基于遥感的千烟洲人工林蒸散及其组分模拟研究

论文通过改进遥感蒸散模型的关键参数,结合遥感数据和气象观测数据,对2003—2008年江西千烟洲人工林生态系统蒸散及其组分进行模拟,并利用涡度相关技术获取的蒸散实测数据对模型模拟结果进行验证和评价。结果表明:①年均蒸散总量模拟值比实测值偏低2.4%,决定系数与均方根误差分别为0.83和0.61 mm·d^-1。②土壤蒸发、林冠截留蒸发和植被蒸腾分别占总蒸散量的12%、23%和65%。其中,土壤蒸发季节及年际变化相对稳定;林冠截留蒸发季节变化明显且在不同年份差异较大;植被蒸腾季节变化明显,但年际变异较小。③1—3月植被光合作用较弱,植被蒸腾与蒸散比小于30%。随着植被蒸腾的增强,从4月开始植被蒸腾与蒸散比迅速增加,在生长旺季(7月底)可达到约90%。由于该模型所需数据在区域尺度较易获取,从而为开展区域尺度中亚热带人工林生态系统蒸散及其组分模拟提供方法支撑。     

温升包络线在核电站温排水监测中的应用

采用核电站温排水遥感监测影像叠加合并方法提取温升包络线,监测核电站附近海域海表温升数据,确定核电温排水的影响范围,对海面温升面积进行计算,判断核电站温排水是否符合我国相关海洋空间管理标准。以某核电站实例表明,通过遥感监测获得的温升包络线在核电站温排水监测中具有应用价值。     

基于遥感和ULSE模型评价1995~2005年江西土壤侵蚀

为了定量评价近10年江西省土壤侵蚀的时空变化,采用USLE模型的月模式,耦合1995年的NOAA-AVHRR归一化植被指数(NDVI)和2005年的Terra-MODIS增强型植被指数(EVI),定量评价了江西省1995年和2005年的土壤侵蚀空间分布.研究结果表明: (1)江西省土壤侵蚀主要受地形条件的影响,存在明显的空间差异,强烈侵蚀级别以上的土壤侵蚀主要发生在鄱阳湖流域各子流域的上游;地形平坦的鄱阳湖平原和吉泰盆地,土壤侵蚀强度低;(2)江西省土壤侵蚀在1995~2005年间得到一定程度的改善,年土壤侵蚀总量由1995年的1.494亿t下降为2005年的1.268亿t,全省平均土壤侵蚀模数由894.68t/(km2.a)下降为759.31t/(km2.a);中度以上侵蚀面积由1995年的16 945km2下降为2005年的13 833km2;(3)全省绝大多数县(市、区)的土壤侵蚀强度降低,其中德兴县、玉山县、婺源县、上饶县、浮梁县、横峰县等赣东北5县的平均土壤侵蚀模数下降最为显著;但也有少量县(市区)的土壤侵蚀呈进一步扩大趋势,主要有井冈山市、永新县、莲花县、瑞昌县、广昌县和永丰县.     

淀山湖叶绿素a的高光谱遥感监测研究

为实现淀山湖水体叶绿素a的快速监测, 于2006年春季对淀山湖进行了现场光谱测量和同步的采样、实验室分析,通过分析叶绿素a和光谱数据之间的关系,建立了多个基于高光谱数据的叶绿素a浓度的估算模型。结果表明：利用归一化反射率建立的模型中精度最高的是采用708 nm反射率的一元二次模型,R2为0.608;利用R710nm/R667nm反射比建立模型精度得到提高,R2接近0.68;利用695.5 nm一阶微分建立的指数模型精度进一步提高,其R2达到了0.76;分别利用708 nm和667 nm的归一化反射率对数作自变量进行多元回归得到的模型精度最高,R2接近0.8,可以用来实现对叶绿素a的定量估算,实现叶绿素a的快速监测。     

利用混合水体指数模型(CIWI)提取滨海湿地水体的信息

根据对滨海湿地区域光谱特性分析和背景信息的反射特点,对主要的水体信息提取方法进行比较分析,将混合水体指数模型（CIWI）应用于TM遥感影像,针对滨海湿地区域水体信息进行提取实验,并与近年得到广泛应用和承认的水体指数模型：归一化差异水体指数（NDWI）、改进的归一化差异水体指数（MNDWI）等进行比较。实验证明CIWI的区分度最高,水体与植被、水体与城镇、植被与城镇之间的区分度分别为505%、292%与267%;水体、植被、城镇像元之间的差异比NDWI和MNDWI增大,减少了混淆信息的影响,对江苏盐城滨海湿地研究区的水体信息提取精度达到949%,实现了滨海湿地区域水体信息的准确快速提取。     

基于FY2C卫星数据的藏北高原降水估算研究

如何确切的掌握降水的时空分布,对区域气候、水文和生态应用等至关重要。以藏北高原典型区为研究区,在大量地面实况降水观测数据与对长时间序列FY 2C 影像光谱特征和云图特征分析的基础上,获取卫星降水模拟参数特征集以刻画云降水的发生与发展过程,选用最值归一化方法对不同量纲云图特征参数进行归一化处理。构建基于三层前向型反向传播神经网络的卫星降水估算模型,用于该地域降水估算,并采用多指标体系分析模型的降水模拟精度。结果表明：静止气象卫星红外波段能较精确地揭示云的降水机理,较高时间分辨率遥感图像可以监测云图的变化细节,并获取能够反映云图降水特征的降水模拟参数;人工神经网络能较好地刻画该地域卫星降水特征的非线性规律;三层前向型反向传播神经网络卫星降水估算模型的估算结果与雨量计实测值间的相关性可以达到0.57。模型估算结果系统性的低估偏小,预示着对该地域弱降水强度将有较好的指示性     

Application of Remote Sensing Detection and GIS in Analysis of Vegetation Pattern Dynamics in the Yellow River Delta

Abstract

Regional vegetation pattern dynamics has a great impact on ecosystem and climate change. Remote sensing data and geographical information system (GIS) analysis were widely used in the detection of vegetation pattern dynamics. In this study, the Yellow River Delta was selected as the study area. By using 1986, 1993, 1996, 1999 and 2005 remote sensing data as basic information resource, with the support of GIS, a wetland vegetation spatial information dataset was built up. Through selecting the land-scape metrics such as class area (CA), class percent of landscape (PL), number of patch (NP), largest patch index (LPI) and mean patch size (MPS) etc., the dynamics of vegetation pattern was analyzed. The result showed that the change of vegetation pattern is significant from 1986 to 2005. From 1986–1999, the area of the vegetation, the percent of vegetation, LPI and MPS decreased, the NP increased, the vegetation pattern tends to be fragmental. The decrease in vegetation area may well be explained by the fact of the nature environment evolution (Climate change and decrease in Yellow River runoff) and the increase in the population in the Yellow River Delta. However, from 1999–2005, the area of the vegetation, the percent of vegetation, LPI and MPS increased, while the NP decreased. This trend of restoration may be due to the implementation of water resources regulation for the Yellow River Delta since 1999.     

On East Asian Sand and Duststorms and Associated Significant Dustfall Observed from January to May 2001

Monitoring of dust pollution at the western shore of Tae-ahn Peninsula (TAP) and in the Chongju area of central Korea was carried out from January to May 2001. It was found that in Koreathere were 9 cases of sand and duststorms (DS) and 16 associatedsignificant dustfall (SD) days. Observed maximum concentrations of DS and SD coming from NW China and Mongolia were in the rangeof 300–920 for TSP, 200–690 for PM₁₀ and 100–170 g m^-3 for PM_2.5.Satellite measurements clearly showed the formation and subsequent movement of DS to the Korean Peninsula and onward to the Korea East Sea, Japan and the Gulf of Alaska. According to satellite image analysis of dust clouds there were abundant coarse particles, measuring in size of 11.0 m. Medium-sizedparticles measuring in the range of 3.5–7.0 were also prevalent,while fine particles of less than 2.0 m were less distinctive in reflectivity. Measured values of PM_2.5 were alsorelatively low with SD events.The measured average pH values of dusty precipitation associatedwith DS were 7.24. Alkaline precipitation can play a `temporary'role in the neutralization of acidified soil until the subsequentevent of acidic rain. The new selection criteria of SD days from PM_2.5 at 85 g m^-3, PM₁₀ 190 g m^-3 and TSP 250 g m^-3 are recommended on dust pollution occurring from the invasion of a DS elsewhere.     

Responses of vegetation to a rise in water level at Kerkini Reservoir (1982–1991), a Ramsar site in northern Greece

The floodplain of the river Strymon at Kerkini (northern Greece) was transformed into an irrigation reservoir by the construction of a dam in 1932 and subsequently enlarged in 1982. The aims of this study were to quantify the changes occurring in the various habitat types following raising of the waterlevel and to assess the stability of the plant communities present at this Ramsar site. The current hydrological regime, which has been stable since 1986, is typified by an increase in mean annual reservoir level of 2.2 m and by an increase in the annual range in level of 1.3 m. Landsat (1980, 1981, 1984, 1986, and 1988) and SPOT (1990) satellite images show a decrease in the area of grassland and shallow water areas, the very rapid disappearance of reedbeds, the appearance of beds ofNymphaea and the disappearance of half the forest area. The flooded forest, dominated bySalix alba, is a key habitat contributing to the biological richness of this wetland of international importance. The decrease in the forested area will continue because of the death of standing trees, the absence of regeneration under the new regime, the felling of trees and grazing. Management could be undertaken to ensure the survival of forested habitat and reedbeds at Kerkini, but this would require that the authorities take into account nature conservation and the protected status of the site and not raise the water level again.     

Mapping Land-Cover and Mangrove Structures with Remote Sensing Techniques: A Contribution to a Synoptic GIS in Support of Coastal Management in North Brazil

This article deals with the development and application of a cartographic database for a synoptic Geographic Information System (GIS). Its purpose is the storage and evaluation of the heterogeneous datasets of the interdisciplinary scientific research program MADAM (Mangrove Dynamics and Management), which aims to develop recommendations for a tailored integrated coastal management scheme for the mangrove ecosystem at Bragança (North Brazil). The article describes the integration of remote sensing data, aerial photographs, as well as point data provided by fieldwork from different scientific fields. Using various innovative processing techniques and different scale-resolution levels, an assessment of temporal–spatial changes of the mangrove peninsula and the adjacent rural socioeconomic impact area, the type of mangrove structure, as well as a land-use cover analyses was undertaken. The definition of the spatial level of detail was found to be a major issue in the development of the GIS, as well as during the processing and analysis procedures. A division between strong and weak patterns in the mangrove ecosystem could be made, which implies different management measures and sets of specific interdisciplinary studies and monitoring at hierarchical scales.