Research on the Application of the Light-Energy Spectrum Fusion Technique to Land and Resources Survey

This paper introduces how to use remote sensing images including Landsat (MSS and TM) andairborne radioactivity images to identify the type of rocks in the areas covered by vegetation. The relationship between light spectrum (Landsat MSS and TM) and energy spectrum (U, Th and K) is discussed on the basis of correlation analysis, and it is proven that there are correlations between the Landsat MSS or TM data and the U, Th and K. data. By using the fusion technique, new images were generated, which contain both the light spectrum and the energy spectrum information.Taking the Lucong basin as the study area, the present paper demonstrates the successful identification of various types of rocks using the fusion technique. Different types of rocks are represented by different colours on the new light-energy spectrum images, so that volcanic rocks of the Jurassic and Cretaceous periods can be discriminated. Another example, in the Lingquan basin in Northeast China, not only the different types of rocks are su     

Developing synergy regression models with space-borne ALOS PALSAR and Landsat TM sensors for retrieving tropical forest biomass

Forest stand biomass serves as an effective indicator for monitoring REDD (reducing emissions from deforestation and forest degradation). Optical remote sensing data have been widely used to derive forest biophysical parameters inspite of their poor sensitivity towards the forest properties. Microwave remote sensing provides a better alternative owing to its inherent ability to penetrate the forest vegetation. This study aims at developing optimal regression models for retrieving forest above-ground bole biomass (AGBB) utilising optical data from Landsat TM and microwave data from L-band of ALOS PALSAR data over Indian subcontinental tropical deciduous mixed forests located in Munger (Bihar, India). Spatial biomass models were developed. The results using Landsat TM showed poor correlation (R² = 0.295 and RMSE = 35 t/ha) when compared to HH polarized L-band SAR (R² = 0.868 and RMSE = 16.06 t/ha). However, the prediction model performed even better when both the optical and SAR were used simultaneously (R² = 0.892 and RMSE = 14.08 t/ha). The addition of TM metrics has positively contributed in improving PALSAR estimates of forest biomass. Hence, the study recommends the combined use of both optical and SAR sensors for better assessment of stand biomass with significant contribution towards operational forestry.     

基于生物气候因子聚类的植物群组划分及其中国生物群区模拟

本研究基于10′×10′空间分辨率运用GIS软件对中国植被分布地理配准数字化,提取获得196个优势植物的地理分布数据以及内插相关的气候因子,运用分位数函数计算获得各个植物地理分布的气候阈值;   在此基础上采用生物气候群组（BAGs）的分类方法,对196个优势植物气候因子数据进行聚类分析,同时结合中国植被的地理分布特征,最后划分出29个BAG组,包括13个乔木组(6个常绿阔叶和2个落叶阔叶群组,4个常绿针叶和1个落叶针叶群组)、10个灌木组(5个常绿灌木和5个落叶灌木群组)和6个草本组。本研究的BAGs划分是建立在较高空间分辨率和定量气候因子阈值的基础上,因此,通过计算获得BAGs气候阈值(包括每日温度变化幅度、地表霜冻频率、月均温度、月降雨量、相对湿度、日照时数和>5℃年积温等),运用生物圈碳循环CARAIB模型,模拟出中国19个生物群区的现代潜在分布。模拟结果较好地表达了中国植被类型的地理分布,尤其较详细地展示了热带、亚热带、温带和高原地区的不同植被类型地理分布格局,但由于东北地区、西部草原以及热带原生植被受人类活动严重干扰,使得模拟结果比该地区的原生植被地理分布的实际面积偏大。总体而言,本研究获得的29个生物气候群组对模拟现代生物群区分布提供了理想的分类依据,其气候阈值也是进行过去和未来生物群区模拟的重要参照数据。     

Estimating gross primary productivity of a tropical forest ecosystem over north-east India using LAI and meteorological variables

Tropical forests act as a major sink of atmospheric carbon dioxide, and store large amounts of carbon in biomass. India is a tropical country with regions of dense vegetation and high biodiversity. However due to the paucity of observations, the carbon sequestration potential of these forests could not be assessed in detail so far. To address this gap, several flux towers were erected over different ecosystems in India by Indian Institute of Tropical Meteorology as part of the MetFlux India project funded by MoES (Ministry of Earth Sciences, Government of India). A 50 m tall tower was set up over a semi-evergreen moist deciduous forest named Kaziranga National Park in north-eastern part of India which houses a significant stretch of local forest cover. Climatically this region is identified to be humid sub-tropical. Here we report first generation of the in situ meteorological observations and leaf area index (LAI) measurements from this site. LAI obtained from NASA’s Moderate Resolution Imaging Spectroradiometer (MODIS) is compared with the in situ measured LAI. We use these in situ measurements to calculate the total gross photosynthesis (or gross primary productivity, GPP) of the forest using a calibrated model. LAI and GPP show prominent seasonal variation. LAI ranges between 0.75 in winter to 3.25 in summer. Annual GPP is estimated to be \(2.11\,\hbox {kg C m}^{-2} \, \hbox {year}^{-1}\).     

Analysing land and vegetation cover dynamics during last three decades in Katerniaghat wildlife sanctuary,India

The change in the tropical forests could be clearly linked to the expansion of the human population and economies. An understanding of the anthropogenic forcing plays an important role in analyzing the impacts of climate change and the fate of tropical forests in the present and future scenario. In the present study, we analyze the impact of natural and anthropogenic factors in forest dynamics in Katerniaghat wildlife sanctuary situated along the Indo-Nepal border in Uttar Pradesh state, India. The study site is under tremendous pressure due to anthropogenic factors from surrounding areas since last three decades. The vegetation cover of the sanctuary primarily comprised of Shorea robusta forests, Tectona grandis plantation, and mixed deciduous forest; while the land cover comprised of agriculture, barren land, and water bodies. The classification accuracy was 83.5%, 91.5%, and 95.2% with MSS, IKONOS, and Quickbird datasets, respectively. Shorea robusta forests showed an increase of 16 km²; while Tectona grandis increased by 63.01 km² during 1975–2010. The spatial heterogeneity in these tropical vegetation classes surrounded by the human dominated agricultural lands could not be addressed using Landsat MSS data due to coarse spatial resolution; whereas the IKONOS and Quickbird satellite datasets proved to advantageous, thus being able to precisely address the variations within the vegetation classes as well as in the land cover classes and along the edge areas. Massive deforestation during 1970s along the adjoining international boundary with Nepal has led to destruction of the wildlife corridor and has exposed the wildlife sanctuary to human interference like grazing and poaching. Higher rates of forest dynamics during the 25-year period indicate the vulnerability of the ecosystem to the natural and anthropogenic disturbances in the proximity of the sanctuary.     

Forest fire spread simulation algorithm based on cellular automata

Traditional models result in low efficiency and poor accuracy when simulating the spread of large-scale forest fires. We constructed an improved model that couples cellular automata with an existing forest fire model to ensure better time accuracy of forest fire spread. Our model considers the impact of time steps on simulation accuracy to provide an optimal time step value. The model was tested using a case study of forest fire spread at Daxing’an Mountain in May 2006. The results show that the optimal time step for the forest fire spread geographic cellular automata simulation algorithm is 1/8 of the time taken for cellular material to be completely combusted. When compared with real fire data from Landsat Thematic Mapper (TM) images, our model was found to have high temporal and spatial consistency, with a mean Kappa coefficient of 0.6352 and mean accuracy of 87.89%. This algorithm can be used to simulate and predict forest fire spread and is also reversible (i.e., it can identify fire source points).     

Trends and variation in vegetation greenness related to geographic controls in middle and eastern Inner Mongolia,China

Extensive studies have investigated the relationships between climate change and vegetation dynamics. However, the geographic controls on vegetation dynamics are rarely studied. In this study, the geographic controls on the trends and variation of vegetation greenness in middle and eastern Inner Mongolia, China (mid-eastern Inner Mongolia) were investigated. The SPOT VEGETATION 10-day period synthesis archive of normalized difference vegetation index (NDVI) from 1999 to 2007 was used for this study. First, the maximum value compositing (MVC) method was applied to derive monthly maximum NDVI (MNDVI), and then yearly mean NDVI (YMNDVI) was calculated by averaging the MNDVIs. The greenness rate of change (GRC) and the coefficient of variation (CV) were used to monitor the trends and variation in YMNDVI at each raster grid for different vegetation types, which were determined from a land use dataset at a scale of 1:100,000, interpreted from Landsat TM images in 2000. The possible effects of geographic factors including elevation, slope and aspect on GRC and CV for three main vegetation types (cropland, forest and steppe) were analyzed. The results indicate that the average NDVI values during the 9-year study period for steppe, forest and cropland were 0.26, 0.41 and 0.32, respectively; while the GRC was 0.008, 0.042 and 0.033 per decade, respectively; and CVs were 10.2, 4.8 and 7.1%, respectively. Cropland and steppe shared a similar trend in NDVI variation, with both decreasing initially and then increasing over the study period. The forest YMNDVI increased throughout the study period. The GRCs of the forest also increased, although GRCs for cropland and steppe decreased with increasing elevation. The GRCs of cropland and steppe increased with increasing slope, but the forest GRCs were not as closely related to slope. All three vegetation types exhibited the same effects in that the GRC was larger on north-facing (shady) slopes than south-facing slopes due to differences in water conditions. The CVs of the three vegetation types showed different features to the GRC. The CVs for all three vegetation types were not affected by aspect. The CVs for forest and cropland showed minor effects with changes in elevation and slope, but the CV for steppe decreased with increasing slope, and increased with increasing elevations to 1,200 m, before decreasing at higher elevations. Our findings suggest that the role of geographic factors in controlling GRC should also be considered alongside climate factors.     

多源遥感数据反演土壤水分方法

基于ASAR-APP影像数据和光学影像数据,根据水云模型研究了小麦覆盖下地表土壤含水量的反演方法。利用TM和MODIS影像构建的植被生物、物理参数与实测小麦含水量进行回归分析,发现TM影像提取的归一化水分指数(NDWI)反演精度较好,相关系数达到0.87。根据这一关系,结合水云模型并联立裸露地表土壤湿度反演模型,建立了基于多源遥感数据的土壤含水量反演模型和参数统一求解方案。反演结果表明:该方案可得到理想的土壤水分反演精度,并可控制参数估计的误差。反演土壤含水量和准同步实测数据的相关系数为0.9,均方根误差为3.83%。在此基础上,分析了模型参数的敏感性,并制作了研究区土壤缺水量分布图。     

Forest biomass carbon dynamics (1980–2009) in western Himalaya in the context of REDD+ policy

Carbon emissions from forests have decreased in the past decade due to conservation efforts, however majority of carbon losses suffered in the past went unnoticed until the role of forests in mitigating climate change was realized. Forestry sector in developing countries is recognized as one of the largest and low cost mitigation options to address climate change. The present study was conducted to assess the multi-temporal biomass carbon mitigation in the temperate forests of western Himalaya using satellite (Landsat MSS, TM, ETM+) and forest inventory data. Forest type density mapping was done through on-screen visual interpretation of satellite data. After conducting preliminary survey in 2009, 45 quadrats (0.1 ha) were laid in six forest types for collecting field inventory data viz., diameter at breast height, tree height, slope and aspect. Biomass carbon (t ha^?1) was estimated for different forest types with different crown densities (open with 10–40% crown density and closed with >40%) using recommended regression equations, ratios and factors. A decreasing trend of carbon (145.13–134.87 mt) was observed over the period of time. Temporal biomass carbon dynamics was analyzed for REDD+ opportunities. The temporal variation of carbon observed was found to be more useful for claiming benefits under negative options (deforestation and forest degradation) of REDD+. The study doesn’t take actual conversions to CO₂ into account. However, the findings are useful in establishing baseline emissions through temporal carbon losses. Further, the study helps in identification of location specific socio-economic drivers of losses that can be used for appropriate mitigation interventions.     

云南种子植物区系地理成分分布格局及其意义

以覆盖云南全境的31个地区植物区系研究报告为基础, 使用Arcview软件制作了14个种子植物属的地理成分多度分布图。 对这些种子植物属的多度分布格局进行分析,获得以下结果:①热带成分合计占所统计的地区植物区系属的29.6%~94.1%, 其最小值是在云南西北部, 最大值出现在云南南部; 温带成分合计占所统计的地区植物区系属的6%~70.4%, 其多度分布恰与热带成分相反。②在热带成分中, 全热带分布属占13.64%~46%, 居热带成分之首, 其最大值出现在云南怒江和元江的干热河谷; 热带亚洲分布属占3.33%~42.26%, 最大值出现在云南南部。③在温带成分中, 北温带分布属占所统计的地区植物区系的2%~30.39%, 居温带成分之首, 最大值出现在云南西北部山地, 其次是东亚分布属, 占2%~14.57%, 最大值出现在滇西北横断山区。④在与云南植被的关系上,热带亚洲分布属、旧世界热带分布属和热带亚洲至热带大洋洲分布属是云南热带森林的主要成分; 全热带分布属和热带亚洲至热带非洲分布属是干热河谷植被的主要成分,北温带分布属主要出现在冷湿山地森林植被中,而东亚分布属、东亚和北美洲间断分布属及旧世界温带分布属与温凉气候地区的森林植被联系相对密切。⑤生物地理上,云南种子植物属的多度分格局暗示了所谓的“田中线”在印度—马来西亚热带植物区系与东亚亚热带—温带植物区系的地理界限上有意义;热带亚洲至热带非洲分布属主要存在于云南的干热河谷,暗示了干热河谷植被具有一定非洲亲缘;云南西部的热带性很强,暗示了在印度板块同亚洲板块相撞、推挤北移及喜马拉雅隆升过程中, 云南西部至东南部的地块可能发生了顺时针转动。⑥在云南, 不论其纬度, 只要是海拔低的区域都是热带区域, 支持云南应是一个基带为热带的大的山原观点。     

Mapping red edge-based vegetation health indicators using Landsat TM data for Australian native vegetation cover

The usefulness of red edge bands, and vegetation indices based on red edge bands, for vegetation health monitoring has already been demonstrated. There are some satellites such as WorldView-2 and Sentinel-2 acquiring images in red edge band data; while, the former data can be expensive and often lack consistent global coverage, the latter does not have a long term archive and consequently cannot be used for a long term time series analysis. This study tests the ability to predict red edge band and red edge-based vegetation indices through freely available Landsat Thematic Mapper data for an Australian Eucalyptus-dominated vegetation cover within and around a mine site. Two modelling strategies including multiple-linear regression as a linear approach and random forests as a non-linear approach were used. The results showed that it is possible to generate red edge derivatives using the Landsat Thematic Mapper data with less than 10% error using both linear and non-linear methods; however, the linear method resulted in higher estimation accuracies than non-linear methods.     

Forest cover change trajectories and their impact on landslide occurrence in the tropical Andes

Tropical mountain regions are prone to landslide hazards. Given the current land pressure with increasing occupation of steep uplands, landslide hazards are expected to increase in the near future. Understanding the factors that control landslide hazards is therefore essential. Rare event logistic regression allows us to perform a robust detection of landslide controlling factors. This technique is here applied to the tropical Andes to evaluate the impact of dynamic land cover changes on landslide occurrences. Land cover change trajectories (i.e. dynamic evolution of land cover through time) were specifically included in the probabilistic landslide analysis. While natural physical processes such as slope undercutting by rivers and failure of oversteepened slopes are important in this tropical mountainous site, landslides are increasingly associated with human activities. The data show that land cover trajectories are associated with landslide patterns. In this humid mountainous site, forest degradation does not lead to a measurable increase in landslide occurrence. However, few years after forests are converted to pastures, a rapid decline of slope stability is observed. Land cover conversion from forest to pasture permanently reduces slope stability. It is assumed that major changes in soil properties and hydrology induced by the vegetation conversion play a role in accelerating landslide hazards.     

Remote Sensing Study of Gold Biogeochemical Effects in the Western Guangdong-Hainan Region——A Case Study of the Hetai Gold Deposit

The vegetation has been poisoned by gold in the western Guangdong-Hainan region. The gold content ofthe leaves there is as high as 10-1961 times the abundance, the chlorophyll content is 10%-30% lower thanthat of the vegetation in metamorphic terrains and 10%-20% higher than that in granite terrains, and thecarotenoid content is 10%-44% lower than the background value. The water content of leaves is 10% to 20%lower than the background value. The cells of leaves are deformed and broken. The leaf surface shows colourspots and becomes yellow or dark green. The spectral reflectance of the leaf surface is 5%-30% higher than thebackground value: the spectral shape has shifted 5-15 nm to the short wavelength. The gray scales of eanopyon images of Landsat TM and airborne imaging scanner (AIS) are 10%-100% higher than the backgroundvalues. On Landsat TM and AIS false colour images, plants poisoned by gold display a yellow color, whichdisinguishes them from background plants. According to the spectral and image features of goldbiogeochemical effects, the author has constructed a gold information system and expert prediction system,and thus two gold target areas and two gold prospect areas have been identified rapidly, economically andaccurately in the western Guangdong-Hainan region which is extensively covered by vegetation.     

Spatial variability of soil quality in the surroundings of a saline lake environment

The spatial variability of the quality of the soil on the shores of a lagoon affected by hydromorphy and/or salinity can be identified (Laguna de Villacañas, Castilla La Mancha) by the use of multitemporal Landsat images in order to analyse these changes. For this purpose, TM and ETM images along with field observations and certain edaphic laboratory parameters are used. In order to identify saline-hydromorphic soils, the spatial variability of chemical and physical properties of a transect, which includes from Solonchaks to Regosols and Cambisols, have been correlated with the Normalised Difference Vegetation Index (NDVI). This index, chosen for specific dates, has proven to be very useful in detecting halophytic vegetation and relating it to the variability of the quality of these soils.     

The relationship between soil geochemistry and die back of montane forests in Sri Lanka: a case study

Tropical montane forests of Sri Lanka form a unique ecosystem with more than 50% of endemic plant species. It has been noted that trees, belonging to different size and age classes of these forest, have been dying due to a yet unknown factor. This phenomenon was first observed in the Horton Plains National Park, which is a high plateau, composed of tropical montane forests. Later dying of forests were observed at several areas including Hakgala montane forest. Physical parameters trace nutrients as well as toxic element concentrations in soils, were studied in order to identify the possible geochemical factors behind the forest die back. Systematic soil sampling was carried out covering the entire Horton Plains National Park and random samples were collected from Hakgala montane forest. Samples were analyzed for available Fe, Mn, Zn, Cu, Ni, Cd, Al and Pb using standard colorimetric and atomic absorption spectrometric procedures. Physical parameters such as pH, moisture content, and conductivity of the samples were also measured. Among extractable micro-nutrients Cu and Zn, and Ni show no deficiency or excess levels. However, the recorded available high concentrations of Fe, Mn and Al can be toxic to certain montane plant species. Acidic moist soil of the area may enhance the toxic effects of these elements. Possible source of these elements should be the underground lithology of the area. According to the results obtained, there is a relationship between forest die back and high Pb concentrations. The same phenomenon was also observed in the Hakgala forest. The distribution pattern of Pb in the Horton Plains coincides well with the die back distribution pattern. The observed Pb values at Horton Plains and Hakgala are almost similar to values observed at Pannipitiya and Dombagaskanda locations, which are located close to main roads carrying heavy traffic. It is quite possible therefore, that Pb toxicity may be a significant factor behind the forest die back even though other factors should not be completely ruled out at this juncture. Strong monsoon winds, bringing Pb from the polluted southwestern part of the country, can be the most possible source of soil Pb. Further studies however are indeed necessary to confirm the conclusion.     

Manifestation of remote sensing data and GIS on landslide hazard analysis using spatial-based statistical models

This paper presents landslide hazard analysis at Cameron area, Malaysia, using a geographic information system (GIS) and remote sensing data. Landslide locations were identified from interpretation of aerial photographs and field surveys. Topographical and geological data and satellite images were collected, processed, and constructed into a spatial database using GIS and image processing. The factors chosen that influence landslide occurrence are topographic slope, topographic aspect, topographic curvature, and distance to rivers, all from the topographic database; lithology and distance to faults were taken from the geologic database; land cover from TM satellite image; the vegetation index value was taken from Landsat images; and precipitation distribution from meteorological data. Landslide hazard area was analyzed and mapped using the landslide occurrence factors by frequency ratio and bivariate logistic regression models. The results of the analysis were verified using the landslide location data and compared with the probabilistic models. The validation results showed that the frequency ratio model (accuracy is 89.25%) is better in prediction of landslide than bivariate logistic regression (accuracy is 85.73%) model.     

Remote Sensing Study of Gold Biogeochemical Effects in the Western Guangdong-Hainan Region1—A Case Study of the Hetai Gold Deposit

Abstract The vegetation has been poisoned by gold in the western Guangdong-Hainan region. The gold content of the leaves there is as high as 10-1961 times the abundance, the chlorophyll content is 10%-30% lower than that of the vegetation in metamorphic terrains and 10%-20% higher than that in granite terrains, and the carotenoid content is 10%-44% lower than the background value. The water content of leaves is 10% to 20% lower than the background value. The cells of leaves are deformed and broken. The leaf surface shows colour spots and becomes yellow or dark green. The spectral reflectance of the leaf surface is 5%-30% higher than the background value; the spectral shape has shifted 5-15 nm to the short wavelength. The gray scales of eanopy on images of Landsat TM and airborne imaging scanner (AIS) are 10%-100% higher than the background values. On Landsat TM and AIS false colour images, plants poisoned by gold display a yellow color, which disinguishes them from background plants. According to the spectral and image features of gold biogeochemical effects, the author has constructed a gold information system and expert prediction system, and thus two gold target areas and two gold prospect areas have been identified rapidly, economically and accurately in the western Guangdong-Hainan region which is extensively covered by vegetation.     

Dynamic analysis of eco-environmental changes based on remote sensing and geographic information system: an example in Longdong region of the Chinese Loess Plateau

The Chinese Loess Plateau is suffering from severe soil erosion. The eco-environmental changes of the plateau are believed to have an important influence on global eco-environmental sustainability; hence, this problem has attracted considerable attention from scientists around the world. This study has two purposes; application of remote sensing (RS) and geographic information system (GIS) techniques in the dynamic analysis of eco-environmental changes in the semiarid zone; and using the Longdong region of the Chinese Loess Plateau as an example, to make dynamic analysis of the eco-environmental changes of the region during the 1986–2004 period and identify controlling factors. Landsat Thematic Mapper (TM) data at a spatial resolution of 30 m were used for analysis. Two training areas were selected in Jingning and Qingcheng counties for analysis using 10-m resolution SPOT and Landsat TM data. The satellite RS images were obtained from the Institute of Remote Sensing Application (IRSA), Chinese Academy of Sciences (CAS). Each images was rectified by Albers Equal Area Conic projection based on 1:50,000 scale topographic maps after spectrum preparation of the images. To make the precision within 1 or 2 pixels, the accurate coordinative control points of the two systems were identified. Then the interpretation key was established based on the land use/cover survey in the study area. The images were classified into six primary environmental types (farmland, forest, grassland, water, construction area, and desert) and 25 sub-types using a visual image interactive interpretation method to obtain vector and attribute data. The resultant accuracy of the land use/cover classification reached 95%. Finally, the transformation areas and ratios of various eco-environmental types in the region were calculated to obtain the transition matrixes of eco-environmental types in the two training areas, Jingning and Qingcheng. This study demonstrates that satellite RS and GIS techniques are effective tools to monitor and analyze the eco-environmental changes in the semiarid region. Visual image interactive interpretation based on GIS technique provides comprehensive information on the direction, rate, and location of eco-environmental changes. The transition matrix model can be used to precisely analyze the variation and rates of the eco-environmental types and their spatial distribution. Great land use changes have taken place Longdong during the 1986–2004 period. These eco-environmental changes were driven by natural and human factors. Natural factors influencing the Longdong region of the Chinese Loess Plateau mainly include temperature, water condition, terrain, soil, and erosion; while human activities include over-cultivation, overgrazing, and fuelwood cutting. As viewed from the extent and severity of the influences, human activities play a very important role in altering the eco-environment of the semiarid region. The study results indicate a need for future research and observation in the semiarid region.     

基于时间序列的地表温度反演:以松江河地区为例

为研究同一地区不同时相的地表温度受植被和地表状况等因素影响后,其温度变化程度的差异,以Landsat 5 TM和Landsat 7 ETM+热红外数据为数据源,采用辐射传输方程法分别反演了松江河地区12个不同月份的地表真实温度,并分析出现稳态温度异常地段的影响因素,计算12个月的温度反演结果的均值和方差。结果表明:1平均值图像中的温度异常区域与绝大部分时相图像的温度异常区域十分相似,图像西北区域和中心区域出现较多温度异常。2方差图像中原系列图像中高温异常的区域,其方差也较大,即时间序列上温度波动较大;原系列图像中植被覆盖度较高的区域,其方差较小,即时间序列上温度变化较为稳定。     

Detection of soil salinity changes and mapping land cover types based upon remotely sensed data

Soil salinity is a major environmental hazard. The global extent of primary and secondary salt affected soils is about 955 and 77 M?ha, respectively. Soil salinity tends to increase in spite of considerable effort dedicated to land reclamation. This requires careful monitoring of the soil salinity status. The objectives of this study were: (a) to evaluate the capability of thematic mapper (TM) and multispectral scanner (MSS) imagery for mapping land cover types, (b) to analyse the spectral features of sail crusts relative to bare soil and gravely soil surface conditions, and (c) to detect the soil salinity changes during the period 1975–2004 in the Ardakan area located in the central Iranian Deserts. The Landsat MSS and TM on two different dates of September 14, 1975 and September 11, 2004, respectively, were used. Due to great confusion between some classes, the TM 6 was included in the band combination. The result of the image classification based on the combination of TM bands 3, 4, 5, and 6 showed of the classification results. For multi-temporal analysis, both TM and MSS images were classified with the same method but with a different number of training classes. The TM-classified image was regrouped to make it comparable with MSS regrouped classified image. The comparison between the classified images showed about 39% of the total area had changed in 29 years. The result of this study revealed the possibility of detecting important soil salinity changes by using Landsat satellite data