Efficient collection of training data for sub-pixel land cover classification using neural networks

Artificial neural networks (ANNs) are a popular class of techniques for performing soft classifications of satellite images. They have successfully been applied for estimating crop areas through sub-pixel classification of medium to low resolution images. Before a network can be used for classification and estimation, however, it has to be trained. The collection of the reference area fractions needed to train an ANN is often both time-consuming and expensive. This study focuses on strategies for decreasing the efforts needed to collect the necessary reference data, without compromising the accuracy of the resulting area estimates. Two aspects were studied: the spatial sampling scheme (i) and the possibility for reusing trained networks in multiple consecutive seasons (ii). Belgium was chosen as the study area because of the vast amount of reference data available. Time series of monthly NDVI composites for both SPOT-VGT and MODIS were used as the network inputs. The results showed that accurate regional crop area estimation (R² > 80%) is possible using only 1% of the entire area for network training, provided that the training samples used are representative for the land use variability present in the study area. Limiting the training samples to a specific subset of the population, either geographically or thematically, significantly decreased the accuracy of the estimates. The results also indicate that the use of ANNs trained with data from one season to estimate area fractions in another season is not to be recommended. The interannual variability observed in the endmembers’ spectral signatures underlines the importance of using up-to-date training samples. It can thus be concluded that the representativeness of the training samples, both regarding the spatial and the temporal aspects, is an important issue in crop area estimation using ANNs that should not easily be ignored.     

Trace Metals in Selected Benthic Invertebrates from Belgian Coastal Waters (1981–1996)

Mercury, cadmium, lead, copper, zinc and chromium were assessed in brown shrimp (Crangon crangon), swimming crab (Liocarcinus holsatus), hermit crab (Pagurus bernhardus), starfish (Asterias rubens) and cut trough shell (Spisula subtruncata) from ten sampling stations off the Belgian coast, including three dredge spoils disposal sites. Calculations of time trends indicated that on the whole concentrations of trace metals had significantly decreased since 1981, also on dredged material dumping sites. Starfish and hermit crab were found to be the most promising benthic indicator species for metal contamination. Globally, the location of the sites, including the dredge spoils disposal zones, showed to be of little influence on the concentrations of trace metals.     

渗流井合理竖井降深确定

通过采用单位面积河流在单位水头差作用下的渗漏量来表征河流渗漏能力,建立渗流井取水理想模型,分别计算了在不同河流渗漏能力和含水层渗透性能条件下,竖井降深对渗流井出水量的影响.建立渗流井取水非稳定流模型,计算了在前期稳定竖井降深不同条件下,河流断流后渗流井出水量衰减过程及竖井降深发展过程.提出渗流井合理竖井降深应根据河流与地下水是否脱节以及含水层渗透性能,在岸边渗流井中部及一侧各布设一个观测孔,根据观测孔水位进行确定.对于含水层渗透性能较强地区,渗流井竖井降深应使得渗流井范围内地下水位与河流脱节,但高于辐射孔顶面;对于含水层渗透性能较差地区,渗流井竖井降深应使得侧部观测孔水位接近河床底面或刚出现脱节.     

QBO Signal in total ozone over Tibet

1. IntroductionOzone is one of the trace gases in the atmosphere distributed in 10--50 km altitude withthe maximum in 20--28 km. Ozone is significant in the following three aspects impacting theclimate and environment: 1) Ozone absorbs harmful solar ultra--violet radiation for protecting the ecological system on the Earth; 2) ozone heats the stratosphere and forces the circulation systems in this layer; 3) ozone variation in the stratosphere can change the incomingradiation at the surface leve…