emissions globally. Processed-based biogeochemical models are useful tools to simulate GHG emissions from cropping systems. However, model comparisons are necessary to provide information for the application of models under different climate, soil, and crop conditions. In this study, two widely-used models (DayCent and DNDC) were evaluated and compared under four main cropping systems in China. The field observations from nine experiments were used for model calibration and validation. The DayCent and DNDC models simulated daily and seasonal CH4 emissions from early rice-late rice and rice-wheat cropping systems reasonably well (r2≥0.49 for daily simulation and nRMSE≤52.9% for seasonal simulation). Both models were able to satisfactorily predict seasonal N2O emissions from maize-wheat fields (0.6≤d≤0.8), but overestimated most daily N2O fluxes at fertilisation and irrigation events. Significantly positive relationships were found between simulated and observed cumulative N2O fluxes in spring maize growing season (0.61≤ r2≤0.85). The DNDC showed smaller differences in simulated and observed cumulative GHG emissions for spring maize and double rice, while DayCent showed better performance on estimating N2O and CH4 for maize-wheat and rice-wheat. This study shows that both models have strengths and weaknesses under a variety of cropping systems and growing regions, which are important to consider when choosing a model for a crop/region-specific simulation.

The methods are very efficient and can handle all ISCAS'85 benchmark circuits in minutes. On average, it was found that the OTR method gave 40%, and the Static/PTL gave 50% delay reductions over SIS, with substantial area savings. Finally, we extend the technology mapping work to interleave it with placement in a single optimization. Conventional methods that perform these steps separately will not be adequate for next-generation circuits. Our approach presents an integerated solution to this problem, and shows an average of 28.19%, and a maximum of 78.42% improvement in the delay over a method that performs the two optimizations in separate steps.