耦合WorldView-2和珠海一号影像的红树林物种分布

红树林物种遥感分类受影像分辨率、波段信息、分类策略、影像特征选择方法等因素影响。现有的红树林物种遥感分类研究大多关注于分类精度的比较,较少报道商业高分与国产高光谱卫星的结合在物种分类的应用。基于湛江高桥红树林保护区的高分（WorldView-2）和高光谱（珠海一号）卫星影像,本文旨在利用随机森林分类方法,比较不同特征选择算法（极端梯度提升（XGBoost）、极端随机树（ERT）、连续投影算法（SPA））和不同空间分辨率（WorldView-2影像从0.5 m的分辨率重采样为1 m、2 m、4 m、8 m、10 m）对红树林物种分类精度的影响,并耦合WorldView-2与珠海一号影像研究红树林物种空间分布格局及影响因素。每种空间分辨率的WorldView-2影像提取248个特征（52个光谱特征及196个纹理特征）,珠海一号影像提取117个光谱特征。结果表明,XGBoost方法优于ERT和SPA,在影像特征选择中具有巨大优势;与其他空间分辨率相比,2 m分辨率的WorldView-2影像具有更高的分类精度;耦合WorldView-2和珠海一号影像的分类精度（分辨率：2 m,总体精度：88.98%,kappa系数：0.846）高于单一的WorldView-2影像（分辨率：2 m,总体精度：83.47%,kappa系数：0.768）和珠海一号影像（分辨率：10 m,总体精度：78.50%,kappa系数：0.703）;另外,林窗、高程、离岸距离一定程度上影响红树林物种分布格局。本文表明WorldView-2与珠海一号高光谱影像的结合在景观及区域尺度上精确提取红树林物种信息具有巨大潜力,有利于红树林生物多样性保护及科学管理。     

On the future navigability of Arctic sea routes: High-resolution projections of the Arctic Ocean and sea ice

The rapid Arctic summer sea ice reduction in the last decade has lead to debates in the maritime industries on the possibility of an increase in cargo transportation in the region. Average sailing times on the North Sea Route along the Siberian Coast have fallen from 20 days in the 1990s to 11 days in 2012–2013, attributed to easing sea ice conditions along the Siberian coast. However, the economic risk of exploiting the Arctic shipping routes is substantial. Here a detailed high-resolution projection of ocean and sea ice to the end of the 21st century forced with the RCP8.5 IPCC emission scenario is used to examine navigability of the Arctic sea routes. In summer, opening of large areas of the Arctic Ocean previously covered by pack ice to the wind and surface waves leads to Arctic pack ice cover evolving into the Marginal Ice Zone. The emerging state of the Arctic Ocean features more fragmented thinner sea ice, stronger winds, ocean currents and waves. By the mid 21st century, summer season sailing times along the route via the North Pole are estimated to be 13–17 days, which could make this route as fast as the North Sea Route.     

Projections of Extreme Rainfall in Hong Kong in the 21st Century

The possible changes in the frequency of extreme rainfall events in Hong Kong in the 21st century wereinvestigated by statistically downscaling 30 sets of the daily global climate model projections (involvinga combination of 12 models and 3 greenhouse gas emission scenarios,namely,A2,A1B,and B1) of theFourth Assessment Report of the Intergovernmental Panel on Climate Change.To cater for the intermittentand skewed character of the daily rainfall,multiple stepwise logistic regression and multiple stepwise linearregression were employed to develop the downscaling models for predicting rainfall occurrence and rainfallamount,respectively.Verification of the simulation of the 1971-2000 climate reveals that the models ingeneral have an acceptable skill in reproducing past statistics of extreme rainfall events in Hong Kong.Theprojection results suggest that,in the 21st century,the annual number of rain days in Hong Kong is expectedto decrease while the daily rainfall intensity will increase,concurrent with the expected increase in annualrainfall.Based on the multi-model scenario ensemble mean,the annual number of rain day is expected todrop from 104 days in 1980-1999 to about 77 days in 2090-2099.For extreme rainfall events,about 90% ofthe model-scenario combinations indicate an increase in the annual number of days with daily rainfall 100mm (R100) towards the end of the 21st century.The mean number of R100 is expected to increase from 3.5days in 1980-1999 to about 5.3 days in 2090-2099.The projected changes in other extreme rainfall indicesalso suggest that the rainfall in Hong Kong in the 21st century may also become more extreme with moreuneven distributions of wet and dry periods.While most of the model-emission scenarios in general projectconsistent trends in the change of rainfall extremes in the 21st century,there is a large divergence in theprojections among different model/emission scenarios.This reflects that there are still large uncertainties inmodel simulations of future extreme rainfall events.     

由锥形束投影监控物体体密度变化的快速方法

体密度是物体特征信息的一个重要组成部分,该信息的动态监控可用来快速判断物体体密度是否发生变化。目前最常用的体密度监控方法是基于图像重建的近似计算,该方法首先利用锥形束投影数据实现体积重建,然后对物体的密度函数积分求得物体的体密度。这种方法算法复杂度较高,很难实现对体密度的实时动态监控。本文提出一种直接基于锥形束投影的体密度快速监控方法,不需要图像重建。仿真实验结果表明,这种方法在动态监控物体体密度变化时,具有较高的实时性和精确性,可以满足实际的动态监控需求,因而在工业检测等领域具有一定的应用价值。