Basic theory and key technologies for marine geographic information system

Marine geographic information system （MGIS） has great ability to deal with the spatio-temporal problems and has potential superiority when it is applied to oceanography. Using the feature extraction of oceanic phenomena as a case study, the functions of the MGIS are analyzed, and thus the position of MGIS in the oceanography is defined. Comparing the requirement of MGIS with that of the traditional GIS which has been developed in the terrestrial applications in the past four decades, the frame for the functions of MGIS is constructed. According to the established MGIS, some key technologies are discussed in detail with emphasis on the specialities which can distinguish the MGIS from the traditional GIS.

Effects of Ulva prolifera dissipation on the offshore environment based on remote sensing images and field monitoring data

Outbreaks of Ulva prolifera have continued in the South Yellow Sea of China(SYS)since 2007,becoming a serious marine ecological disaster.Large amounts of U.prolifera drift to the coast of the Shandong Peninsula to dissipate under the action of southeast monsoons and ocean surface currents.This causes serious harm to the ecological environment and economic activities of coastal cities.To investigate the impact of U.prolifera dissipation,this study extracted the spatiotemporal distribution of U.prolifera in the SYS from 2012 to 2020 based on the Google Earth Engine.The outbreak cycle of U.prolifera was determined by fitting analysis of outbreak time and coverage area through MATLAB.This study also looked at the effect of U.prolifera dissipation on water quality through field monitoring data.The results showed that the growth curve of the U.prolifera has a significant Gaussian distribution.The U.prolifera dissipates in Haiyang,China,in July and August every year and affects the offshore environment.Water quality parameters of seawater at different depths had significant differences after the U.prolifera dissipation.Changes in pH,chemical oxygen demand,nitrite nitrogen,nitrate nitrogen,ammonia nitrogen,chlorophyll a,total phosphorus,and suspended solids were more significant in surface seawater than in deeper water.Changes in the concentrations of dissolved oxygen and total nitrogen were more significant in the deep seawater(1.63 and 1.1 times higher than those in the surface seawater,respectively).The dissipation of U.prolifera releases a large amount of carbon and nitrogen into the seawater,which provides rich nutrients for phytoplankton and may cause secondary disasters such as red tide.These findings are useful for further understanding the rules of U.prolifera dissipation,as well as preventing and controlling green tide disasters.

Soil Quality Assessment Using Weighted Fuzzy Association Rules

Fuzzy association rules (FARs) can be powerful in assessing regional soil quality, a critical step prior to land planning and utilization; however, traditional FARs mined from soil quality database, ignoring the importance variability of the rules, can be redundant and far from optimal. In this study, we developed a method applying different weights to traditional FARs to improve accuracy of soil quality assessment. After the FARs for soil quality assessment were mined, redundant rules were eliminated according to whether the rules were significant or not in reducing the complexity of the soil quality assessment models and in improving the comprehensibility of FARs. The global weights, each representing the importance of a FAR in soil quality assessment, were then introduced and refined using a gradient descent optimization method. This method was applied to the assessment of soil resources conditions in Guangdong Province, China. The new approach had an accuracy of 87%, when 15 rules were mined, as compared with 76% from the traditional approach. The accuracy increased to 96% when 32 rules were mined, in contrast to 88% from the traditional approach. These results demonstrated an improved comprehensibility of FARs and a high accuracy of the proposed method.

A Review of Researches on Deep Learning in Remote Sensing Application

In recent years, deep learning has been widely used in the field of image understanding and made breakthroughs research progress in image understanding. Because remote sensing application and image understanding are inseparable, researchers have carried out a lot of research on the application of deep learning in remote sensing field, and extended the deep learning method to various application fields of remote sensing. This paper summarizes the basic principles of deep learning and its research progress and typical applications in remote sensing, introduces the current main deep learning model and its development history, focuses on the analysis and elaboration of the research status of deep learning in remote sensing image classification, object detection and change detection, and on this basis, summarizes the typical applications and their application effects. Finally, according to the current application of deep learning in remote sensing, the main problems and future development directions are summarized.

北京市区绿色空间滞尘分布反演研究

北京市作为中国政治中心和京津冀特大型城市,近40年来城市化进程迅速,大气颗粒物和尘埃粒子污染问题较为突出,发挥绿色空间滞尘功能具有重要现实意义。将高光谱技术与遥感技术相结合,反演了市域尺度绿色空间滞尘分布。以北京市区绿色空间常见植被大叶黄杨(Euonymus japonicus)为研究对象,通过室外采样与室内试验获取叶片样本的滞尘量、光谱反射率和叶面积等数据,比较叶片滞尘前后的原始光谱曲线和反射率一阶导数特征,分析不同滞尘量对光谱反射率的影响,探究对叶片滞尘量高度敏感的波段。利用光谱响应函数将地面采集的窄波段光谱反射率数据分别转化遥感卫星的宽波段光谱反射率数据,建立对应卫星波段植被指数比值与滞尘量的回归模型,选取拟合效果最好的回归模型作为滞尘反演模型。结合GF-2影像所提取北京市区绿色空间范围,采用滞尘反演模型获取北京市区绿色空间滞尘分布。进而插值得到北京市区尘埃污染分布,采用空间自相关模型检验其空间聚集特性。结果表明:740~1870 nm波段,滞尘后光谱反射率明显低于滞尘前反射率,滞尘对红边、黄边、蓝边位置没有明显影响,对“红边幅值”和“红边面积”影响较明显,利用Sentinel-2影像计算的EVI指数与滞尘量相关性最高,所构建的线性与二次滞尘回归模型决定系数(R^(2))分别为0.705和0.751;利用2021年4月7日和2021年6月3日的Sentinel-2影像反演获取了北京市区绿色空间滞尘分布,其滞尘分布趋势表现为:市中心高于郊区,北部高于南部,东部高于西部。北京市区中部、北部和东部易产生尘埃污染。污染分布具有明显聚集性,并非完全随机。

A model supported by GIS for locating and quantifying PM_(2.5) emission originated from residential wood burning

A research method was presented for spatially quantifying and allocating the potential activity of a fine particle matter emission （ PM2.5 ）, which originated from residential wood burning （RWB） in this study. Demographic, hypsographic, climatic and topographic data were compiled and processed within a geographic information system（GIS）, and as independent variables put into a linear regression model for describing spatial distribution of the potential activity of residential wood burning as primary heating source. In order to improve the estimation, the classifications of urban, suburban and rural were redefined to meet the specifications of this application. Also, several definitions of forest accessibility were tested for estimation. The results suggested that the potential activity of RWB was mostly determined by elevation of a location, forest accessibility, urban/non-urban position, climatic conditions and several demographic variables. The linear regression model could explain approximately 86% of the variation of surveyed potential activity of RWB. The analysis results were validated by employing survey data collected mainly from a WebGIS based phone interview over the study area in central California. Based on lots free public GIS data, the model provided an easy and ideal tool for geographic researchers, environmental planners and administrators to understand where and how much PM2.5 emission from RWB was contributed to air quality. With this knowledge they could identify regions of concern, and better plan mitigation strategies to improve air quality. Furthermore, it allows for future adjustment on some parameters as the spatial analysis method is implemented in the different regions or various eco-social models.

GIS SUPPORTED HEDONIC MODEL FOR ASSESSING PROPERTY VALUE IN WEST OAKLAND, CALIFORNIA

A hedonic linear regression model is constructed in this paper to estimate property value. In our model, the property value (sales price) is a function of several selected variables such as the property characteristics, social neighborhoods, level of neighborhood environmental contaminations, level of neighborhood crimes, and locational accessibility to jobs or services. Definitions and calculation of these variables are approached by using Geographic Information System tools. For improving estimation, gravity model is employed to measure both levels of neighborhood toxic sites and crimes; and a time-based method is used to measure the locational accessibility rather than simple straight-line distance measurement. This study discovers that the relationship between house value and its nearby highway is nonlinear. The methodology could help policy makers assess the external effects of a property. Our model also could be used potentially to identify the current and historic trends of development caused by neighborhood or environments change in the study area.

Estimation and Characteristic Analysis of Biomass within the Haihe River Basin Based on CASA Model

Using CASA model, biomass within the Haihe River basin during 2002 -2007 was estimated based on remote sensing images, corresponding data of temperature, precipitation and solar radiation, and 1：400 000 0 maps of vegetation coverage in China. Variations in the biomass with vegetation type and vegetation coverage in 2007 were analyzed. Meanwhile, its temporal and spatial changes were discussed. The results validate the applicability of CASA model in the estimation of biomass within the Haihe River basin. During the past 6 years, annual average biomass within the basin was 405.5 Tg in total; annual average biomass in the basin was high in the southeast but low in the northwest, namely plains 〉 mountains 〉 plateaus.

Value Assessment of Wetland Ecosystem Services in the Da Hinggan Mountains, China

This study examined regional differences in ecosystem services for the Da Hinggan Mountains(DHM),China.A correction index was constructed based on ten-year average net primary productivity(NPP)data.A new equivalent factor table that was suitable for the assessment of wetlands in the DHM was formed by using the expert weight determination method(EWDM).An evaluation model was established for evaluating the ecosystem service value(ESV)of wetlands in the DHM.The results show that in 2020,the total ESV of wetlands reached 93361×10^(6) USD,with the forest swamp and marsh ecosystems contributing the most.From the perspective of value composition,regulating services and supporting services are the main service functions of wetlands in the DHM.From 2010 to2020,ESV provided by wetlands increased by 4337×10^(6) USD/yr in the DHM.The value of forest swamp and peatland ecosystems increased by 18.6%and 12.7%,respectively,whereas the value of swamp,shrub swamp,and marsh decreased.The research results are of significance for contributing to local government performance evaluation and determining financial compensation for the provision of wetland ecosystem services.

Mapping regional forest management units:a road-based framework in Southeastern Coastal Plain and Piedmont

Management practices are one of the most important factors affecting forest structure and function.Landowners in southern United States manage forests using appropriately sized areas,to meet management objectives that include economic return,sustainability,and esthetic enjoyment.Road networks spatially designate the socioenvironmental elements for the forests,which represented and aggregated as forest management units.Road networks are widely used for managing forests by setting logging roads and firebreaks.We propose that common types of forest management are practiced in road-delineated units that can be determined by remote sensing satellite imagery coupled with crowd-sourced road network datasets.Satellite sensors do not always capture roadcaused canopy openings,so it is difficult to delineate ecologically relevant units based only on satellite data.By integrating citizen-based road networks with the National Land Cover Database,we mapped road-delineated management units across the regional landscape and analyzed the size frequency distribution of management units.We found the road-delineated units smaller than 0.5 ha comprised 64%of the number of units,but only0.98%of the total forest area.We also applied a statistical similarity test(Warren's Index)to access the equivalency of road-delineated units with forest disturbances by simulating a serious of neutral landscapes.The outputs showed that the whole southeastern U.S.has the probability of road-delineated unit of 0.44 and production forests overlapped significantly with disturbance areas with an average probability of 0.50.

Forest Fragmentation and Its Potential Implications in the Brazilian Amazon between 2001 and 2010

In recent decades, human development pressures have results in conversions of vast tracts of Amazonian tropical rain forests to agriculture and other human land uses. In addition to the loss of large forest cover, remaining forests are also fragmented into smaller habitats. Fragmented forests suffer several biological and ecological changes due to edge effects that can exacerbate regional forest degradation. The Brazilian Amazon has had greatly contrasting land cover dynamics in the past decade with the highest historical rates of deforestation (2001-2005) followed by the lowest rates of forest loss in decades, since 2006. Currently, the basin-wide status and implications of forest fragmentation on remnant forests is not well known. We performed a regional forest fragmentation analysis for seven states of the Brazilian Amazon between 2001 and 2010 using a recent deforestation data. During this period, the number of forest fragments (>2 ha) doubled, nearly 125,000 fragments were formed by human activities with more than 50% being smaller than 10 ha. Over the decade, forest edges increased by an average of 36,335 km/year. However, the rate was much greater from 2001-2005 (50,046 km/year) then 2006-2010 (25,365 km/year) when deforestation rates dropped drastically. In 2010, 55% of basin-wide forest edges were < 10 years old due to the creation of large number of small fragments where intensive biological and ecological degradation is ongoing. Over the past decade protected areas have been expanded dramatically over the Brazilian Amazon and, as of 2010, 51% of remaining forests across the basin are within protected areas and only 1.5% of protected areas has been deforested. Conversely, intensive forest cover conversion has been occurred in unprotected forests. While 17% of Amazonian forests are within 1 km of forest edges in 2010, the proportion increases to 34% in unprotected areas varying between 14% and 95% among the studied states. Our results indicate that the Brazilian Amazon now largely consists of two contrasting forest conditions: protected areas with vast undisturbed forests and unprotected forests that are highly fragmented and disturbed landscapes.

Changes in global potential vegetation distributions from 1911 to 2000 as simulated by the Comprehensive Sequential Classification System approach

Vegetation classification models play an important role in studying the response of the terrestrial ecosystem to global climate change. In this paper, we study changes in global Potential Natural Vegetation (PNV) distributions using the Comprehensive Sequential Classification System (CSCS) approach, a technique that combines geographic information systems. Results indicate that on a global scale there are good agreements among maps produced by the CSCS method and the globally well-accepted Holdridge Life Zone (HLZ) and BIOME4 PNV models. The potential vegetation simulated by the CSCS approach has 6 major latitudinal zones in the northern hemisphere and 2 in the southern hemisphere. In mountainous areas it has obvious altitudinal distribution characteristics due to topographic effects. The distribution extent for different PNV classes at various periods has different characteristics. It had a decreasing trend for the tundra and alpine steppe, desert, sub-tropical forest and tropical forest categories, and an increasing trend for the temperate forest and grassland vegetation categories. The simulation of global CSCS-based PNV classes helps to understand climate-vegetation relationships and reveals the dynamics of potential vegetation distributions induced by global changes. Compared with existing statistical and equilibrium models, the CSCS approach provides similar mapping results for global PNV and has the advantage of improved simulation of grassland classes.

Combined analysis of land cover change and NDVI trends in the Northern Eurasian grain belt

We present an approach to regional environ- mental monitoring in the Northern Eurasian grain belt combining time series analysis of MODIS normalized difference vegetation index （NDVI） data over the period 2001-2008 and land cover change （LCC） analysis of the 2001 and 2008 MODIS Global Land Cover product （MCD 12Q 1）. NDVI trends were overwhelmingly negative across the grain belt with statistically significant （p ≤0.05） positive trends covering only 1% of the land surface. LCC was dominated by transitions between three classes; cropland, grassland, and a mixed cropland/natural vegeta- tion mosaic. Combining our analyses of NDVI trends and LCC, we found a pattern of agricultural abandonment （cropland to grassland） in the southern range of the grain belt coinciding with statistically significant （p≤0.05） negative NDVI trends and likely driven by regional drought. In the northern range of the grain belt we found an opposite tendency toward agricultural intensification; in this case, represented by LCC from cropland mosaic to pure cropland, and also associated with statistically significant （p≤0.05） negative NDVI trends. Relatively small clusters of statistically significant （p ≤ 0.05） positive NDVI trends corresponding with both localized land abandonment and localized agricultural intensification show that land use decision making is not uniform across the region. Land surface change in the Northern Eurasian grain belt is part of a larger pattern of land cover land use change （LCLUC） in Eastern Europe, Russia, and former territories of the Soviet Union following realignment of socialist land tenure and agricultural markets. Here, we show that a combined analysis of LCC and NDVI trends provides a more complete picture of the complex- ities of LCLUC in the Northern Eurasian grain belt,involving both broader climatic forcing, and narrower anthropogenic impacts, than might be obtained from either analysis alone.

Grain production trends in Russia, Ukraine and Kazakhstan： New opportunities in an increasingly unstable world？

Grain production in the countries of the former USSR sharply declined during the past two decades and has only recently started to recover. In the context of the current economic and food-price crisis, Russia, Ukraine, and Kazakhstan might be presented with a window of opportunity to reemerge on the global agricultural market, if they succeed in increasing their productivity. The future of their agriculture, however, is highly sensitive to a combination of internal and external factors, such as institutional changes, land-use changes, climate variability and change, and global economic trends. The future of this region＇s grain production is likely to have a significant impact on the global and regional food security over the next decades.

The global Landsat imagery database for the FAO FRA remote sensing survey

To collect and provide periodically updated information on global forest resources,their management and use,the United Nations Food and Agriculture Organization(FAO)has been coordinating global forest resources assessments(FRA)every 5
10 years since 1946.To complement the FRA national-based statistics and to provide an independent assessment of forest cover and change,a global remote sensing survey(RSS)has been organized as part of FAO FRA 2010.In support of the FAO RSS,an image data set appropriate for global analysis of forest extent and change has been produced.Landsat data from the Global Land Survey 1990
2005 were systematically sampled at each longitude and latitude intersection for all points on land.To provide a consistent data source,an operational algorithm for Landsat data pre-processing,normalization,and cloud detection was created and implemented.In this paper,we present an overview of the data processing,characteristics,and validation of the FRA RSS Landsat dataset.The FRA RSS Landsat dataset was evaluated to assess overall quality and quantify potential limitations.

Rapid flood inundation mapping by differencing water indices from pre-and post-flood Landsat images

Following flooding disasters,satellite images provide valuable information required for generating flood inundation maps.Multispectral or optical imagery can be used for generating flood maps when the inundated areas are not covered by clouds.We propose a rapid mapping method for identifying inundated areas based on the increase in the water index value between the pre-and post-flood satellite images.Values of the Normalized Difference Water Index(NDWI)and Modified NDWI(MNDWI)will be higher in the post-flood image for flooded areas compared to the pre-flood image.Based on a threshold value,pixels corresponding to the flooded areas can be separated from non-flooded areas.Inundation maps derived from differencing MNDWI values accurately captured the flooded areas.However the output image will be influenced by the choice of the pre-flood image,hence analysts have to avoid selecting pre-flood images acquired in drought or earlier flood years.Also the inundation maps generated using this method have to be overlaid on the post-flood satellite image in order to orient personnel to landscape features.Advantages of the proposed technique are that flood impacted areas can be identified rapidly,and that the pre-existing water bodies can be excluded from the inundation maps.Using pairs of other satellite data,several maps can be generated within a single flood which would enable emergency response agencies to focus on newly flooded areas.

SAR image water extraction using the attention U-net and multi-scale level set method:flood monitoring in South China in 2020 as a test case

Level set method has been extensively used for image segmentation,which is a key technology of water extraction.However,one of the problems of the level-set method is how to find the appropriate initial surface parameters,which will affect the accuracy and speed of level set evolution.Recently,the semantic segmentation based on deep learning has opened the exciting research possibilities.In addition,the Convolutional Neural Network(CNN)has shown a strong feature representation capability.Therefore,in this paper,the CNN method is used to obtain the initial SAR image segmentation map to provide deep a priori information for the zero-level set curve,which only needs to describe the general outline of the water body,rather than the accurate edges.Compared with the traditional circular and rectangular zero-level set initialization method,this method can converge to the edge of the water body faster and more precisely;it will not fall into the local minimum value and be able to obtain accurate segmentation results.The effectiveness of the proposed method is demonstrated by the experimental results of flood disaster monitoring in South China in 2020.