REGRESSION ON MULTIVARIATE IMAGES:PRINCIPAL COMPONENT REGRESSION FOR MODELING,PREDICTION AND VISUAL DIAGNOSTIC TOOLS

Regression between two blocks(usually called‘dependent’or Y and‘independent’or X)of data is a veryimportant scientific and data-analytical tool.Regression on multivariate images is possible and constitutesa meaningful addition to existing univariate and multivariate techniques of image analysis.The regressioncan be used as a modeling tool or for prediction.The form of the regression equation chosen is dependentupon problem specification and information at hand.This paper describes the use of principal componentregression(PCR).Both model building and prediction are presented for continuous Y-variables.The finalgoal is to supply new image material that can be used for visual inspection on a screen.Also,visual toolsfor diagnosis of model and prediction are provided,often based on derived image material.Examplesof modeling and prediction are given for six channels in a seven-channel satellite image     

30年来呼伦贝尔地区草地植被对气候变化的响应(英文)

Global warming has led to significant vegetation changes especially in the past 20 years. Hulun Buir Grassland in Inner Mongolia, one of the world’s three prairies, is undergoing a process of prominent warming and drying. It is essential to investigate the effects of climatic change (temperature and precipitation) on vegetation dynamics for a better understanding of climatic change. NDVI (Normalized Difference Vegetation Index), reflecting characteristics of plant growth, vegetation coverage and biomass, is used as an indicator to monitor vegetation changes. GIMMS NDVI from 1981 to 2006 and MODIS NDVI from 2000 to 2009 were adopted and integrated in this study to extract the time series characteristics of vegetation changes in Hulun Buir Grassland. The responses of vegetation coverage to climatic change on the yearly, seasonal and monthly scales were analyzed combined with temperature and precipitation data of seven meteorological sites. In the past 30 years, vegetation coverage was more correlated with climatic factors, and the correlations were dependent on the time scales. On an inter-annual scale, vegetation change was better correlated with precipitation, suggesting that rainfall was the main factor for driving vegetation changes. On a seasonal-interannual scale, correlations between vegetation coverage change and climatic factors showed that the sensitivity of vegetation growth to the aqueous and thermal condition changes was different in different seasons. The sensitivity of vegetation growth to temperature in summers was higher than in the other seasons, while its sensitivity to rainfall in both summers and autumns was higher, especially in summers. On a monthly-interannual scale, correlations between vegetation coverage change and climatic factors during growth seasons showed that the response of vegetation changes to temperature in both April and May was stronger. This indicates that the temperature effect occurs in the early stage of vegetation growth. Correlations between vegetation growth and precipitation of the month before the current month, were better from May to August, showing a hysteresis response of vegetation growth to rainfall. Grasses get green and begin to grow in April, and the impacts of temperature on grass growth are obvious. The increase of NDVI in April may be due to climatic warming that leads to an advanced growth season. In summary, relationships between monthly-interannual variations of vegetation coverage and climatic factors represent the temporal rhythm controls of temperature and precipitation on grass growth largely.     

An approach to spatially explicit reconstruction of historical forest in Northeast China

The spatially explicit reconstruction of historical land-cover datasets plays an important role in studying the climatic and ecological effects of land-use and land-cover change（LUCC）. Using potential natural vegetation（PNV） and satellite-based land use data, we determined the possible maximum distribution extent of forest cover in the absence of human disturbance. Subsequently, topography and climate factors were selected to assess the suitability of land for cultivation. Finally, a historical forest area allocation model was devised on the basis of the suitability of land for cultivation. As a case study, we used the historical forest area allocation model to reconstruct forest cover for 1780 and 1940 in Northeast China with a 10-km resolution. To validate the model, we compared satellite-based forest cover data with our reconstruction for 2000. A one-sample t-test of absolute bias showed that the two-tailed significance was 0.12, larger than the significant level 0.05, suggesting that the model has strong ability to capture the spatial distribution of forests. In addition, we calculated the relative difference of our reconstruction at the county scale for 1780 in Northeast China. The number of counties whose relative difference ranged from-30% to 30% is 99, accounting for 74.44% of all counties. These findings demonstrated that the provincial forest area could be transformed into forest cover maps well using the model.     

基于垂直带谱的南北过渡带1:5万植被类型图遥感制图研究——以太白山为例

The compilation of 1:250,000 vegetation type map in the North-South transitional zone and 1:50,000 vegetation type maps in typical mountainous areas is one of the main tasks of Integrated Scientific Investigation of the North-South Transitional Zone of China.In the past,vegetation type maps were compiled by a large number of ground field surveys.Although the field survey method is accurate,it is not only time-consuming,but also only covers a small area due to the limitations of physical environment conditions.Remote sensing data can make up for the limitation of field survey because of its full coverage.However,there are still some difficulties and bottlenecks in the extraction of remote sensing information of vegetation types,especially in the automatic extraction.As an example of the compilation of 1:50,000 vegetation type map,this paper explores and studies the remote sensing extraction and mapping methods of vegetation type with medium and large scales based on mountain altitudinal belts of Taibai Mountain,using multi-temporal high resolution remote sensing data,ground survey data,previous vegetation type map and forest survey data.The results show that:1)mountain altitudinal belts can effectively support remote sensing classification and mapping of 1:50,000 vegetation type map in mountain areas.Terrain constraint factors with mountain altitudinal belt information can be generated by mountain altitudinal belts and 1:10,000 Digital Surface Model(DSM)data of Taibai Mountain.Combining the terrain constraint factors with multi-temporal and high-resolution remote sensing data,ground survey data and previous small-scale vegetation type map data,the vegetation types at all levels can be extracted effectively.2)The basic remote sensing interpretation and mapping process for typical mountains is interpretation of vegetation type-groups→interpretation of vegetation formation groups,formations and subformations→interpretation and classification of vegetation types&subtypes,which is a combination method of top-down method and bottom-up method,not the top-down or the bottom-up classification according to the level of mapping units.The results of this study provide a demonstration and scientific basis for the compilation of large and medium scale vegetation type maps.     

Effects of grazing and climate change on species diversity in sandy grassland, Inner Mongolia, China

To understand the effects of animal grazing activities and climate change on sandy grassland vegetation in northern China, a field grazing and protected enclosure experiment was conducted from 1992 through 2006 in Horqin Sand Land, Inner Mongolia. The results showed that (1) the grazing was primary responsible for changes of the vegetation richness and diversity in the grazing grassland and that changing climate was the main reason for changes in the species richness and diversity in the grassland protected from grazing; (2) light and moderate grazing can promote restoration of the richness and the diversity in the degraded grassland, and heavy grazing could result in a decrease of the richness and diversity; (3) heavy grazing can result in significant decrease of the perennial diversity, and moderate and light grazing promotes increase of the perennial diversity; the grazing, whether heavy or moderate and light grazing, was beneficial to increase of the annual diversity; (4) heavy grazing was not beneficial to diversity of Graminean and Chenopodiaceae, and moderate and light grazing was favorable the diversity of Compositae and Chenopodiaceae; (5) the warm-humid climate was favorable to increase of the richness and the diversity, and the warm-drought climate could result in decease of the richness and the diversity; (6) increased precipitation was favorable to perennial diversity and the diversity of Graminean, Leguminosae, and Compositae, and decreased precipitation had few effects on the annual diversity and Chenopodiaceae diversity.     

基于遥感的黄河三角洲蒸散的动态分析

Evapotranspiration (ET) is an important parameter for water resource management. Compared to the traditional ET computation and measurement methods, the ET computation method based on remote sensing has the advantages of quickness, precision, raster mapping and regional scale. SEBAL, an ET computation model using remote sensing method is based on the surface energy balance equation which is a function of net radiance flux, soil heat flux, sensible heat flux and latent heat flux. The former three fluxes can be computed through the parameters retrieved from remote sensing image, then the latent heat flux can be obtained to provide energy for ET. Finally we can obtain the daily ET. In this study SEBAL was applied to compute ET in the Yellow River Delta of China where water resource faces a rigorous situation. Three Landsat TM images and meteorology data of 1999 were used for ET computation, and spatial and temporal change patterns of ET in the Yellow River Delta were analysed.     

内蒙古植被覆盖度的时空动态变化及其与气候因子的关系(英文)

The vegetation coverage dynamics and its relationship with climate factors on different spatial and temporal scales in Inner Mongolia during 2001-2010 were analyzed based on MODIS-NDVI data and climate data.The results indicated that vegetation coverage in Inner Mongolia showed obvious longitudinal zonality,increasing from west to east across the region with a change rate of 0.2/10°N.During 2001-2010,the mean vegetation coverage was 0.57,0.4 and 0.16 in forest,grassland and desert biome,respectively,exhibiting evident spatial heterogeneities.Totally,vegetation coverage had a slight increasing trend during the study period.Across Inner Mongolia,the area of which the vegetation coverage showed extremely significant and significant increase accounted for 11.25% and 29.13% of the area of whole region,respectively,while the area of which the vegetation coverage showed extremely significant and significant decrease accounted for 7.65% and 26.61%,respectively.On inter-annual time scale,precipitation was the dominant driving force of vegetation coverage for the whole region.On inter-monthly scale,the change of vegetation coverage was consistent with both the change of temperature and precipitation,implying that the vegetation growth within a year is more sensitive to the combined effects of water and heat rather than either single climate factor.The vegetation coverage in forest biome was mainly driven by temperature on both inter-annual and inter-monthly scales,while that in desert biome was mainly influenced by precipitation on both the two temporal scales.In grassland biome,the yearly vegetation coverage had a better correlation with precipitation,while the monthly vegetation coverage was influenced by both temperature and precipitation.In grassland biome,the impacts of precipitation on monthly vegetation coverage showed time-delay effects.     

BOX-COX TRANSFORMATIONS IN THE ANALYSIS OF COMPOSITIONAL DATA

The statistical analysis of compositional data is of fundamental importance to practitioners in generaland to chemists in particular.The existing methodology is principally due to Aitchison,who effectivelyuses two transformations,a ratio followed by the logarithmic,to create a useful,coherent theory thatin principle allows the plethora of normal-based multivariate techniques to be used on the transformeddata.This paper suggests that the well-known class of Box-Cox transformations can be employed inplace of the logarithmic to significantly improve the existing methodology.This is supported in part byshowing that one of the most basic problems that Aitchison managed to overcome,namely thespecification of an interpretable covariance structure for compositional data,can be resolved,or nearlyresolved,once the ratio transformation has been applied.Hence the resolution is not directly dependenton the logarithmic transformation.It is then verified that access to the general Box-Cox family will allowa more accurate use of the normal-based multivariate techniques,simply because better fits to normalitycan be achieved.Finally,maximum likelihood estimation and some associated asymptotics are employedto construct confidence intervals for ratios of the true,unknown compositional constituents.Heretoforethis had not been done even in the context of the logarithmic transformation.Applications to real dataare presented.     

Vegetation greenness modeling in response to climate change for Northeast Thailand

In Northeast Thailand, the climate change has resulted in erratic rainfall and tem- perature patterns. The region has experienced both periods of drought and seasonal floods with the increasing severity. This study investigated the seasonal variation of vegetation greenness based on the Normalized Difference Vegetation Index （NDVI） in major land cover types in the region. An assessment of the relationship between climate patterns and vegeta- tion conditions observed from NDVI was made. NDVI data were collected from year 2001 to 2009 using multi-temporal Terra MODIS Vegetation Indices Product （MOD13Q1）. NDVI pro- files were developed to measure vegetation dynamics and variation according to land cover types. Meteorological information, i.e. rainfall and temperature, for a 30 year time span from 1980 to 2009 was analyzed for their patterns. Furthermore, the data taken from the period of 2001-2009, were digitally encoded into GIS database and the spatial patterns of monthly rainfall and temperature maps were generated based on kriging technique. The results showed a decreasing trend in NDVI values for both deciduous and evergreen forests. The highest productivity and biomass were observed in dry evergreen forests and the lowest in paddy fields. Temperature was found to be increasing slightly from 1980 to 2009 while no significant trends in rainfall amounts were observed. In dry evergreen forest, NDVI was not correlated with rainfall but was significant negatively correlated with temperature. These re- sults indicated that the overall productivity in dry evergreen forest was affected by increasing temperatures. A vegetation greenness model was developed from correlations between NDVI and meteorological data using linear regression. The model could be used to observe the change in vegetation greenness and dynamics affected by temperature and rainfall.     

ESTIMATION IN COMPOSITIONAL DATA ANALYSIS

Compositional data arise naturally in several branches of science,including chemistry,geology,biology,medicine,ecology and manufacturing design.In chemistry,these constrained data seem to occur typicallywhen raw data are normalized or when output is obtained from a constrained estimation procedure,suchas might be used in a source apportionment problem.It is important not only for chemists to be awarethat the usual multivariate statistical techniques are not applicable to constrained data,but also to haveaccess to appropriate techniques as they become available.The currently available methodology is dueprincipally to Aitchison and is based on log-normal models.This paper suggests new parametric andnon-parametric approaches to significantly improve the existing methodology.In the parametric setting,some recent work of Rayens and Srinivasan is extended and a practical regression model is proposed.In the development of the non-parametric approach,minimum distance methods coupled withmultivariate bootstrap techniques are used to obtain point and region estimators.     

A method for estimating vegetation change over time and space

Plant diversity is used as an indicator of the well-being of vegetation and ecological systems. Human activities and global change drive vegetation change in composition and competition of species through plant invasions and replacement of existing species on a given scale. However, species diversity indices do not consider the effects of invasions on the diversity value and on the functions of ecosystems. On the other hand, the existing methods for diversity index can not be used directly for cross-scale evaluation of vegetation data. Therefore, we proposed a 3-dimensional model derived from the logistic equation for estimating vegetation change, using native and non-native plant diversity. The two variables, based on the current and the theoretical maximum diversity of native plants on a given scale, and the result of the model are relative values without units, and are therefore scale-independent. Hence, this method developed can be used directly for cross-scale evaluations of vegetation data, and indirectly for estimating ecosystem or environmental change.     

Vegetation heterogeneity in Monte Desert ecosystems: A multi-scale approach linking patterns and processes

We review the current state of research on vegetation heterogeneity in the Monte Desert at scales varying from landscape to intra-patch. Different factors are related to vegetation heterogeneity at every scale. At a coarse scale (i.e. landscapes and communities) vegetation heterogeneity is commonly determined by abiotic factors, whereas biotic interactions usually influence fine scale (patch, intrapatch) heterogeneity. Communities are distributed at the landscape scale according to differences in precipitation, topography and soil attributes. On the other hand, there is evidence that the spatial pattern of plant patches within communities is determined by runon –runoff processes, although biotic influences such as grazing can induce changes in the spatial pattern of patches when plant cover is strongly reduced. In the same way, at patch and intrapatch scales, biotic interactions (i.e. plant –plant interactions, grazing) determine the size and the species composition of plant patches as well as the distribution of species inside plant patches. However, the mechanisms operating behind such biotic interactions at small scales are commonly related to plant-induced changes in the physical environment. We also found evidence of cross-scale interactions, feedbacks, and non-linear effects such as those induced by grazing disturbance. Our analysis showed some regional differences in patterns and processes related to vegetation heterogeneity along the Monte Desert which can be ascribed to climatic and taxonomic variation among areas. Although information about vegetation heterogeneity and its causes and consequences in the Monte Desert is abundant, some areas where knowledge is scarce are detailed.     

基于NOAA NDVI的植被生长季模拟方法研究

在近几年里, 大尺度的植被生长季监测已经成为全球气候变化研究中的一个重要科学问 题。NOAA/NDVI 数据为研究植被生长季时空变化规律提供了重要手段。文章综述并分析比较了 基于NDVI 估测植被生长季开始、结束、长度等特征参数的方法: NDVI 阈值、时间序列分析、物候 期的频率分布型与NDVI 相结合、主分量分析、利用曲线进行拟合等方法。受不同因素影响, 各方 法有不同的应用局限性, 因此, 在以前研究的基础上, 利用较常用的四种方法: 阈值法、滑动平均 法、最大变化斜率、曲线拟合模型模拟了锡林浩特1991～1999 年的草原生长季, 最后利用野外实 测的草原返青期验证了监测结果。结果表明: 与地面观测数据相结合, 基于阈值可得到较好的草 原返青期; 基于曲线拟合模型能适用于大尺度上的植被生长季变化监测, 但存在问题是拟合曲线 很难接近于实际曲线, 因此, 需要进一步研究的内容是选择合适的曲线估测模型, 监测不同植被 类型生长季的年际变化规律。     

Symbiosis of Biodiversity and Tea Production Through Chagusaba

Chagusaba, which is traditional tea-grass integrated system in Shizuoka, is one of Globally Important Agricultural Heritage Systems (GIAHS) sites in Japan. Chagusaba provide an important habitat for natural plant species. In this study, we investigated the flora in Chagusaba and environmental factors to affect the number of native species estimated their diversity. Fifty Chagusaba were classified in to five vegetation types by Two-way indicator species analysis. In addition, high number of native plant species Chagusaba had low soil pH, covered large areas, and were in places where the land had not been modified. Through statistical analysis (GLM), we found that land modification had an especially significant effect on the diversity of native plants, and that the diversity of native plants was reflected in the history of the use of the land.     

洪河自然保护区湿地主要植被类型物种多样性与表层土壤性质的相关性研究

对洪河湿地主要植被类型物种多样性特征、土壤性质特征及二者之间的相互关系进行了研究。结果表明:按沼泽—沼泽化草甸—湿草甸—灌丛草甸—岛状林的序列,物种丰富度呈逐渐增加的变化格局,物种多样性、均匀度呈先下降后上升的格局,优势度呈先上升后下降的格局;而土壤性质特征则呈不规则变化格局。植物物种多样性与土壤性质之间只有均匀度与速效氮之间存在着极显著正相关,其他均无显著相关。这说明在洪河湿地,土壤性质不是决定上述序列的植被类型的物种多样性的主要因素,而土壤速效氮含量是影响物种均匀度的关键因素,同时,对其他物种多样性特征也有一定影响。     

陆地生态系统模拟研究中IBIS模型开发应用的回顾和展望

全球动态植被模型(DGVM)是研究生态系统复杂过程和全球变化的强有力工具。本文基于过去20多年全球已发表的文献,对得到广泛关注和应用的DGVM之一——集成生物圈模拟器(Integrated Biosphere Simulator,IBIS)的开发、改进、发展及应用进行了总结。IBIS是一个在全球变化等多领域中有着广泛应用的模型。自1996年诞生以来,IBIS在陆地生态系统的碳、氮、水循环,植被动态、陆气耦合、水域系统耦合和气候变化影响等多个方面取得了验证和应用。本文较为系统地阐述了IBIS模型在V2.5版本后的不同发展方向,主要针对IBIS模型在水文过程(蒸散、土壤水分、地下水、径流)、植被动态(植被功能型、土地覆盖变化)、植被生理过程(植被物候、光合作用、植被生长、碳分配)、土壤生物地球化学过程(土壤碳氮循环及反馈、温室气体排放等)以及包括土地利用变化、干扰与管理等人类活动过程等方面的改进与发展进行了全面的综述;在此基础上,对模型在生态系统生产力、碳水收支、水分利用效率、温室气体排放、自然干扰(干旱、火灾、虫害)和人类活动(土地利用变化、农业经营)等方面的应用,以及模型的技术性改进方面进行了回顾;最后对模型的前景和进一步发展提出了一些见解。     

Effects of climate change and urban development on the distribution and conservation of vegetation in a Mediterranean type ecosystem

Climate and land-use changes are projected to threaten biodiversity over this century. However, few studies have considered the spatial and temporal overlap of these threats to evaluate how ongoing land-use change could affect species ranges projected to shift outside conservation areas. We evaluated climate change and urban development effects on vegetation distribution in the Southwest ecoregion, California Floristic Province, USA. We also evaluated how well a conservation network protects suitable habitat for rare plant species under these change projections and identified primary sources of uncertainty. We used consensus-based maps from three species distribution models (SDMs) to project current and future suitable habitat for 19 species representing different functional types (defined by fire-response – obligate seeders, resprouting shrubs – and life forms – herbs, subshrubs), and range sizes (large/common, small/rare). We used one spatially explicit urban growth projection; two climate models, emission scenarios, and probability thresholds applied to SDMs; and high-resolution (90 m) environmental data. We projected that suitable habitat could disappear for 4 species and decrease for 15 by 2080. Averaged centroids of suitable habitat (all species) were projected to shift tens (up to hundreds) of kilometers. Herbs showed a small-projected response to climate change, while obligate seeders could suffer the greatest losses. Several rare species could lose suitable habitat inside conservation areas while increasing area outside. We concluded that (i) climate change is more important than urban development for vegetation habitat loss in this ecoregion through 2080 due to diminishing amounts of undeveloped private land in this region; (ii) the existing conservation plan, while extensive, may be inadequate to protect plant diversity under projected patterns of climate change and urban development, (iii) regional assessments of the dynamics of the drivers of biodiversity change based on high-resolution environmental data and consensus predictive mapping, such as this study, are necessary to identify the species expected to be the most vulnerable and to meaningfully inform regional-scale conservation.     

地球暖化促进植物迁移与入侵

在当今快速的全球气候变化 ,特别是快速而急剧的地球暖化情况下 ,现有植物及其系统被迫 :1)重新适应这种变化了及仍在变化着的环境条件 ;2 )往高海拔及极地方向迁移寻找合适的生存环境 ;或者 3)因其迁移的速度赶不上气候变化的速度而消亡。同时 ,气温升高和降水模式 (包括降水量及其分配 )的变化 ,以及与此相关联的干扰型式 (包括干扰种类、强度及延续时间 )的变化 ,一方面将使现有植物及其生态系统对外来生物的抗性弱化 ,另一方面将激活外来物种的活性 ,导致外来生物的快速扩散与大规模入侵 ,进而排挤和“杀死”当地乡土种 ,减少生物多样性 ,改变原有生态系统的组成、结构和功能 ,最终导致严重的社会经济与生态环境问题。在制定域、国家及全球经济与环境政策时 ,应充分考虑到这些问题 ,以防患于未然     

Using long-term datasets to study exotic plant invasions on rangelands in the western United States

Invasions by exotic species are generally described using a logistic growth curve divided into three phases: introduction, expansion and saturation. This model is constructed primarily from regional studies of plant invasions based on historical records and herbarium samples. The goal of this study is to compare invasion curves at the local scale to the logistic growth curve using long-term datasets. Five datasets ranging 41–86 years in length were recovered from five sites in four western states. Data for the following seven exotic species were analyzed using regression analysis to evaluate fit to a non-linear sigmoidal logistic curve: crested wheatgrass (Agropyron cristatum), dwarf alyssum (Alyssum desertorum), cheatgrass (Bromus tectorum), Lehmann lovegrass (Eragrostis lehmanniana), halogeton (Halogeton glomeratus), Russian thistle (Salsola tragus) and tumble mustard (Sisymbrium altissimum). A greater variety of curve shapes was documented by long-term datasets than those published based on herbaria sampling. Only two species from three different sites and with three different data types met the criteria for fitting a logistic curve. Many of the other species/location combinations were characterized by sporadic spikes and crashes. The general lack of fit with the model may be the results of the complex interactions that drive vegetation change in rangeland environments.     

植物响应气候变化模型模拟研究进展

尽管温室气体的增温效应及幅度大小具有很大的不确定性 ,存在许多分歧 ,然而植物对于气候变化的响应研究仍然成为众多学者关注的热点。近十年来 ,国内外学者从模型模拟的角度进行了多层面定量研究。归纳起来 ,已有的模型可以归并为生物地理相关模型、生态响应面模型、立地模型、植物生理模型以及统计模型等五类。本文逐一进行了评述 ,分析了它们的优势与不足 ,最后对模型模拟的发展趋势进行了展望。