摘要：
Carbon dioxide (CO2) is the most important anthropogenic greenhouse gas causing global warming. An increasing number of studies have focused on urban areas recently because cities are major anthropogenic sources of CO2 and also the main habitats of most human beings. However, the complicated nature of urban landscapes and the inhomogeneous distributions of CO2 sources and sinks lead to methodological difficulties in CO2 observation. This paper introduces a new approach to estimate CO2 concentration from satellite imagery using a Bayesian network. An estimation model based on Bayesian network was built to characterize the quantitative relationships between remote-sensing data and CO2 concentrations. Comparative analysis of the proposed model and multiple regression models was then carried out. The feasibility of estimating carbon dioxide concentrations in urban areas from satellite imagery was analyzed, and the advantages of modeling land-surface parameters using the Bayesian network were addressed.
摘要：
Miscanthus has been rated as one of the most promising bioenergy crops due to its potential for biomass production. The sustainable production of Miscanthus for bioenergy feedstock partly depends on the varieties that are efficient in terms of nutrient use for the production of biomass. In this study, 23 Miscanthus accessions, collected from wide range of geographic regions, were established early in March 2010 in Wuhan, China. The feedstock was sampled for nutrient concentration determination late in November 2010 and 2011 (at physiological maturity), and harvested early in February 2011 and 2012 (after a killing frost) to evaluate the biomass yield, nutrient concentration and removal. Across these two years, the biomass yield was negatively related to the latitude of the original collection sites. A significant increase in biomass production was observed in the second growth year relative to that in the first growth year among almost all of the Miscanthus accessions. The accessions of MS267 and MS321 only yielded 1.32 and 1.91 ton ha(-1) in 2010, respectively, but the biomass yield increased dramatically to 11.23 and 22.76 ton ha(-1) in 2011, leading to greater nutrient removal by the final harvest and thus the requirement of much more fertilizer in the following years. The accessions MS92, MS145, MS262, MS436 and MS438 established poor biomass yields, averaging <1 ton ha(-1) in the first two years, which suggests that they may be unsuitable for planting in the present environment. Significant differences between accessions were found for the nutrient concentration at maturity and after frost. Notable differences in the nutrient concentration after frost and nutrient removal were presented among the Miscanthus accessions. In addition, the significant difference conferred the possibility of achieving a desirable cultivar with significant biomass yield and relative nutrient removal by harvest. The accessions MS434, MS461 and MS296 had consistently high biomass yield and relatively low nutrient removal, demonstrating desirable characteristics as a low-input bioenergy crop. The results are important for guiding the agronomical practices of nutrient management and genetic improvement for nutrient-use efficiency to increase biomass production with low fertilizer input. (C) 2013 Elsevier B.V. All rights reserved.
作者机构：
[周勇; 夏天; 于雷] College of Urban and Environmental Science, Huazhong Normal University, Wuhan 430079, China;[郧文聚] Land Consolidation and Rehabilitation Center, Ministry of Land and Resources, Beijing 100035, China
通讯机构：
College of Urban and Environmental Science, Huazhong Normal University, China
