Comparison of land cover maps using fuzzy agreement

A generic problem associated with different land cover maps that cover the same geographical area is the use of different legend categories. There may be disagreement in many areas when comparing different land cover products even though the legend shows the same or very similar land cover class. To capture the uncertainty associated with both differences in the legend and the difficulty in classification when comparing two land cover maps, expert knowledge and a fuzzy logic framework are used to map the fuzzy agreement. The methodology is illustrated by comparing the Global Land Cover 2000 data set and the MODIS global land cover product. Overall accuracy measures are calculated, and the spatial fuzzy agreement between the two land cover products is provided. This approach can be used to improve the overall confidence in a land cover product, since areas of severe disagreement can be highlighted, and areas can be identified that require further attention and possible re-mapping.     

Assessing the applicability of OpenStreetMap data to assist the validation of land use/land cover maps

The validation of land use/land cover (LULC) maps is usually performed using a reference database consisting of a sample of points or regions to which the ‘real’ class is assigned. This assignment is usually performed by specialists using photointerpretation (PI) of high-resolution imagery and/or field visits, which are time consuming and expensive processes. The aim of this article is to assess if the data available in the collaborative project OpenStreetMap (OSM) may be used as a source of data to assist the creation of these reference databases, reducing the time spent and costs associated with their generation. For this aim, two case studies were used, where the validation of the Global Monitoring for Environment and Security Urban Atlas (UA) was performed. The used methodology requires the harmonization of the data available in OSM with the UA nomenclature, and the subsequent creation of a LULC map from the OSM data. This map was then compared to UA to assess the similarity of the regions mapped in both. To test the usefulness of OSM data to assess the accuracy of UA, a sample of points was created and two reference databases generated, one assigning the data extracted automatically from OSM to the points where these data were available, and PI for the remaining points, and the other using only PI. The accuracy assessment of UA for the two case studies was then made building confusion matrixes and computing accuracy indicators. The results showed that for the two study areas, only low percentages of points had to be photo interpreted in the first reference database (respectively, 12% and 2% for the two study areas), decreasing the work load considerably. The results obtained with both reference databases are comparable for level 1 classes. For level 2 classes, worse results were obtained for some classes, showing that the OSM data used are not enough to create reliable reference data.     

Identifying strengths and limitations of pan-European forest cover maps through spatial comparison

Detailed and harmonized information on spatial forest distribution is an essential input for forest-related environmental assessments, in particular, for biomass and growing stock modeling. In the last years, several mapping approaches have been developed in order to provide such information for Europe in a harmonized way. Each of these maps exhibits particular properties and varies in accuracy. Yet, they are often used in parallel for different modeling purposes. A detailed spatial comparison seemed necessary in order to provide information on the advantages and limitations of each of these forest cover maps in order to facilitate their selection for modeling purposes.

This article confronts the high-resolution forest cover map recently developed by the Joint Research Centre for the year 2000 (FMAP2000) with previously existing maps for the same time period: the CORINE Land Cover 2000 (CLC2000) and the Calibrated European Forest Map 1996 (CEFM1996). The spatial comparison of these three maps was carried out based on forest proportion maps of 1 km derived from the original maps. To characterize differences according to biogeographic regions, two criteria were used: detail of thematic content within each map and local spatial agreement.

Concerning thematic content, CLC2000 displayed a surfeit of non-forested areas at the cost of low forest proportions, while FMAP2000 showed a more balanced distribution likely to preserve more detail in forest spatial pattern. Good spatial agreement was found for CLC2000 and FMAP2000 within about 70% of the study area, while only 50% agreement was found when compared with CEFM1996. The largest spatial differences between all maps were found in the Alpine and Mediterranean regions. Reasons for these might be different input data and classification techniques and, in particular, the calibration of CEFM1996 to reported national statistics.     

Spatiotemporal changes in agricultural land cover in Nepal over the last 100 years

In order to advance land use and land cover change (LUCC) research in Nepal, it is essential to reconstruct both the spatiotemporal distribution of agricultural land cover as well as scenarios that can explain these changes at the national and regional levels. Because of rapid population growth, the status of agricultural land in Nepal has changed markedly over the last 100 years. Historical data is used in this study, encompassing soils, populations, climatic variables, and topography. Data were revised to a series of 30 m grid cells utilized for agricultural land suitability and allocation models and were analyzed using a suite of advanced geographical tools. Our reconstructions for the spatiotemporal distribution of agricultural land in Nepal reveal an increasing trend between 1910 and 2010 (from 151.2 × 10² km² to 438.8 × 10² km²). This expanded rate of increase in agricultural land has varied between different eco, physiographic, and altitudinal regions of the country, significantly driven by population changes and policies over the period of this investigation. The historical dataset presented in this paper fills an existing gap in studies of agricultural land change and can be applied to other carbon cycle and climate modeling studies, as well as to impact assessments of agricultural land change in Nepal.     

Variable resolution block kriging using a hierarchical spatial data structure

Abstract

Kriging is an optimal method of spatial interpolation that produces an error for each interpolated value. Block kriging is a form of kriging that computes averaged estimates over blocks (areas or volumes) within the interpolation space. If this space is sampled sparsely, and divided into blocks of a constant size, a variable estimation error is obtained for each block, with blocks near to sample points having smaller errors than blocks farther away. An alternative strategy for sparsely sampled spaces is to vary the sizes of blocks in such away that a block's interpolated value is just sufficiently different from that of an adjacent block given the errors on both blocks. This has the advantage of increasing spatial resolution in many regions, and conversely reducing it in others where maintaining a constant size of block is unjustified (hence achieving data compression). Such a variable subdivision of space can be achieved by regular recursive decomposition using a hierarchical data structure. An implementation of this alternative strategy employing a split-and-merge algorithm operating on a hierarchical data structure is discussed. The technique is illustrated using an oceanographic example involving the interpolation of satellite sea surface temperature data. Consideration is given to the problem of error propagation when combining variable resolution interpolated fields in GIS modelling operations.     

Combining land cover products using a minimum divergence and a Bayesian data fusion approach

Land cover mapping plays an important role for a wide spectrum of applications that are ranging from climate modeling to food security. However, it is a common case that several and partially conflicting land cover products are available at the same time over a same area, where each product suffers from specific limitations and lack of accuracy. In order to take advantage of the best features of each product while at the same time attenuating their respective weaknesses, this paper is proposing a methodology that allows the user to combine these products together based on a general framework involving maximum entropy/minimum divergence principles, Bayesian data fusion and Bayesian updating. First, information brought by each land cover product is coded in terms of inequality constraints so that a first estimation of their quality can be computed based on a maximum entropy/minimum divergence principle. Information from these various land cover products can then be fused afterwards in a Bayesian framework, leading to a single map with an associated measure of uncertainty. Finally, it is shown how the additional information brought by control data can help improving this fused map through a Bayesian updating procedure. The first part of the paper is briefly presenting the most important theoretical results, while the second part is illustrating the use of this suggested approach for a specific area in Belgium, where five different land cover products are at hand. The benefits and limitations of this approach are finally discussed by the light of the results for this case study.     

The spatial local accuracy of land cover datasets over the Qiangtang Plateau,High Asia

We analyzed the spatial local accuracy of land cover (LC) datasets for the Qiangtang Plateau, High Asia, incorporating 923 field sampling points and seven LC compilations including the International Geosphere Biosphere Programme Data and Information System (IGBPDIS), Global Land cover mapping at 30 m resolution (GlobeLand30), MODIS Land Cover Type product (MCD12Q1), Climate Change Initiative Land Cover (CCI-LC), Global Land Cover 2000 (GLC2000), University of Maryland (UMD), and GlobCover 2009 (Glob-Cover). We initially compared resultant similarities and differences in both area and spatial patterns and analyzed inherent relationships with data sources. We then applied a geographically weighted regression (GWR) approach to predict local accuracy variation. The results of this study reveal that distinct differences, even inverse time series trends, in LC data between CCI-LC and MCD12Q1 were present between 2001 and 2015, with the exception of category areal discordance between the seven datasets. We also show a series of evident discrepancies amongst the LC datasets sampled here in terms of spatial patterns, that is, high spatial congruence is mainly seen in the homogeneous southeastern region of the study area while a low degree of spatial congruence is widely distributed across heterogeneous northwestern and northeastern regions. The overall combined spatial accuracy of the seven LC datasets considered here is less than 70%, and the GlobeLand30 and CCI-LC datasets exhibit higher local accuracy than their counterparts, yielding maximum overall accuracy (OA) values of 77.39% and 61.43%, respectively. Finally, 5.63% of this area is characterized by both high assessment and accuracy (HH) values, mainly located in central and eastern regions of the Qiangtang Plateau, while most low accuracy regions are found in northern, northeastern, and western regions.

     

Impact of cultivated land fragmentation on spatial heterogeneity of agricultural agglomeration in China

Systematically revealing the impact of cultivated land fragmentation(CLF) on the geographical agglomeration pattern of agricultural specialization(AS) has positive significance for national agricultural production management. Based on the data of the second national land survey and agricultural production, this study has explored the impact of CLF on spatial heterogeneity of agricultural agglomeration in China by comprehensively using the Theil index, ordinary least square model and geographically weighted regression. Results showed that:(1) the regional differentiation of the CLF in China is obvious, and the cultivated land fragmentation index is generally characterized by increasing pattern from northwest to southeast.(2) Spatially, the development level of AS in China has formed three high-value clusters in the Northeast China Plain, the Qinghai-Tibet Plateau, and the middle of the Middle-lower Yangtze Plain; and the low-value contiguous areas centered on the Yunnan-Guizhou Plateau and the Sichuan Basin and surrounding regions, with significant spatial differences. The contribution of grain crops, economic crops, and vegetables and melon to the level of AS was 74.63%, 9.09%, and 16.28%, respectively, and the pattern of agricultural geographical aggregation dominated by grain crops has primarily taken in shape.(3) CLF is significantly negatively correlated with AS, and every 1% increase in the degree of CLF will result in a decrease of about 0.2% in AS. However, the impact of CLF on the geographic agglomeration of different crop categories or groups varies significantly. Among them, CLF has a prominent impact on the specialization level of grain crops and vegetables and melon. Each 1% increase in the CLF will reduce the specialization level of grain crops by 0.38%, and increase the level of vegetables and melon by about 0.22%.(4) According to the landscape characteristics of cultivated land, the degree of spatial division and agglomeration of cultivated land patches have a significant impact on the formation of geographical agglomerationpattern of AS, and the intensity and direction of influence show significant regional differentiation, while the patch size has no significant impact.     

The potential of old maps and encyclopaedias for reconstructing historic European land cover/use change

Continental to global reconstructions of historic land cover/use are important inputs for many environmental, ecological and biogeochemical studies. While local to regional reconstructions frequently make use of old topographic maps and land use statistics, continental to global reconstructions are mostly model-based reconstructions. As a result they are subject to large uncertainties. A wealth of historic land cover/use maps and statistics have been produced and these are now more accessible due to the ending of copyrights and secrecy statuses, enthusiastic hobby communities and national cartographic institutes or cadastres that have a strategy towards data sharing with society. In this paper we made use of historic statistics and old topographic maps to demonstrate the added value for model-based reconstructions of historic land cover/use for Central Europe back to 1900. We harmonized these diverse data types and different types of historic land data were incorporated into the land use reconstructions. The added value of using these data was evaluated using historical maps by performing a reconstruction with and without the historic information. The accuracy of the land allocation in the historic reconstruction was improved by 16.5% using historic maps. Additionally, historic maps improved the representation of the spatial structure of landscapes. The historic land cover/use statistics used showed a strong agreement with independent estimates, like historic maps.     

Geographically neural network weighted regression for the accurate estimation of spatial non-stationarity

ABSTRACT

Geographically weighted regression (GWR) is a classic and widely used approach to model spatial non-stationarity. However, the approach makes no precise expressions of its weighting kernels and is insufficient to estimate complex geographical processes. To resolve these problems, we proposed a geographically neural network weighted regression (GNNWR) model that combines ordinary least squares (OLS) and neural networks to estimate spatial non-stationarity based on a concept similar to GWR. Specifically, we designed a spatially weighted neural network (SWNN) to represent the nonstationary weight matrix in GNNWR and developed two case studies to examine the effectiveness of GNNWR. The first case used simulated datasets, and the second case, environmental observations from the coastal areas of Zhejiang. The results showed that GNNWR achieved better fitting accuracy and more adequate prediction than OLS and GWR. In addition, GNNWR is applicable to addressing spatial non-stationarity in various domains with complex geographical processes.     

Simulating urban land use change by integrating a convolutional neural network with vector-based cellular automata

ABSTRACT

Vector-based cellular automata (VCA) models have been applied in land use change simulations at fine scales. However, the neighborhood effects of the driving factors are rarely considered in the exploration of the transition suitability of cells, leading to lower simulation accuracy. This study proposes a convolutional neural network (CNN)-VCA model that adopts the CNN to extract the high-level features of the driving factors within a neighborhood of an irregularly shaped cell and discover the relationships between multiple land use changes and driving factors at the neighborhood level. The proposed model was applied to simulate urban land use changes in Shenzhen, China. Compared with several VCA models using other machine learning methods, the proposed CNN-VCA model obtained the highest simulation accuracy (figure-of-merit = 0.361). The results indicated that the CNN-VCA model can effectively uncover the neighborhood effects of multiple driving factors on the developmental potential of land parcels and obtain more details on the morphological characteristics of land parcels. Moreover, the land use patterns of 2020 and 2025 under an ecological control strategy were simulated to provide decision support for urban planning.     

基于分布式感知深度神经网络的高分辨率PM2.5值估算

细颗粒物(PM2.5)个体暴露水平是健康效应研究中的关键问题,然而历史数据缺失和地面监测点覆盖范围小阻碍了相关研究.基于美国国家航空航天局遥感数据提供的气溶胶光学厚度(AOD),融合地面监测、气象等多源数据进行建模来估算近地面PM2.5浓度,所得结果的空间覆盖范围广、时间连续性强、方法成本低.本文基于2018年京津冀鲁...     

Object-oriented classification of very high resolution airborne imagery for the extraction of hedgerows and field margin cover in agricultural areas

The recent advancement in technology for the airborne collection and subsequent processing and analysis of digital remotely sensed data has been significant. An investigation into the contributions that recent developments have made to the assessment and extraction of hedgerow and field margins is presented. Research into the active legislation by which hedgerows and agricultural biodiversity in England are bound is brought together in the concept of an ideal classification. Using Leica Geosystems ADS40 data, collected for Berkshire (UK), the degree to which the ideal classification can be reached is investigated, focusing on the extraction of hedgerows and unploughed field margin protective boundary strips as image objects. Elevation data were derived from the ADS40 imagery data to accompany true-colour and colour-infrared spectral information. By buffering the detected hedgerow objects in a GIS, we facilitated an investigation of the presence of measures that satisfy the minimum hedgerow protection requirements stipulated by the UK Department for Environment, Food and Rural Affairs (DEFRA). In our study area, it was estimated that 68% of areas that need to be classified as ‘grassy hedgerow boundary strip’ to satisfy the minimum requirements of the legislation were established with low-lying grassy cover before the introduction of the legislation in July 2005. The implications of this research for the development of automated classification methods for field scale agricultural mapping are discussed.     

基于多时相陆地卫星图像的锡林河流域土地覆盖类型特征

1IntroductionThe growing concern over the impact of changes in land use and land cover on environmental conditions and the increasing human impact on the natural resources has captured worldwide attention of the political and scientific community (Ojima etal., 1991; Smith etal., 2001). It is expected that the changing land use/land cover pattern will be one of the driving forces of environmental changes superimposed on the natural changes at regional scale (Fu etal., 1993; Bonan, 1995). Arid…     

Identification of land cover/land use changes in the greater area of the Preveza peninsula in Greece using Landsat satellite data

This study investigates the Land Use & Land Cover (LULC) changes in a coastal area of the southwest part of Epirus region, called Preveza, situated in North-western Greece. Remote sensing imagery coming from the Enhanced Thematic Mapper (ETM+) sensor on board at the Landsat 7 satellite platform is used for this purpose. More specifically, we identified LULC changes in this environmentally sensitive coastal area, using Landsat image scenes for the dates of June 19th, 2000 and July 22nd, 2009. During this period, there was an increasing tourist activity and a high growth in the construction sector of the study area. The land-use changes were identified, examining several vegetation indices and band combinations, along with the implementation of different well-known classification techniques. The Normalized Difference Vegetation Index (NDVI) and the Brightness Index (BI) have proved to be the most suitable indices to successfully identify discrete land surface classes for this study area. Regarding the classifiers, a series of traditional and modern algorithms were tested. The Artificial Neural Networks (ANNs) and the Support Vector Machines (SVMs) gave improved results in comparison to other more traditional classification techniques. The best overall accuracy for the study area was achieved with the SVM classifier and reached 96.25% and 97.15% on the dates of June 19th, 2000 and July 22nd, 2009 respectively. The classification results depicted notable urbanization, small deforestation and important LULC changes in the agriculture sector, indicating a rapid coastal environment change in the region of interest.     

Characterization of land cover types in Xilin River Basin using multi-temporal Landsat images

This study conducted computer-aided image analysis of land use and land cover in Xilin River Basin, Inner Mongolia, using 4 sets of Landsat TM/ETM+ images acquired on July 31, 1987, August 11, 1991, September 27, 1997 and May 23, 2000, respectively. Primarily, 17 sub-class land cover types were recognized, including nine grassland types at community level: F. sibiricum steppe, S. baicalensis steppe, A. chinensis + forbs steppe, A. chinensis + bunchgrass steppe, A. chinensis + Ar. frigida steppe, S. grandis + A. chinensis steppe, S. grandis + bunchgrass steppe, S. krylavii steppe, Ar. frigida steppe and eight non-grassland types: active cropland, harvested cropland, urban area, wetland, desertified land, saline and alkaline land, cloud, water body + cloud shadow. To eliminate the classification error existing among different sub-types of the same gross type, the 17 sub-class land cover types were grouped into five gross types: meadow grassland, temperate grassland, desert grassland, cropland and non-grassland. The overall classification accuracy of the five land cover types was 81.0% for 1987, 81.7% for 1991, 80.1% for 1997 and 78.2% for 2000.     

最优子集神经网络高温预报模型

 利用2003-2007年6～9月ECMWF格点场资料,使用差分法、天气诊断、因子组合等方法构造出能反映本地天气动力学特征的预报因子库,采用press准则初选因子,尝试用最优子集方法进行神经网络夏季6～9月≥35℃高温预报模型的建模方法研究。2008年7月预报系统投入业务应用,检验证明所构造的神经网络高温预报模型具有更好的拟合和预报效果,为神经网络在灾害性天气预报的应用研究提供了新的思路和方法。     

Spatial features of land use/land cover change in the United States

With the classification data covering American land-use/land-cover (LUCC) with 30 m resolution from the project of National Land Cover Data (NLCD), we normalized them and made their resolution changed into 1 km ×1 km, created the data of American land-use grade and analyzed the spatial distribution and features of American LUCC as well as the influence of population and altitude on the land-use grade in light of methods of sampling analysis and correlation study. Based on the analysis, we concluded that forestry and grassland, accounting for 71.24% of the whole country, has taken the main part of American land cover, and besides, construction and arable land has occupied 19.22% of the total land, the rest of land cover types, including water area, wetland and underdeveloped land, is 9.54% of the country's total. The developing potential of American land resources is enormous with less destroyed and disturbed ecological environment. Although, in some sense, the population and altitude influence the spatial variation of American land-use grade respectively, the influence of spatial variation of altitude and population density on that of land-use grade is not significanct.     

美国土地利用/土地覆盖变化的空间特征

1IntroductionInrecentyears,studyonglobalchangehasbecomethefocus,onwhichstresshasbeenlaidbytodayhumansociety.Inordertodealwithseriesofenvironmentalproblemssuchasdegradationoflandresources,globalwarming,destructionofozonosphereandsoon,andassurethestabilizationandsustainabledevelopmentofworldeconomy,IGBPhasestablishedthestudyframeforglobalchange.Withavastrangeofstudy,researchonglobalchangehasbroughtforwardagreatdealofimportantscientificquestionsrelatedtofeasibilityforhumanelivingoftheEarth.Inth…     

Classification of mineral deposits into types using mineralogy with a probabilistic neural network

In order to determine whether it is desirable to quantify mineral-deposit models further, a test of the ability of a probabilistic neural network to classify deposits into types based on mineralogy was conducted. Presence or absence of ore and alteration mineralogy in well-typed deposits were used to train the network. To reduce the number of minerals considered, the analyzed data were restricted to minerals present in at least 20% of at least one deposit type. An advantage of this restriction is that single or rare occurrences of minerals did not dominate the results. Probabilistic neural networks can provide mathematically sound confidence measures based on Bayes theorem and are relatively insensitive to outliers. Founded on Parzen density estimation, they require no assumptions about distributions of random variables used for classification, even handling multimodal distributions. They train quickly and work as well as, or better than, multiple-layer feedforward networks. Tests were performed with a probabilistic neural network employing a Gaussian kernel and separate sigma weights for each class and each variable. The training set was reduced to the presence or absence of 58 reported minerals in eight deposit types. The training set included: 49 Cyprus massive sulfide deposits; 200 kuroko massive sulfide deposits; 59 Comstock epithermal vein gold districts; 17 quartzalunite epithermal gold deposits; 25 Creede epithermal gold deposits; 28 sedimentary-exhalative zinc-lead deposits; 28 Sado epithermal vein gold deposits; and 100 porphyry copper deposits. The most common training problem was the error of classifying about 27% of Cyprus-type deposits in the training set as kuroko. In independent tests with deposits not used in the training set, 88% of 224 kuroko massive sulfide deposits were classed correctly, 92% of 25 porphyry copper deposits, 78% of 9 Comstock epithermal gold-silver districts, and 83% of six quartzalunite epithermal gold deposits were classed correctly. Across all deposit types, 88% of deposits in the validation dataset were correctly classed. Misclassifications were most common if a deposit was characterized by only a few minerals, e.g., pyrite, chalcopyrite,and sphalerite. The success rate jumped to 98% correctly classed deposits when just two rock types were added. Such a high success rate of the probabilistic neural network suggests that not only should this preliminary test be expanded to include other deposit types, but that other deposit features should be added