排序方式： 共有34条查询结果，搜索用时 312 毫秒
1.
In recent years,landslide susceptibility mapping has substantially improved with advances in machine learning.However,there are still challenges remain in landslide mapping due to the availability of limited inventory data.In this paper,a novel method that improves the performance of machine learning techniques is presented.The proposed method creates synthetic inventory data using Generative Adversarial Networks(GANs)for improving the prediction of landslides.In this research,landslide inventory data of 156 landslide locations were identified in Cameron Highlands,Malaysia,taken from previous projects the authors worked on.Elevation,slope,aspect,plan curvature,profile curvature,total curvature,lithology,land use and land cover(LULC),distance to the road,distance to the river,stream power index(SPI),sediment transport index(STI),terrain roughness index(TRI),topographic wetness index(TWI)and vegetation density are geo-environmental factors considered in this study based on suggestions from previous works on Cameron Highlands.To show the capability of GANs in improving landslide prediction models,this study tests the proposed GAN model with benchmark models namely Artificial Neural Network(ANN),Support Vector Machine(SVM),Decision Trees(DT),Random Forest(RF)and Bagging ensemble models with ANN and SVM models.These models were validated using the area under the receiver operating characteristic curve(AUROC).The DT,RF,SVM,ANN and Bagging ensemble could achieve the AUROC values of(0.90,0.94,0.86,0.69 and 0.82)for the training;and the AUROC of(0.76,0.81,0.85,0.72 and 0.75)for the test,subsequently.When using additional samples,the same models achieved the AUROC values of(0.92,0.94,0.88,0.75 and 0.84)for the training and(0.78,0.82,0.82,0.78 and 0.80)for the test,respectively.Using the additional samples improved the test accuracy of all the models except SVM.As a result,in data-scarce environments,this research showed that utilizing GANs to generate supplementary samples is promising because it can improve the predictive capability of common landslide prediction models. 相似文献
2.
Within the frame of the ESCOMPTE program, a spatial emission inventory and an emission database aimed at tropospheric photochemistry intercomparison modeling has been developed under the scientific supervision of the LPCA with the help of the regional coordination of Air Quality network AIRMARAIX. This inventory has been established for all categories of sources (stationary, mobile and biogenic sources) over a domain of 19,600 km2 centered on the cities of Marseilles–Aix-en-Provence in the southeastern part of France with a spatial resolution of 1 km2. A yearly inventory for 1999 has been established, and hourly emission inventories for 23 days of June and July 2000 and 2001, corresponding to the intensive measurement periods, have been produced. The 104 chemical species in the inventory have been selected to be relevant with respect to photochemistry modeling according to available data. The entire list of species in the inventory numbers 216 which will allow other future applications of this database. This database is presently the most detailed and complete regional emission database in France. In addition, the database structure and the emission calculation modules have been designed to ensure a better sustainability and upgradeability, being provided with appropriate maintenance software. The general organization and method is summarized and the results obtained for both yearly and hourly emissions are detailed and discussed. Some comparisons have been performed with the existing results in this region to ensure the congruency of the results. This leads to confirm the relevance and the consistency of the ESCOMPTE emission inventory. 相似文献
3.
B. H. Luckman 《地球表面变化过程与地形》1976,1(3):287-298
The major controls of rockfall activity are divided into two interacting groups. (a)Climatic factors which, through their control of temperatures and the availability and state of water, are primary controls of rockfall trigger mechanisms. (b) Geologic factors which, via cliff form (plan, profile, dissection etc.) and the character and availability of materials, influence the type, spatial distribution and intensity of rockfall activity. Detailed examination of these controls suggest that both seasonal and daily patterns of rockfall activity can vary markedly over a very small area. The implications of this variability for the design and interpretation of rockfall inventories are discussed using sample data from Surprise Valley, Jasper National Park, for the period from May-October 1969. On a seasonal basis rockfall activity showed a major spring peak with secondary maxima in the fall and associated with major summer storms. Two sites of differing aspect and morphology were studied in more detail revealing marked differences in the mean hourly frequency (0·94:0·20), mean hourly probability (0·40:0·16, hours with rockfall/hours observed) and daily pattern of rockfall activity. The west facing site showed equal hourly probability of rockfalls from 1000 to 2000 hr whereas the east facing slope had a greater hourly probability when it was in the sun (1100 1400hr) than in the shade (1400-1900hr). The differences in rockfall frequency and probabilities reflect the physical characteristics of the individual sites whereas the daily pattern of rockfalls is related to microclimate. These results suggest that if future inventory studies are to make a significant contribution, they must be carefully designed field experiments in which. (i) Data are restricted to the study or comparison of single continuously observed sites. (ii) Study sites and/or data periods are carefully selected to isolate and investigate specific controls e.g. the influence of cliff form, aspect, periods of rainfall etc. on rockfall patterns. (iii) Good on-site microclimate data are available. Also comparison of frequency measures based on the arithmetic mean should be avoided since the distribution of rockfalls per hour closely follows a Poisson distribution with occasional high values which unduly influence the value of the arithmetic mean. It is suggested that rockfall probabilities, based on the binary decision of whether or not a rockfall occurs in a given period, are a more useful measure for daily patterns of rockfall activity. 相似文献
4.
Wild mushrooms are recognized as important non-wood forest products in mountainous ecosystems, but their real potential for generating rural economies has not been fully evaluated due to the difficulties in obtaining reliable productivity data, minimizing their true potential as contributor to rural economies. Mushroom yield models based on large data series from Pinus forest ecosystems in the region of Catalonia(Spain), combined with data from the Spanish National Forest Inventory allow us to estimate the potential mushroom productivity by forest ecosystems. The results of 24,500 tons/yr of mushrooms of which 16,300 tons are classified as edible and 7,900 tons are commonly marketed demonstrate the importance of mushroom productions in Catalonian pine forests, mostly located in mountainous areas where the development of agricultural activities is limited. Economic mushroom value is estimated at 48 million € for the edible mushroom and 32 million € for those corresponding to marketable yields, confirming the potential of this non-wood forest product. These production results and corresponding economic values provide a basis for the incorporation of wild mushrooms as significant non-wood forest products in the development of forest policies in mountainous areas. 相似文献
5.
Ling Wang Mengyuan Zhang Yinghua Lou Runhui Ke Minggang Zheng 《Marine pollution bulletin》2017,114(1):577-582
The levels and distribution of tris-(2,3-dibromopropyl) isocyanurate (TBC) and hexabromocyclododecanes (HBCDs) of surface sediments in the Yellow River Delta wetland had been investigated. The concentrations of TBC and ∑ HBCDs ranged from 0.20 to 29.03 ng·g? 1 dw and below limits of detections (LODs) to 20.25 ng·g? 1 dw. The average composition profile of three HBCDs isomers were 10.1%, 6.1% and 83.8% for α-, β- and γ-HBCD, respectively. Moreover, correlation analysis indicated there are similar sources among three isomers and positive correlations between total organic carbon (TOC) content and concentrations of TBC and HBCDs. The mass inventory of TBC,α-, β-, γ-HBCD, ΣHBCDs in surface sediments of Yellow River Delta wetland were estimated about 725.50, 72.76, 44.29, 548.34, 665.39 kg. Therefore, further investigations on potential human health and environmental risk assessments of TBC and HBCDs were needed. 相似文献
6.
1968-2009年叶尔羌河流域冰川变化——基于第一、二次中国冰川编目数据 总被引：6，自引：1，他引：5
利用"中国冰川资源及其变化调查"项目最新冰川编目成果和中国第一次冰川编目结果, 对中国叶尔羌河流域1968-2009年冰川变化进行了分析. 结果表明:叶尔羌河流域冰川总体上处于退缩状态, 面积减少了927 km2, 年平均面积减少23.2 km2, 年均面积缩小比例为0.36%·a-1, 与中国其他地区冰川退缩程度相比属于中等水平. 叶尔羌河流域不同规模冰川的退缩幅度存在差异, 小冰川大幅萎缩, 甚至消失; 规模较大的冰川相对变化幅度较小, 一些冰川出现过跃动. 从朝向分布来看, 位于南坡的冰川退缩最为严重, 而西坡较小. 冰川集中分布在海拔5 100~5 500 m和5 500~5 900 m区间, 海拔4 700~5 100 m区间的冰川面积减少最为显著. 消失冰川大多数为面积在0.2~0.5 km2的小冰川, 且朝向东北坡的冰川消失数量最多. 研究区有冰川分裂现象, 也出现了支冰川前进超覆现象, 统计表明该流域有13条冰川在前进后形成6条冰川. 1968-2009年研究区气温升高、降水增加, 总体上看, 降水增加缓解了因升温而导致的冰川退缩. 相似文献
7.
Measuring tree stem diameters using intensity profiles from ground-based scanning lidar from a fixed viewpoint 总被引：1，自引：0，他引：1
This paper presents a method for using the intensity of returns from a scanning light detection and ranging (lidar) system from a single viewing point to identify the location and measure the diameter of tree stems within a forest. Such instruments are being used for rapid forest inventory and to provide consistent supporting information for airborne lidars. The intensity transect across a tree stem is found to be consistent with a simple model parameterised by the range and diameter of the trunk. The stem diameter is calculated by fitting the model to transect data. The angular span of the stem can also be estimated by using a simple threshold where intensity values are tested against the expected intensity for a stem of given diameter. This is useful when data are insufficient to fit the model or the stem is partially obscured. The process of identifying tree positions and trunk diameters is fully automated and is shown to be successful in identifying a high proportion of trees, including some that are partially obscured from view. The range and bearing to trees are in excellent agreement with field data. Trunk angular span and diameter estimations are well correlated with field measurements at the plot scale. The accuracy of diameter estimation is found to decrease with range from the scanning position and is also reduced for stems subtending small angles (less than twice the scanning resolution) to the instrument. A method for adjusting survey results to compensate for trees missed due to obscuration from the scanning point and the use of angle count methods is found to improve basal area estimates and achieve agreement within 4% of field measurements. 相似文献
8.
Spatial predictions of forest variables are required for supporting modern national and sub-national forest planning strategies, especially in the framework of a climate change scenario. Nowadays methods for constructing wall-to-wall maps and calculating small-area estimates of forest parameters are becoming essential components of most advanced National Forest Inventory (NFI) programs. Such methods are based on the assumption of a relationship between the forest variables and predictor variables that are available for the entire forest area. Many commonly used predictors are based on data obtained from active or passive remote sensing technologies. Italy has almost 40% of its land area covered by forests. Because of the great diversity of Italian forests with respect to composition, structure and management and underlying climatic, morphological and soil conditions, a relevant question is whether methods successfully used in less complex temperate and boreal forests may be applied successfully at country level in Italy.For a study area of more than 48,657 km2 in central Italy of which 43% is covered by forest, the study presents the results of a test regarding wall-to-wall, spatially explicit estimation of forest growing stock volume (GSV) based on field measurement of 1350 plots during the last Italian NFI. For the same area, we used potential predictor variables that are available across the whole of Italy: cloud-free mosaics of multispectral optical satellite imagery (Landsat 5 TM), microwave sensor data (JAXA PALSAR), a canopy height model (CHM) from satellite LiDAR, and auxiliary variables from climate, temperature and precipitation maps, soil maps, and a digital terrain model.Two non-parametric (random forests and k-NN) and two parametric (multiple linear regression and geographically weighted regression) prediction methods were tested to produce wall-to-wall map of growing stock volume at 23-m resolution. Pixel level predictions were used to produce small-area, province-level model-assisted estimates. The performances of all the methods were compared in terms of percent root mean-square error using a leave-one-out procedure and an independent dataset was used for validation. Results were comparable to those available for other ecological regions using similar predictors, but random forests produced the most accurate results with a pixel level R2 = 0.69 and RMSE% = 37.2% against the independent validation dataset. Model-assisted estimates were more precise than the original design-based estimates provided by the NFI. 相似文献
9.
Roger K. Wiebusch 《社会与自然资源》2017,30(1):16-30
While conversion of wetlands to nonagricultural purposes has persisted, the five U.S. national wetland inventory reports spanning 1955–2009 show a large decline in the rate of conversion of wetlands for agriculture, especially from the mid-1970s to the mid-1990s. Through regression and path analysis, this study identifies primary policy and economic drivers of this decline in wetland conversions. The Clean Water Act §404 is strongly and negatively related to conversion rates. Crop prices affect wetland conversions both directly, by influencing returns to investments for conversion, and indirectly, by influencing Conservation and Wetland Reserve Program enrollments as well as the regulatory impact of Swampbuster. Along with the proportion of wetlands remaining, these factors explain 90–94% of the annual variation in total wetland conversions, agricultural conversions, and conversion of forested and emergent wetlands. 相似文献
10.
基于第二次冰川编目数据的中国冰川高度结构特征分析 总被引：2，自引：0，他引：2
主要基于第二次中国冰川编目数据,计算和分析中国西部各个山脉冰川面积随高度分布特征、冰川平衡线高度场的分布规律和积累区比率的分布特征。结果表明：① 各个山脉冰川面积随高度分布呈近似正态分布,冰川最大面积所占总面积的百分比与冰川分布高度差的比值可以作为描述冰川面积随高度分布的形状参数;② 平衡线高度的分布特征受气候和地形影响,由南向北逐渐降低,由东向西逐渐升高;西北和南部高大山脉边缘比较密集,青藏高原内部比较稀疏。③ 冰川积累区比率的分布特征与水汽、地形和物理冰川属性有关。各大山脉外侧和海洋型冰川区积累区比率较小（<0.5）,山脉内侧及高原内陆地区和极大陆型冰川区的积累区比率较大（>0.7）。 相似文献