Article

On the reliability of Landsat TM for estimating forest variables by regression techniques: a methodological analysis

December 1998
IEEE Transactions on Geoscience and Remote Sensing 36(6):1888 - 1897

DOI:10.1109/36.729360

Source
IEEE Xplore

Authors:

Rosana da Cruz Lino Salvador

São Paulo State University

Xavier Pons

Universitat Autònoma de Barcelona, Catalonia, Spain

In order to build models that relate thematic mapper (TM) imagery to field forest variables, several regression techniques, such as the ones based on the Mallows' C_p and the adjusted R² statistics, were applied. Nevertheless, although the best created models had good fittings (R²>0.65) apparently supported by a clear statistical significance (p<0.0001), later trials tested with additional plots showed that these models were, in fact, nonrobust models (models with very low-predictive capabilities). Two factors were pointed out as causes of these inconsistencies between predicted and observed values: a relatively small number of available field plots and a relatively high number of possible independent variables. Actually, different trials suggested much lower fittings for the expected “really” predictive models. Some restrictions of TM satellite data, such as its radiometric, spectral, and spatial limitations, together with restrictions arising from gathering and processing of field data, might have led to these poor relations. This study shows the need for guarantees stronger than the usual ones before concluding that there is a clear possibility of using satellite information to estimate forest parameters by means of regression techniques

Evaluation of the effect of temporality on forest residual biomass estimation using summer Landsat TM imagery

Chapter

Jan 2008

The use of forest residual biomass (FRB) has grown slower than other biomass resources for renewable energy production, mainly because of a lack of methodologies to assess its quantity at regional scale. In this regard, previous works have demonstrated the utility of satellite images in estimating FRB, considering radiometric variables related to wetness as the most useful. In this context, the objective of this paper is to validate the utility of wetness variables obtained from Landsat TM images to estimate FRB, regardless of the image date in the summer period, which are the most suitable for estimating forest parameters. For this purpose, correlations between FRB data calculated using field work and the Second National Forest Inventory (NFI-2) plots, and spectral variables from three summer Landsat images, which were contemporary with the NFI-2 field work, were analyzed. As a result, it is concluded that both wetness variables considered, MID57 and TC3, are good predictors of FRB independently of the summer moment considered, and map of the study area is created. In addition, as complementary data, the moisture variation of the pine species considered were analyzed in the summer period by means of field work, verifying that significant differences do not exist.

A model-based approach for mapping rangelands covers using Landsat TM image data

Article

Full-text available

Oct 2009

Empirical models are important tools for relating field-measured biophysical variables to remotely sensed data. Regression analysis has been a popular empirical method of linking these two types of data to estimate variables such as biomass, percent vegetation canopy cover, and bare soil. This study was conducted in a semi-arid rangeland ecosystem of Qazvin province, Iran. This paper presents the development of a regression model for predicting rangeland biophysical variables using the original image data of Landsat TM nonthermal bands. The biophysical variables of interest within the rangeland ecosystem were percent vegetation canopy cover, bare soil extent, and stone and gravel which their correlations were analyzed in relation to Landsat TM original data. The results of applying stepwise multiple regression showed that there is a significant correlation between Landsat TM band 2 reflectance values and biophysical variables. The developed models were applied to Landsat TM band 2 and relevant maps were generated. We concluded that such problems as an inexact location of field samples on the image, small size of samples, vegetation heterogeneity may significantly affect the modeling of real rangeland Landsat TM data relationships.

Using Remote Sensing to Estimate a Renewable Resource: Forest Residual Biomass

Chapter

Mar 2012

Improved Biomass Estimation Using the Texture Parameters of Two High-Resolution Optical Sensors

Article

Full-text available

Apr 2011

Accurate forest biomass estimation is essential for greenhouse gas inventories, terrestrial carbon accounting, and climate change modeling studies. Unfortunately, no universal and transferable technique has been developed so far to quantify biomass carbon sources and sinks over large areas because of the environmental, topographic, and biophysical complexity of forest ecosystems. Among the remote sensing techniques tested, the use of multisensors and the spatial as well as the spectral characteristics of the data have demonstrated a strong potential for forest biomass estimation. However, the use of multisensor data accompanied by spatial data processing has not been fully investigated because of the unavailability of appropriate data sets and the complexity of image processing techniques in combining multisensor data with the analysis of the spatial characteristics. This paper investigates the texture parameters of two high resolution (10 m) optical sensors (Advanced Visible and Near Infrared Radiometer type 2 (AVNIR-2) and SPOT-5) in different processing combinations for biomass estimation. Multiple regression models are developed between image parameters extracted from the different stages of image processing and the biomass of 50 field plots, which was estimated using a newly developed "allometric model" for the study region. The results demonstrate a clear improvement in biomass estimation using the texture parameters of a single sensor (r² = 0.854 and rmse = 38.54) compared to the best result obtained from simple spectral reflectance (r² = 0.494) and simple spectral band ratios (r2 = 0.59). This was further improved to obtain a very promising result using the texture parameter of both sensors together (r² = 0.897 and rmse = 32.38), the texture parameters from the principal component analysis of both sensors (r² = 0.851 and rmse = 38.80), and the texture parameters from the av eraging of both sensors (r² = - - 0.911 and rmse = 30.10). Improvement was also observed using the simple ratio of the texture parameters of AVNIR-2 (r² = 0.899 and rmse = 32.04) and SPOT-5 (r² = 0.916), and finally, the most promising result (r² = 0.939 and rmse = 24.77) was achieved using the ratios of the texture parameters of both sensors together. This high level of agreement between the field and image data derived from the two novel techniques (i.e., combination/fusion of the multisensor data and the ratio of the texture parameters) is a very significant improvement over previous work where agreement not exceeding r² = 0.65 has been achieved using optical sensors. Furthermore, biomass estimates of up to 500 t/ha in our study area far exceed the saturation levels observed in other studies using optical sensors.

Geostatistical prediction of vegetation amount using ground and remotely sensed data

Thesis

Jan 1999

Jennifer Lee Dungan

p>Geostatistical methods, which assume spatial dependence, have an untapped potential to map vegetation amount using ancillary data from remote sensing images. Two geostatistical methods, cokriging and conditional simulation, were constructed with aspatial regression in terms of their accuracy and uncertainty description. For a synthetic data set constructed from imaging spectrometer data, spatial regression was most accurate when ground and spectral variables were very closely related ( r between the data exceeding .89). Cokriging was more accurate in all other situations. Conditional simulation, though not as accurate, was superior to the other two methods in reproducing the univariate and spatial characteristics of vegetation amount. The sample size was 300 and the sampling fraction was .3%. For a real data set from western Montana, USA, over 300 ground measurements of conifer canopy cover made in each of two years by the US Forest Service and collocated NDVI values from Landsat TM were used to predict canopy cover in a 97 square km² subarea where the sampling fraction was .03%. The nonlinear aspatial regression model between canopy cover and NDVI had statistically identical parameters in both years, but prediction intervals were very wide and accuracy was low at test points. Cokriged maps had much higher accuracy, but were affected by the small sampling fraction and clumped of ground measurements. Conditionally simulated realizations using collocated cokriging displayed the desirable aspects of cokriging at the same time as presenting plausible global and spatial distributions of canopy cover and were therefore preferable to the cokriged maps.</p

Evaluation of MODIS for groundcover and biomass/feed availability estimates in tropical savannah systems

Technical Report

Full-text available

Mar 2007

This project has developed a remote sensing index for monitoring bare ground1 in all seasons over tropical savannah grasslands at a spatial resolution of 1 km. The MODIS bare ground index (ModBGI) was developed over the Charters Towers Landsat –TM scene area. It uses data from combined AM and PM overpasses of the MODerate Resolution Imaging Spectroradiometers (MODIS) on board the Terra and Aqua satellites and was established using a Landsat –TM ground cover product. The ModBGI model is sufficiently robust and stable to return meaningful results in all seasons. At 1 km the ModBGI index is useful at a catchment and regional planning scale but is of limited use at the paddock/property scale. The project has made use of the Bi-directional Reflectance Distribution Function (BRDF) parameters of MODIS in combination with semi-empirical models to standardise a time series of MODIS image data. Standardising MODIS data to a common sun angle has reduced seasonal variability within data captured at different times of the year and enabled derivation of a biomass model. The biomass model was developed using field data collected from 31 sites during April 2004/5/6 and October 2004/5 and sun angle corrected MODIS data. The biomass model requires more field validation to properly assess its accuracy and suitability to provide stand alone products. It is of interest from a research perspective and with further development could become a source of calibration for existing models such as AussieGRASS.

Forest regeneration in a cold climate – alternative methods.

Article

Full-text available

Jan 2001

Article in English - Icelandic summary SAMANTEKT Meginmarkmið þessara rannsókna er að afla þekkingar til þess að geta hagnýtt beinar sáningar barrtrjáfræs í nýskógrækt á Íslandi. Helstu niðurstöður þessa verkefnis eru þær að í kjölfar beinna sáninga barrtrjáfræs spírar fræið ágætlega og sáðplöntur komast á legg. Ekki er sjáanlegur marktækur munur á því milli Suður- og Austurlands. Af þeim trjátegundum sem notaðar voru gaf sáning stafafuru jafn bestan árangur. Á Héraði skilaði þó sáning rússalerkis viðunandi árangri, en vegna þess hve fræverð er hátt um þessar mundir er aðferðin tæpast raunhæfur kostur í lerkiskógrækt. Á sama hátt gaf notkun plastkeilu jafnbestan árangur við sáningar, bæði hvað varðar spírun, lifun og vöxt plantnanna. Þess ber þó að geta að á Héraði voru áhrif keilunnar mun minni en í Mosfelli, sérstaklega á lifun og vöxt plantna. Skýrist það væntanlega af hagstæðu veðurfari svæðisins. Í Mosfelli virðist skjólið af keilunni ráða úrslitum fyrir lífslíkur plantna. Keilan kemur einnig í veg fyrir afrán á fræi, en það getur verið mikið vandamál við beinsáningar hér, bæði í Mosfelli og Höfða. Athygli vekur hversu vöxtur á sáðplöntunum er lítill fyrstu sumrin. Þær virðast lenda í vaxtarstöðnun (stagnation) sem væntanlega skýrist af næringarskorti og hugsanlega skorti á sambýlisörverum, s.s. svepprót. Tilraunir sem hófust haustið 1997 með áburðargjöf benda til þess að hún auki vöxt plantna og sé nauðsynleg við beinar sáningar hérlendis. Afföll plantna orsakast af svipuðum þáttum og þekkt eru úr gróðursetningarstarfinu, þ.e.a.s. frostlyftingu, frostskemmdum, þurrki og ranabjöllunagi. Skipulag beinna sáninga þarf að taka tillit til allra þessara þátta ef vel á að takast til. Ávinningur beinna sáninga virðist vera mestur þegar notuð er stafafura. Orsakast það af nokkrum þáttum. Fræ hennar er tiltölulega ódýrt og fyrirsjáanlega verður töluvert framboð af því hér innanlands. Hún er ræktuð í tiltölulega þurru mólendi, en þar er helst hægt að mæla með sáningum og einnig ættu rótarvansköpun og stofnsveigjur að vera minna vandamál eftir beina sáningu en eftir gróðursetningu. Þessar tilraunir renna stoðum undir það, að beinar sáningar á barrtrjáfræi í útjörð geti gefið áþekkan árangur og vænta má af hefðbundnum gróðursetningum. Mikill munur er þó á milli einstakra sáningaraðferða og ljóst að ekki dugir að sá fræinu einu og sér. Hjálparaðgerðir eru því nauðsynlegar ef fullnægjandi árangur á að nást.

Improved forest biomass estimates using ALOS AVNIR-2 texture indices

Article

Apr 2011
REMOTE SENS ENVIRON

Estimation of Crown Biomass of Pinus spp. From Landsat TM and Its Effect on Burn Severity in a Spanish Fire Scar

Article

Jan 2009

Remote sensing has been shown to be an efficient tool in the study of forest-fire processes. However, a lack of information on the amount of biomass burnt reduces the accuracy of fire severity and emission models. In this study, we use imagery from the Landsat Thematic Mapper to map crown biomass and burn severity for a large Mediterranean area. Considering the specific characteristics of the Mediterranean environment, two methods to extract useful remote sensing data were employed; both sought to analyze relationships between crown biomass and spectral information. As a result, a crown biomass map of Pinus spp . was created for the entire study area, applying nonlinear regression using the variable MID57 (TM5 + TM7) (R² = 0.651). Considering only P. halepensis pixels that were burnt in the selected fire scar, the relationships between crown biomass and burn severity were found to be high and significant, yielding an R² value of 0.516. Finally, a logistic regression model was constructed to map the presence or otherwise of high burn severity levels using crown biomass as the independent variable, yielding in the confusion matrix an overall percentage of data points correctly classified of 77% and a Kappa statistic in the validation sample of 0.554.

Evaluación de los recursos de biomasa residual forestal mediante imágenes del satélite Landsat y SIG

Article

Full-text available

Jan 2006
GeoFocus

Diversos trabajos han puesto de manifiesto la existencia de correlaciones entre la biomasa forestal y la información de imágenes de satélite. La aplicación de la teledetección para cuantificar esta biomasa presenta ventajas respecto a la de los inventarios tradicionales. En este contexto, existe una falta de trabajos de teledetección dirigidos al estudio de la biomasa residual forestal, cuyo aprovechamiento energético presenta beneficios medioambientales y socio-económicos. El objetivo es desarrollar una metodología para evaluar -mediante regresión logística- la biomasa residual de los bosques de pináceas de Teruel a partir de una imagen Landsat TM, de información topográfica y de variables derivadas del Mapa Forestal de Aragón, tomando como referencia las parcelas del Inventario Forestal Nacional y trabajo de campo. Los resultados indican que el neocanal MSI, TM4 y la variable nivel de madurez son los predictores más importantes para evaluar estos recursos.

A regression model approach for mapping woody foliage projective cover using landsat imagery in Queensland, Australia

Conference Paper

Oct 2004

This paper describes the development of a regression model for predicting foliage projective cover (FPC) using an extensive set of over 2000 field observations for Queensland, Australia. The model includes Landsat TM and ETM+ imagery and a climatological ancillary variable, vapour pressure deficit. The resulting model was validated using independent site data and preliminary validation against FPC estimates from airbourne laser scanner data is presented. Results suggest the model is robust and performing well over a range of soil types and vegetation communities. This regression-based methodology is currently included in the process of monitoring annual woody vegetation change over Queensland and will form the basis of new products for monitoring longer term trends in FPC

Understanding spatial uncertainty in empirical remote sensing models

Thesis

Jan 2007

Nicholas Hamm

p>This thesis addresses the uncertainty in empirical remote sensing models. Specifically the empirical line method (ELM) for atmospheric correction of airborne remotely sensed data is examined. First, the pairing of the field and remotely sensed data for input into the regression model is considered. The typical approach to the ELM averages over all field measurements in each ground target (GT). However, this approach is problematic. Disadvantages were addressed by pairing the field measurements directly with the pixel-based remotely sensed data either using the point-pixel approach or block-pixel approach. The latter is favoured since it explicitly addresses the support issue. It is recommended that at least 50 and preferably 100 measurements should be obtained for each GT. This thesis quantified the impact of positional uncertainty on the outcome of the ELM. When a moderate level of positional uncertainty was introduced, this led to bias in the parameter estimates for the point-pixel approach, although this could be minimised by using a sample size of at least 50 and preferably 100 measurements for each GT. For the geostatistical block-pixel approach introducing positional uncertainty led to an increase in the variogram at short lags but did not, affect parameter estimation for the ELM. Finally, adopting the point-pixel or block-pixel approach led to a regression model with heteroskedastic and spatially correlated residuals. However, these conditions are not handled in standard regression models. Hence a model that incorporates both weighting and spatial correlation was adopted. When this approach was applied to real data, it led to an increase in the uncertainty in the ELM. Spatial correlation may still be presenting a random sample. Hence adopting a random sampling strategy does not obviate the need to model this phenomenon.</p

Monitoring Land Use/Cover Change and Urban Sprawl Using Remote Sensing Data: A Study of Siliguri and Raiganj Urban Agglomerations, India

Chapter

Feb 2022

Spatial assimilation and the dynamicity of urban land use are significant issues in the study of modern towns and cities. Many studies have been conducted to monitor urban land use and sprawl of metropolitan cities or other big cities in India and other countries. But the same kinds of studies conducted for small and medium towns/cities are lesser in number. In this chapter, supervised image classification technique with maximum likelihood classifier algorithm has been applied to estimate the land use/land cover (LULC) change over two time periods using ERDAS imagine (v.14). For assessing the supervised classification technique’s accuracy, confusion or error matrix and kappa coefficient (K) have been applied. A conversion map has been generated from the classified image pairs to measure the quantitative characteristic of changes. Shannon entropy method has been used to find out the urban sprawls. The result of this analysis indicates that the built-up increased significantly from 32.86 km2 in 1990 to 61.16 km2 in 2019 in Siliguri (UA), and for Raiganj (UA), it increased from 4.76 km2 in 1991 to 22.41 km2 in 2019, resulting in a loss in prime agricultural land, fallow land, and vegetation. Shannon entropy has provided excellent assistance for quantifying the sprawling mechanism in both areas to obtain the result. The findings of this chapter may help planners and policymakers guiding urban land management in the context of rapid conversion, as seen in the recent past.KeywordsLand use/coverChange detectionUrban sprawlRemote sensing and GIS

Role of Geospatial Techniques in Soil Erosion Modelling in South Koel Basin, Jharkhand, India

Chapter

Feb 2022

Land deterioration affects cropland and land suitability throughout the globe and is one of the forces that lead to the harming of land richness, thus resulting in meager productivity, caused by both natural as well as anthropogenic factors. Deterioration of crop areas by land abrasion is a global encounter prompting deprivation of supplement-intense surface soil, enhanced effluents from more impervious lands, thus less availability of water flora, and also leads to other calamities such as landslides. That one may examine the aftermath of this very issue, the demand of the hour is to comprehend the reasons, effects, and seriousness of this issue. Researchers, and environmentalists all over the world, are creating strategies for evaluation. Ground appraisals, professional assessments, object notations, end user’s judgments, aptness acclimation, remote sensing, and modeling methods are some of the ways to approach this issue. In this study, RUSLE was used to generate a soil erosion map of the study area. Five years of mean yearlong information is used for the computation of the annual rainfall factor. The soil survey was done by the NBSS & LUP, and parameters such as sand, silt, clay, and organic matter content were taken into consideration to generate the K factor. Likewise, to generate the value of LS factor, a DEM was used. To generate the cover factor values, NDVI was used in a geographic information system (GIS) environment. All these factors were generated in the GIS environment to obtain the results. The erosion maps achieved were categorized into slight (0–5 t/h/year), moderate (5–10 t/h/year), high (10–20 t/h/year), very high (20–40 t/h/year), severe (40–80 t/h/year), and very severe (>80 t/h/year). The map thus generated will serve as a tool for planners for proper conservation practices in the area.

An Introduction to Anthropogeomorphology and Geospatial Technology

Chapter

Feb 2022

In analyzing the Earth’s surface pattern, geochronology, natural resources, natural hazards and landscape growth Anthropogeomorphological mapping has played an important role. It includes dividing the field into metaphysical space entities that use criteria such as morphology (form), genetics (process), shape and function, chronology, correlations between the biophysical environment (land cover, soils, ecology), and spatial and topological connections between surface characteristics (landforms). This chapter explores the importance of geospatial technologies in Anthropogeomorphology and its use of different ways for the study of geomorphology and human involvement. Instead of seeking to cover wider areas of GIS-based scientific research, this chapter focuses on geomorphological problems and knowledge criteria dictating application creation and refinement. It also illustrates the relevance of the shortcomings of perception with regard to depiction, size, measurement and remote sensing.Keywords Anthropogeomorphology LandformRemote sensingGIS

Use of Remote Sensing Techniques in Land Degradation Mapping

Chapter

Feb 2022

Optimal Sampling Intensity in South Korea for a Land-Use Change Matrix Using Point Sampling

Article

Full-text available

Jun 2021

To report changes in land use, the forestry sector, and land-use change matrix (LUCM), monitoring is necessary in South Korea to adequately respond to the Post-2020 climate regime. To calculate the greenhouse gas statistics observing the principle of transparency required by the Climate Change Convention, a consistent nationwide land-use classification and LUCM are required. However, in South Korea, land-use information is available from the 5th National Forest Inventory conducted in 2006 onwards; therefore, developing methods to determine historical LUCM information, including the base year required by the Intergovernmnetal Panel on Climate Change (IPCC), is essential. To determine the optimal sampling intensity for measuring systematic land-use changes and to estimate the corresponding area of land-use categories for previously unmeasured years, seven intensities—2 × 2 km to 8 × 8 km—were tested using the areas of the 3rd and 4th aerial photographs in time series for forestland, cropland, grassland, wetland, and settlements, according to their standard deviations and estimates of uncertainty. Analyses of statistical accuracy, statistical efficiency, economic efficiency, and convenience showed that a sampling intensity of 4 × 4 km was ideal. Additionally, the categorized areas of unmeasured land-use years were calculated through linear interpolation and extrapolation. Our LUCM can be utilized for developing a national greenhouse gas inventory.

Multi-objective ant colony optimization algorithm based on decomposition for community detection in complex networks

Article

Full-text available

Dec 2019
SOFT COMPUT

Community detection aims to identify topological structures and discover patterns in complex networks, which presents an important problem of great significance. The problem can be modeled as an NP hard combinatorial optimization problem, to which multi-objective optimization has been applied, addressing the common resolution limitation problem in modularity-based optimization. In the literature, ant colony optimization (ACO) algorithm, however, has been only applied to community detection with single objective. This is due to the main difficulties in defining and updating the pheromone matrices, constructing the transition probability model, and tuning the parameters. To address these issues, a multi-objective ACO algorithm based on decomposition (MOACO/D-Net) is proposed in this paper, minimizing negative ratio association and ratio cut simultaneously in community detection. MOACO/D-Net decomposes the community detection multi-objective optimization problem into several subproblems, and each one corresponds to one ant in the ant colony. Furthermore, the ant colony is partitioned into groups, and ants in the same group share a common pheromone matrix with information learned from high-quality solutions. The pheromone matrix of each group is updated based on updated nondominated solutions in this group. New solutions are constructed by the ants in each group using a proposed transition probability model, and each of them is then improved by an improvement operator based on the definition of strong community. After improvement, all the solutions are compared with the solutions in the external archive and the nondominated ones are added to the external archive. Finally each ant updates its current solution based on a better neighbor, which may belong to an adjacent group. The resulting final external archive consists of nondominated solutions, and each one corresponds to a different partition of the network. Systematic experiments on LFR benchmark networks and eight real-world networks demonstrate the effectiveness and robustness of the proposed algorithm. The ranges of proper values for each parameter are also analyzed, addressing the key issue of parameter tuning in ACO algorithms based on a large number of tests conducted.

Dynamic analysis of Robinia pseudoacacia forest health levels from 1995 to 2013 in the Yellow River Delta, China using multitemporal Landsat imagery

Article

Jun 2018

The Robinia pseudoacacia forest in the Yellow River Delta (YRD), China, was planted in the 1970s and has continuously suffered dieback and mortality since the 1990s. Timely and accurate information on forest growth and forest condition and its dynamic change as well is essential for assessing and developing effective management strategies. In this study, multitemporal Landsat imagery was used to analyze and monitor changes of the R. pseudoacacia forest in the YRD from 1995 to 2013. To do so, Landsat image band reflectance, three fraction images calculated by using a multiple endmember spectral mixture analysis (MESMA) method, and four vegetation indices (VIs) were used to discriminate three health levels of R. pseudoacacia forest in years 1995, 2007, and 2013 with a random forest (RF) classifier. The four VIs include a difference infrared index (DII) developed in this study, normalized difference vegetation index, soil-adjusted vegetation index, and normalized difference infrared index (NDII), all of which were computed from Landsat Thematic Mapper and Operational Land Imager multispectral (MS) bands. The dynamic changes of the forest health levels during the periods of 1995–2007 and 2007–2013 were analysed. The analysis results demonstrate that three fraction images created by MESMA method and four VIs were powerful in separating the three forest health levels. In addition to the Landsat MS bands, the additional three fraction images increased the classification accuracy by 14−20%; if coupled with the four VIs, the overall accuracy was further increased by 5−6%. According to the importance values calculated by RF classifier for all input features, the DII vegetation index was the second effective feature, outperforming NDII. From 1995 to 2013, a total of 2615 ha of forest in the study area suffered from mortality or loss.

Modelling and mapping of above ground biomass (AGB) of oil palm plantations in Malaysia using remotely-sensed data

Article

Full-text available

Aug 2017

Estimating accurate above ground biomass (AGB) of oil palm plantation in Malaysia is crucial as it serves as an important indicator to assess the role of oil palm plantations in the global carbon cycle, particularly whether it serves as carbon source or sink. Research on oil palm AGB in Malaysia using remote sensing is almost insignificant and it has known that remote sensing provides easy, inexpensive and less time consuming over larger areas. Therefore, this study focuses on evaluating the potential of Landsat Thematic Mapper (TM) data with combination of field data survey to predict AGB estimates and mapping the oil palm plantations. The relationships of AGB with individual TM bands and various selected vegetation indices were examined. In addition, various possibilities of data transform were explored in statistical analysis. The potential models selected were obtained using backward elimination method where R², adjusted R² (R²adj), standard error of estimate (SEE), root mean squared error (RMSE) and Mallows’s Cp criterion were examined in model development and validation. It was found that the most promising model provides moderately good prediction of about 62% of the variability of the AGB with RMSE value of 3.68 tonnes (t) ha⁻¹. In conclusion, Landsat TM offers the low cost AGB estimates and mapping of oil palm plantations with moderate accuracy in Malaysia.

Remote Sensing of Forest Biomass

Chapter

Full-text available

Dec 2014

Forest biomass reflects sequestration or release of carbon between terrestrial ecosystems and the atmosphere. Measuring the size and complexity of forest biomass over large areas can enable us to better understand the environmental processes, availability of renewable energy, and global carbon cycle. This chapter reviews recent progress in measuring forest biomass from remote sensing. In quantifying forest biomass, forest properties are often characterized from three types of remote sensing data. Passive optical spectral reflectances are sensitive to vegetation structure (leaf area index, crown size and tree density), texture and shadow. Radar data measure dielectric and geometrical properties of forests. Lidar data characterize vegetation vertical structure and height. Because these instruments have their advantages and disadvantages in reflecting forest properties, data fusion techniques can combine data from multiple sensors and related information from associated databases to achieve improved accuracy in biomass estimation. The remote sensing data or derived forest attributes are commonly correlated to forest biomass using empirical regression models, non-parametric methods, and physically-based allometric models. Although forest biomass is widely estimated at various scales from remote sensing data, models tend to underestimate large biomass densities and overestimate small ones because of saturation issues. Finally, the assessment and validation of forest biomass obtained from remote sensing is critical because current biomass estimates at large area are of large uncertainties.

Mapping health levels of Robinia pseudoacacia forests in the Yellow River delta, China, using IKONOS and Landsat 8 OLI imagery

Article

Feb 2015

The largest artificial Robinia pseudoacacia forests in the Yellow River delta of China have been infected by dieback diseases. Over the past several decades, this has caused a large amount of mortality of Robinia pseudoacacia forests in this area. Timely and accurate information on the health levels of the forests is crucial to improving local ecological and economic conditions. Remote sensing has been demonstrated to be a useful tool to map forest diseases over a large area. In this study, IKONOS and Landsat 8 Operational Land Imager (OLI) sensor data were collected for comparing their capability of accurately mapping health levels of the artificial forests. There were three health levels (i.e. healthy, medium dieback, and severe dieback) based on explicit tree crown symptoms. After the IKONOS and OLI images were preprocessed, both spatial and spectral features were extracted from the IKONOS and OLI imagery, and a maximum likelihood classification method was used to identify and map health levels of Robinia pseudoacacia forests. The experimental results indicate that the IKONOS sensor has greater potential for identifying and mapping forest health levels. Furthermore, texture features, especially texture variance, derived from the IKONOS panchromatic band, contributed greatly to the accuracy of classification results, achieving an overall accuracy (OA) of 96% for the IKONOS sensor and an OA of 88% for the OLI 2, which used both OLI spectral and IKONOS spatial features, compared with an OA of 74% for the OLI sensor alone. Our results indicate that the texture features extracted from high resolution imagery can improve the classification accuracy of health levels of planted forests with a regular spatial pattern. Our experimental results also demonstrate that classification of an image with a spatial resolution similar to, or finer than, tree crown diameter outperforms that of relatively coarse resolution imagery for differentiating living tree crowns and understorey dense green grass.

Assessing the effects of site heterogeneity and soil properties when unmixing photosynthetic vegetation, non-photosynthetic vegetation and bare soil fractions from Landsat and MODIS data

Article

Feb 2015
REMOTE SENS ENVIRON

The importance of spatial accuracy in characterizing stand types using remotely sensed data

Article

Full-text available

Oct 2011
AFR J BIOTECHNOL

Fatih Sivrikaya

This study assessed the potential use of Landsat 7 ETM+ (15 and 30 m spatial resolutions) images to estimate forest stand attributes such as development stages, crown closure and stand types. The study evaluates the performance of spatial and image classification accuracies between Landsat images (15 and 30 m spatial resolutions) and the forest cover type map (FCTM) with the spatial analysis functions of Geographical Information System (GIS). As a base study, the stand parameters were determined by forest cover type generated with high spatial accuracy of infrared color aerial photography interpretation. The study compared the performance of classification accuracies of satellite images into the forest cover type map (FCTM). The result shows that crown closure was the most successfully classified stand parameters with a 0.92 kappa statistic value and 94.2% overall accuracy assessments in 30 m resolution Landsat 7 image and 0.94 and 95.8% in 15 m resolution Landsat image, respectively. The results indicate that 15 m resolution Landsat 7 image can lead to more accurate mapping of stand type with development stages and crown closures, than 30 m resolution Landsat image according to classification accuracy. However, spatial accuracy was lower than classification accuracy in both images. Spatial analysis clearly showed that the spatial accuracy might be more important than the image accuracy in classification of satellite images to determine forest cover types. This study reveals the differences between image accuracy and spatial accuracy of stand parameters in both Landsat images. The differences were quite significant and should be taken into consideration in forest inventory and land use planning.

An efficient regression strategy for extracting forest biomass information from satellite sensor data

Article

Full-text available

Apr 2005

Monitoring of biomass in forest ecosystems is important as forest is diminishing rapidly in many parts of the world, which is one of the major sources of global carbon emission. Remote sensing is a useful tool for rapid estimation of biomass. Most of the studies currently available to assess biomass from satellite sensor data using regression provide low correlation. The study explores the possibilities to increase it. Various spectral channels and transformations of Landsat Enhanced Thematic Mapper Plus (ETM+) data for predicting biomass in a tropical forest ecosystem of south-eastern Bangladesh were tested. One of the interesting findings of the study is the incorporation of dummy variables based on forest types can dramatically increase the correlation.

Structural attributes of three forest types in central Spain and Landsat ETM+ information evaluated with redundancy analysis

Article

Full-text available

Oct 2008

Antonio Vázquez de la Cueva

Structural attributes of forest, such as canopy crown closure, stand height, stem density and basal area, derived from the third Spanish National Forest Inventory (IFN-3) were used in combination with spectral information derived from Landsat Enhanced Thematic Mapper Plus (ETM+) imagery and topographic information to evaluate their relationships. To deal with the variability found in the literature, three different types of vegetation, dominated by conifers, evergreen sclerophyll and broad-leaved deciduous trees, were analysed. In addition, the analyses were performed using three sets of plots filtered to be successively more homogeneous. A multivariate canonical ordination method, redundancy analysis (RDA), was used to enable the simultaneous evaluation of the two data sets and provide a useful graphical output highlighting the relationships between response (structural attributes) and explanatory (spectral and topographic) variables. Rank correlation analyses were also performed. The low percentage of explained variance at the multivariate analyses and low rank correlation coefficients made it difficult to derive practical empirical models. The strong influence of vegetation type on the results was confirmed, given that each type was sensitive to a different kind of spectral information. Finally, the results did not allow validation of the hypothesis that the relationship should be better when using a more homogeneous set of plots.

Improved landsat-based forest mapping in steep mountainous terrain

Article

Full-text available

Jan 2003

Evaluación de la influencia de la heterogeneidad de los bosques mediterráneos en la estimación de biomasa residual a partir de imágenes Landsat

Article

Full-text available

Fernando Perez-Cabello

This paper evaluates the influence of heterogeneity -mediterranean forest characteristic which induces difficulty on forest parameters estimation using remote sensing- in the performance of logistic regressions to estimate residual biomass using Landsat TM images and ancillary data (DEM and forest map) in the pine forest of Teruel. The Pearson's coefficient of variation (CV) was calcula- ted in the six spectral reflectance TM bands applying a 3x3 pixel window centred at each of the 482 plots of the 2 nd Spanish Forest Inventory. Several models according to the degree of heterogeneity were performed.

Geospatial Strategy for Tropical Forest-Wildlife Reserve Biomass Estimation

Article

Full-text available

Jan 2013

This study focus on the biomass estimation of Sariska Wildlife Reserve using forest inventory and geospatial approaches to develop a model based on the statistical correlation between biomass measured at plot level and the associated spectral characteristics. The multistage statistical technique with incorporated the satellite data of IRS P-6 LISS III gives a precise estimation of biomass. Forest cover, forest stratum, and biomass maps were generated in the study. Spectral signatures along with tonal and textural variations were used to classify different forest types validated with GPS and ground truth data. Altitude dependent vegetation and contour information from toposheets were also considered while classifying imagery during interpretation. Sample plots were laid in study area with 0.1 ha area at intersect of the diagonals of the plots. DBH and height of all the trees inside the plot were measured and converted to biomass using volumetric equations depending upon specific gravity. The specific gravity of each tree species differ from each other and sometimes unique in different regions and varies from forest type of different regions. Estimation of tree biomass can serve as useful benchmark for future studies in related areas. Linear equation obtained was used as the model to generate final biomass map where predicted and estimated biomass were compared for each band of the satellite imageries. Linear, logarithm and power exponential models were compared to each other for correlation coefficient. Correlation between estimated and predicted AGB is 0.835 and coefficient of determination (r2) value is 0.698.

Analysis of Landsat 5 TM Images for Monitoring the State of Restoration of Abandoned Quarries

Article

Full-text available

Mar 2004
Int J Min Reclamat Environ

Quarrying causes high damage to the ecosystems, consequently abandoned sites monitoring is a very important issue. This study describes a methodology to assess the sites’ state of restoration through the analysis of Landsat 5 TM images. The procedure does not require any radiometric corrections for atmospheric effects nor for irradiance effects related to the relief. Such corrections would require a model for the atmosphere at the time the images were taken and a detailed digital topographic model (DEM). In addition, the use of a GIS allowed a direct model calibration from satellite images without any previous field inventory and measurements. The results from the proposed procedure have been compared to in-situ surveyed data.

Approaches for monitoring landscape composition and pattern using remote sensing

Article

Full-text available

Jul 2001

Integration of Radar and Landsat-derived foliage projected cover for woody regrowth mapping, Queensland, Australia

Article

Feb 2006
REMOTE SENS ENVIRON

In Queensland, Australia, forest areas are discriminated from non-forest by applying a threshold ( 12%) to Landsat-derived Foliage Projected Cover (FPC) layers (equating to 20% canopy cover), which are produced routinely for the State. However, separation of woody regrowth following agricultural clearing cannot be undertaken with confidence, and is therefore not mapped routinely by State Agencies. Using fully polarimetric C-, L- and P-band NASA AIRSAR and Landsat FPC data for forests and agricultural land near Injune, central Queensland, we corroborate that woody regrowth dominated by Brigalow (Acacia harpophylla) cannot be discriminated using either FPC or indeed C-band data alone, because the rapid attainment of a canopy cover leads to similarities in both reflectance and backscatter with remnant forest. We also show that regrowth cannot be discriminated from non-forest areas using either L-band or P-band data alone. However, mapping can be achieved by thresholding and intersecting these layers, as regrowth is unique in supporting both a high FPC (> 12%) and C-band SAR backscatter (> ~ ? 18 dB at HV polarisation) and low L-band and P-band SAR backscatter (e.g.

Improved Landsat-based forest mapping in steep mountainous terrain using object-based classification

Article

Sep 2003
FOREST ECOL MANAG

The accuracy of forest stand type maps derived from a Landsat Thematic Mapper (Landsat TM) image of a heterogeneous forest covering rugged terrain is generally low. Therefore, the first objective of this study was to assess whether topographic correction of TM bands and adding the digital elevation model (DEM) as additional band improves the accuracy of Landsat TM-based forest stand type mapping in steep mountainous terrain. The second objective of this study was to compare object-based classification with per-pixel classification on the basis of the accuracy and the applicability of the derived forest stand type maps. To fulfil these objectives different classification schemes were applied to both topographically corrected and uncorrected Landsat TM images, both with and without the DEM as additional band. All the classification results were compared on the basis of confusion matrices and kappa statistics. It is found that both topographic correction and classification with the DEM as additional band increase the accuracy of Landsat TM-based forest stand type maps in steep mountainous terrain. Further it was found that the accuracies of per-pixel classifications were slightly higher, but object-based classification seemed to provide better overall results according to local foresters. It is concluded that Landsat TM images could provide basic information at regional scale for compiling forest stand type maps especially if they are classified with an object-based technique.

Ground-measured spectral signatures as indicators of ground cover and leaf area index: The case of paddy rice

Article

Jan 2001
FIELD CROP RES

A methodology is described to use spectral signatures as indicators of the vegetative status in rice paddy cultures. Ground cover and leaf area index (LAI), considered as indicators of above-ground biomass, and were measured in the field using indirect techniques of digitized close-range vertical photography and Licor 2000 instrument readings, as well as direct destructive sampling. Simultaneously, field reflectance values were collected over specific spectral bandwidths using a hand-held radiometer. Several vegetation indices were derived from these spectral measurements and their predictive power (individually or in combination) with respect to field-measured ground cover and LAI quantified. The additional effects of plant chlorophyll content, paddy depth, water sediment load, and bottom layer color were also investigated. None of these variables added significantly to the predictive power of the models. The models were refined for intra-seasonal variability and a new growth-stage-dependent variable improved the models’ predictive capabilities.The results demonstrated that the monitoring of paddy rice crop status by means of its spectral signatures appears very promising.

Retrieving forest structure variables based on image texture analysis and IKONOS-2 imagery

Article

Full-text available

Jun 2006
REMOTE SENS ENVIRON

Remote sensing techniques have been seen as valuable and low-cost tools for frequent forest inventory purposes. However, estimation errors of relevant forest structure variables remain too high for operational use of high spatial resolution satellite imagery, such as Landsat TM/ETM and SPOT HRV, in temperate forests. Very high spatial resolution images that have been acquired by new commercial satellites, such as IKONOS-2 or QuickBird, are expected to reduce estimation errors to a level that is acceptable by foresters. This study assessed the capability of 1-m resolution IKONOS-2 imagery to estimate the five main forest variables—age, top height, circumference, stand density and basal area—in even-aged common spruce stands. They were estimated on the basis of texture features that were derived from the grey-level co-occurrence matrix (GLCM). The coefficients of determination, R2, of the best models ranged from 0.76 to 0.82 for top height, circumference, stand density and age variables. Basal area was found to be weakly correlated to texture variables (R2 = 0.35). Relative prediction errors of four out of the five studied forest variables were comparable to the usual sampling inventory errors (top height: 10%; circumference: 15%; basal area: 16%; age: 18%), but the stand density estimation error (29%) remained too high for use in forest planning. The sensitivity analysis to the GLCM parameters showed that the most important parameters were the texture feature, the displacement and the window size. The orientation parameter had minimal effects on the R2 values, even if it influenced the values of the texture features.

Classification of natural broad-leaved evergreen forests based on mulit-data for forest inventory in the central highlands of Vietnam

Article

Jan 2009

Huong Nguyen

Für die nachhaltige Bewirtschaftung von Wäldern sind forstliche Karten mit einer zweckmäßigen Waldeinteilung von großer praktischer Bedeutung. Um diese wichtige Datengrundlage zu erstellen und fortzuführen müssen geeignete technische Lösungen gefunden werden, welche Kosten und zeitlichen Aufwand in vernünftigen Grenzen zu halten. In dieser Arbeit wird nach einer optimierten Verknüpfung von Fernerkundungsdaten und terrestrisch erhobenen Daten für die flächige Klassifizierung von tropischen, naturnahen Wäldern gesucht. Dabei wird auch der Frage nachgegangen, welche forstlichen Parameter eine enge Beziehung zu den spektralen Werten in den Satellitendaten haben. Des Weiteren wurden Untersuchungen zur Schätzung der Holzvorräte durchgeführt. Das Untersuchungsgebiet liegt im zentralen Hochland von Vietnam. Dort kommen noch verbreitet natürliche tropische Wälder vor. Als Fernerkundungsmittel wurde eine SPOT 5 Szene vom 5. Februar 2006 verwendet. Für die geometrischen und topographischen Korrekturen des Satellitenbildes wurden Passpunkte (Ground Control Points) aus GPS Messungen sowie ein Geländemodell aus digitalisierten Höhenschichtlinien eingesetzt. Für die topographische Normalisierung wurden drei verschiedene Algorithmen getestet und zwar Cosine, Minnaert und C-correction. Die besten Ergebnisse lieferte die C-correction, weshalb diese Ergebnisse in die weiteren Verarbeitungsschritte einflossen. Die vorbearbeiteten Satellitendaten wurden sowohl mit unüberwachter Klassifizierung, als auch mit überwachten Klassifizierungsverfahren ausgewertet. Für letzteres und auch für die Genauigkeitsüberprüfung der Ergebnisse wurden eine Reihe von Trainingsgebieten in dem Untersuchungsgebiet festgelegt. Insgesamt 120 Plots wurden im Gelände angelegt, verteilt auf die vier unterschiedenen, natürlichen Waldklassen sowie für vier weiteren Landbedeckungsklassen. In den Plots wurden forstliche Parameter wie BHD, Baumhöhe, Kronendurchmesser, Position der Bäume und deren relative Lage zu den Nachbarbäumen gemessen, Kronenschluß und Bedeckungsgrad wurden geschätzt. Zur Berechnung der Zusammenhänge zwischen den erhobenen forstlichen Variablen selbst sowie den Variablen und den spektralen Fernerkundungsdaten wurden Regressionsanalysen angewandt. Dabei wurden enge Zusammenhänge zwischen spektralen Eigenschaften mit der Dichte des Kronendaches, dem mittleren Durchmesser, sowie der Dichte aller Bäume über 35 cm Durchmesser gefunden. Für die Berechnung von charakteristischen spektralen Eigenschaften der einzelnen Waldklassen wurde J-shaped Verteilung Meyer N-BHD zu Grunde gelegt. Basierend auf den Ergebnissen der Satellitenbildklassifikation und den berechneten Kennziffern wird in einem letzten Schritt ein neues, verbessertes Klassifikationsschema für natürliche tropische Wälder in vorgeschlagen. Für die Schätzung des Holzvorrates wurden die vier Kanäle der Spot 5 Szene, drei Principle Component Berechnungen sowie der Normalized Difference Vegetation Index (NDVI) herangezogen. Dabei kamen verschiedene Methoden aus der Geostatistik, Regressionsanalyse und kNN vergleichend zur Anwendung. Die besten Ergebnisse erzielte der geostatistische Ansatz des „Regression Krigings“ mit den SPOT 5 daten. Als zusätzliche Information wurde das potentielle Risiko für ungeregelte Waldnutzungen in Abhängigkeit von der Zugänglichkeit der Waldflächen modelliert. Vier abgestufte „Impact“ Klassen wurden für die zwei betroffenen Waldformationen mit GIS Techniken ausgeschieden. Die daraus resultierenden thematischen Karten zeigen die gefährdeten Flächen und dienen für die Planung entsprechender Maßnahmen bei der Waldbewirtschaftung. Sowohl für Beobachtung der Waldentwicklung als auch für die nachhaltige Bewirtschaftung der Wälder Vietnams werden durch die Kombination von Fernerkundungsdaten und Feldaufnahmen sehr brauchbare Forstkarten zur Verfügung gestellt. In Verbindung mit der Abschätzung potentieller Risiken sind diese ein Planungsinstrument von großer praktischer Bedeutung für die Forstwirtschaft in Vietnam.

Object-Based Mapping and Modelling of Sundarban Mangrove Forests in India

Chapter

Feb 2022

Deforestation because of man-made (anthropogenic) factors is quite high in the developing countries rather than the developed countries. Mangrove forests can be categorized as one of the important natural resources in India. Recently, the mangrove forests in India are declining very rapidly, and it is more prominent in the eastern parts of India, i.e. in the Sundarban area. In this paper, an attempt has been made to find out the spatio-temporal changes of Sundarban mangrove forests in India. Side by side, an attempt has also been made to find out the reasons behind this deforestation. Object-based modelling and normalized difference vegetation index (NDVI) techniques have been applied for mapping and modelling of mangrove forests in Indian Sundarban. This study revealed that mangrove forests in Sundarban areas have been declining very rapidly, and the rate of forest depletion is very high in recent decades than the previous decades. Various anthropogenic factors like agricultural practices, deforestation, animal husbandry activities, etc. have been increased day by day.This study will be throwing a new light in the mapping and modelling of mangrove forests in India. It will also be helpful for the policy planners and decision-makers to take decisions to save the mangrove forests. This is helpful for the sustainable planning and management of the study area.KeywordsObject-based mappingNormalized difference vegetation indexModelling forest coverSustainable planning

Estimation of forest biophysical properties using SPOT HRG data (Case Study: Darabkola Experimental Forest)

Article

Full-text available

Feb 2014

Siavash Kalbi

The main objective of this study was estimation of three forest biophysical properties including volume, basal area and number per hectare using SPOT-HRG data in district 1 Darabkola’s forest. Images were orthorectified to reduce relief displacement errors. Different synthetic bands were generated trough principal component analysis and suitable vegetation indices were created by rationing. After take territorial information, accomplished calculation volume, basal area and number per hectare values in sample plat. Since analyze normalized data, evaluating correlation between territorial and spectral by linear regression. Cross validation getting best correlation model investigating by some sample plat and by relevant evaluation criterions. For volume estimation, results of this study showed that linear combination of NIR and SWIR bands by adjusted determination coefficient 74.5% and root mean square relative error 77.1 m3/ha were better predictors than other combination. For basal area estimation, linear combination NIR band and MSAVI2 index showed best results (R2 adj=80.5; RMSE=4.02). Linear combination NIR and SWIR band by adjusted determination coefficient of 85.7% and root mean square relative error of 44.6 numbers per hectare could better predict the number trees per hectare. Results this study showed SPOT data has high capability for estimation biophysical properties as volume, basal area and number per hectare. Of results this study can use for operational management forest stand

Forest aboveground biomass estimation using SPOT-5 texture indices and spectral derivatives

Conference Paper

Jul 2014

In this paper, spectral derivatives and textural indices derived from a SPOT-5 image have been investigated for the potential of estimating aboveground biomass (AGB) in the region of Liangshui (4710'N, 12853'E), Heilongjiang Provence, China. The results indicated that textural indices were more accurate than spectral derivatives and the best model could be obtained by combining textural indices and spectral derivatives (R=0.77 and RMSE=55.96 t/ha).

Estimation of forest stand diameter class using airborne lidar and field data

Article

Jun 2015

In this study, a method for estimating the stand diameter at breast height (DBH) classes in a South Korea forest using airborne lidar and field data was proposed. First, a digital surface model (DSM) and digital terrain model (DTM) were generated from the lidar data that have a point density of 4.3 points/m2, then a tree canopy model (TCM) was created by subtracting the DTM from the DSM. The tree height and crown diameter were estimated from the rasterized TCM using local maximum points, minimum points and a circle fitting algorithm. Individual tree heights and crown diameters were converted into DBH using the allometric equations obtained from the field survey data. We calculated the proportion of the total number of individual trees belonging to each DBH class in each stand to determine the stand DBH class according to the standard guidelines. More than 60% of the stand DBH classes were correctly estimated by the proposed method, and their area occupied over 80% of the total forest area. The proposed method generated more accurate results compared to the digital forest type map provided by the government.

Advanced Exploratory Data Analysis for Mapping Regional Canopy Cover

Article

Jan 2006
PHOTOGRAMM ENG REM S

USGS Digital Orthophoto Quadrangles (DOQs) are a form of inexpensive, high spatial resolution (1 m ground resolution) imagery available for most regions of the United States. Typically, DOQs have been used in the construction of “basemaps” or as training datasets. In this paper we present a technical approach that uses DOQs as the primary data source to map regional forest canopy cover. This approach, a form of advanced exploratory data analysis (AEDA), can separate areas of crown, shadow, and non-crown vegetation using a single value threshold and a “value range” threshold obtained by analyzing the statistical plots built on a multi-fractal model. By applying AEDA, we can distinguish the crown, crown boundary zone, shadow areas, and non-crown areas within a DOQ mosaic by their distinctive multifractal properties. Over a period of two years, we mapped canopy cover of 20,000 km2 of ponderosa pine-dominated forest across northern Arizona using this technique. We used two hundred ground plot measurements from an independent source to assess the accuracy of the canopy cover map across an 8,000 km2 region.

Forest Structure Estimates with GEOBIA and Multi-scale optical Sensors

Article

Oct 2013
Proceedings of SPIE

The GEOBIA (GEOgraphic Object-Based Image Analysis) approach persists to reveal its effectiveness in remote sensing data analysis, which provides paradigms that integrate analyst's expert knowledge to generate semantically meaningful image-segments. These segments might contribute to the reduction of problems associated with the analysis of the discrete spectral value of a pixel, such as illumination and shaded tree crowns. However, the challenge in this paper is to introduce a GEOBIA as a sophisticated framework toward the automation of forest structural attributes estimate. Optical sensor was examined to develop models for the estimation of forest attributes. Analyses were performed over a forested selected site in the Blue Nile region of Sudan. The framework of the present research involved; segment extraction, field sample selection, forest attributes generalization, model validation, and mapping the predicted attributes. The rationale for incorporating these data is to offer a semi-automatic GEOBIA approach from which forest attribute is acquired through automated segmentation algorithms at the delineated tree crowns or clusters of crowns level. Correlation and regression analyses were applied to identify the relation between a wide range of spectral, textural, contextual metrics, and the field derived forest attributes. Forest structural attribute estimation results acquired from our GEOBIA framework reveal strong relationships and precise estimates. Furthermore, the best fitted models were cross-validated with independent set of field samples, which revealed a high degree of precision.

Empirical relations between Landsat TM spectral response and forest stands near Fort Simpson, Northwest Territories, Canada

Article

Feb 2002

Empirical relationships between forest stand variables, such as age and crown closure, and spectral response measured by the Landsat Thematic Mapper (TM) satellite sensor have long been suggested as an information source to support forest inventories in many regions of the world. Using regression and correlation techniques, the authors have identified the form and strength of these empirical relationships for a sample of forest stands near Fort Simpson, Northwest Territories. Models were strongest for pioneer forest species such as jack pine and trembling aspen as these relationships were characterized by reasonably consistent changes in stand structure and composition. White spruce, a secondary successional species, produced the statistically weakest models from Landsat TM spectral response patterns for all stand variables with the exception of crown closure. The authors attribute the differences in model strength to variable trends in stem growth and stand structural changes caused by differing successional pathways for white spruce stands in this region.

Mountain geoecosystems. GIS modelling of rockfall and protection

Article

Full-text available

Luuk Dorren

An assessment of the spatial variability of basal area in a terrain covered by Mediterranean woodlands

Article

Oct 2000
AGR ECOSYST ENVIRON

Rosana da Cruz Lino Salvador

Spatial patterns of Mediterranean woodlands are largely unknown. An exploratory analysis of the spatial structure of basal area was carried out in this type of community. Specifically, an area of non-disturbed woodlands located in the north-east of the Iberian peninsula was analysed. Direct and indirect evidences for the hypothesis stating that environmental factors lead to the establishment of spatial patterns on the basal area of non-disturbed Mediterranean woodlands were investigated. The results showed that the variability of this parameter occurs within really short spatial intervals (of less than 100m) and neither radiation nor the slope seemed to have an important effect on it. In consequence, results did not support the previously mentioned hypothesis. Among others, long term effects of forest management, an uneven presence of relief conditions and a complex dynamics of basal area within plots were suggested as explanations for the results found. Finally, some recommendations and practical advises were set for future inventories and studies of spatial patterns in Mediterranean woodlands.

Estimating forest canopy density of tropical mixed deciduous vegetation using Landsat data: A comparison of three classification approaches

Article

Feb 2012
INT J REMOTE SENS

Although a number of image classification approaches are available to estimate forest canopy density (FCD) using satellite data, assessment of their relative performances with tropical mixed deciduous vegetation is lacking. This study compared three image classification approaches – maximum likelihood classification (MLC), multiple linear regression (MLR) and FCD Mapper – in estimating the FCD of mixed deciduous forest in Myanmar. The application of MLC and MLR was based on spectral reflectance of vegetation, whereas FCD Mapper was operated on integrating the biophysical indices derived from the reflectance of the vegetation. The FCD was classified into four categories: closed canopy forest (CCF; FCD ≥ 70%), medium canopy forest (MCF; 40% ≥ FCD < 70%), open canopy forest (OCF; 10% ≥ FCD < 40%) and non-forest (NF; FCD < 10%). In the three classification approaches, producer's and user's accuracies were higher for more homogeneous vegetation such as NF and CCF than for heterogeneous vegetation density (VD) such as OCF and MCF. FCD Mapper produced the best overall accuracy and kappa coefficient. This study revealed that only spectral reflectance is not enough to get good results in estimating FCD in tropical mixed deciduous vegetation. This study indicates that FCD Mapper, an inexpensive approach because it requires only validation data and thus saves time, can be applied to monitor tropical mixed deciduous vegetation over time at lower cost than alternative methods.

Canopy-cover thematic-map generation for Military Map products using remote sensing data in inaccessible areas

Article

Jul 2010

Canopy cover is one of the most important elements in concealing military structures and enemy reconnaissance. In this study, we propose an algorithm for automatic generation of density measure of percent canopy cover, which is an attribute of the digital Military Map product, using high-resolution satellite images of inaccessible areas. The thematic mapping process of canopy cover can be divided into image classification, segmentation, and texture analysis. QuickBird and SPOT-5 high-resolution images are classified using Landsat images and military maps. Then, forested areas are extracted from the classified images, and closing and opening operations are executed through morphology filtering. The extracted region is divided into unit-zone objects using a segmentation technique, and the percentage of canopy cover of each object is categorized as one of four levels (0–25, 26–50, 51–75, 76–100%). Two methods were used to establish the percentage of canopy cover for each segment: the discriminant method, using statistical analysis, and the classified canopy ratio method, which calculates the percentage of forest in the segment. The discriminant method showed 48% (QuickBird) and 61% (SPOT-5) accuracy and classified canopy ratio method showed 71% (QuickBird) and 87% (SPOT-5) accuracy. KeywordsCanopy cover–Thematic map–Satellite image–Classification–Segmentation

SEGSAMP: A pixel window sampling method based on image segmentation

Article

Jul 2009

Pixel windows are often used to average the image spectral response values that represent the area over which field measurements are collected to estimate biophysical parameters such as forest stand structure, aboveground biomass, and leaf area index. Averaging spectral values within a pixel window ensures the resulting spectral response is representative of the biophysical parameter, and reduces sampling error from spatial mis-registrations between the image and plot locations. These spectral values are related to field plot measurements through empirical models that may result in poor estimates if (a) the plot is located too close to a spectrally different land feature; and (b) natural and human-caused disturbance occurs between the field data collection and image acquisition. This paper introduces SEGSAMP (SEGmentation SAMPle), an image sampling method to extract pixel values within a segmented sampling window that would be more spectrally representative of biophysical parameters measured from field plots. Written in Arc Macro Language, the method uses a point and polygon shapefile from which to sample only those pixels that are within a defined pixel window size and the image segment the plot is contained in. The output forms a segmented pixel window grid, which can be used to extract pixel values from imagery. A case-study application demonstrates that segmented pixel windows resulted in models whose coefficient of determination values were higher for prediction of stand height and crown closure (height R_adj ² = 0.65 ; crown closure R_adj ² = 0.57) than from square windows (height R_adj ² = 0.57 ; crown closure R_adj ² = 0.48).

Discrimination of Conifer Height, Age and Crown Closure Classes Using Landsat-5 Tm Imagery in the Canadian Northwest Territories

Article

Full-text available

May 2003

Empirical relationships between forest stand attributes and Landsat-5 Thematic Mapper spectral response were developed in order to assess its informational value in support of forest inventory operations in the Northwest Territories. An existing large-area classification procedure, based on a supervised methodology, has been able to generate classes of white spruce and jack pine. Within individual forest species groups, spectral variability related to differences in crown closure, height and age is of interest to forest managers in the region. The objective of this study was to determine the accuracy with which stands within the white spruce and jack pine classes could be further separated into two stand height classes (< 15 m and S 15 m), two age classes (< 100 years and S 100 years) and two crown closure classes (< 30% and S 30%) with a single (summer) Landsat-5 Thematic Mapper (TM) image. Discrimination generally improved with the addition of spectral texture measures where independently assessed accuracies ranged from 60 to 90%. A look-up table was devised for conifer-dominated areas (> 80% dominant species) which could subsequently be assigned for height, age and crown closure class values based on Landsat TM spectral response patterns.

A change detection method for remotely sensed multispectral and multitemporal images using 3-D segmentation

Article

Jun 2001

A new method is proposed fur detection of the temporal changes using three-dimensional (3D) segmentation. The method is a kind of clustering methods for temporal changes. In the method, multitemporal images form a image block in 3D space; x-y plane and time axis. The image block is first divided into spatially uniform sub-blocks by applying binary division process. The division rule is based on the statistical t-test using Mahalanobis distance between spatial coefficient vectors of a local regression model fitted to neighboring sub-blocks to be divided. The divided sub-blocks are then merged into clusters using a clustering technique. The block-based processing, like the spatial segmentation technique, is very effective in reduction of apparent changes due to noise. Temporal change is detected as a boundary perpendicular to the time axis in the segmentation result. The proposed method is successfully applied to actual multitemporal and multispectral LANDSAT/TM images

Percentage vegetation cover of a degrading rangeland from SPOT

Article

Full-text available

Jul 1992
INT J REMOTE SENS

Multispectral SPOT imagery is used to map percentage vegetation cover in the Mackenzie Basin of the South Island, New Zealand. Twenty ground measurements of quadrats 60 by 60 m2 relate percentage vegetation cover to normalized vegetation index. The non-linear calibration curve is C=50 tanh (6.1(NVI-0.22)) + 50 where C is percentage vegetation cover and NVI is normalized vegetation index. The vertical spread of quadrat points about the calibration curve gives the accuracy of a percentage cover prediction from NVI. In the linear part of the calibration curve the upper and lower prediction limits (80% confidence interval) are plus or minus 15% cover. -Authors

Incorporation of relief polynomial-based geometric corrections

Article

Full-text available

Jul 1995
PHOTOGRAMM ENG REM S

This study focuses on the geometrical deformations introduced by relief in images captured by the TM sensor of Landsat satellites and by the HRV sensor of SPOT satellites. Different correction alternatives are presented in order to incorporate altitude data into correction procedures based on first-degree polynomial models. Column and row determinations from the corresponding map coordinates are carried out independently. Three different models for columns and two for rows are proposed. The results have been contrasted with those obtained using classic first- and second-degree polynomial calculations, and with those obtained using an orbital model (for SPOT images). The models presented are easy to implement and provide a level of precision similar to that of the orbital model used, while they are much more efficient in calculation time. In view of the results, the model which integrates altimetric data into a single first-degree polynomial seems of particular interest.

Remote sensing of forest health

Article

Full-text available

May 2012

Extensive monitoring of forest health in Europe has been carried out for two decades, based mainly on defoliation and discolouration. Together these two variables reflect chlorophyll amounts in the tree crown, i.e. as an indicator of foliar mass, and chlorophyll concentration in the foliage, respectively. In a current project we try to apply remote sensing techniques to estimate canopy chlorophyll mass, being a suitable forest health variable. So far, we limit this to Norway spruce only. LIDAR data here play an important role, together with optical and spectral data, either from survey flights or from satellites. We intend to model relationships between foliar mass and LIDAR data for sample trees, and then scale up this to foliar mass estimates for the entire LIDAR area. Similarly, we try to scale up chlorophyll concentrations in sample trees, by modelling a relationship between sample tree chlorophyll and hyper-spectral data. The estimates of foliar mass and chlorophyll concentrations are then aggregated to every 10x10 m pixel of a SPOT satellite scene which is also covered by airborne data, providing an up-scaled ground truth. If we are successful with this, it might be a starting point for developing a new nationwide forest health monitoring system in Norway.

Satellite Inventory of Minnesota Forest Resources

Article

Full-text available

Aug 1993
PHOTOGRAMM ENG REM S

The methods and results of using Landsat Thematic Mapper (TM) data to classify and estimate the acreage of forest covertypes in northeastern Minnesota are described. Portions of six TM scenes covering five counties with a total area of 14,679 square miles were classified into six forest and five nonforest classes. The approach involved the integration of cluster sampling, image processing, and estimation. Using cluster sampling, 343 plots, each 88 acres in size, were photo interpreted and field mapped as a source of reference data for classifier training and calibration of the TM data classifications. Classification accuracies of up to 75 percent were achieved; most misclassification was between similar or related classes. An inverse method of calibration, based on the error rates obtained from the classifications of the cluster plots, was used to adjust the classification class proportions for classification errors. The resulting area estimates for total forest land in the five-county area were within 3 percent of the estimate made independently by the USDA Forest Service. Area estimates for conifer and hardwood forest types were within 0.8 and 6.0 percent respectively, of the Forest Service estimates. A trial of a second method of estimating the same classes as the Forest Service resulted in standard errors of 0.002 to 0.015. A study of the use of multidate TM data for change detection showed that forest canopy depletion, canopy increment, and no change could be identified with greater than 90 percent accuracy. The project results have been the basis for the Minnesota Department of Natural Resources and the Forest Service to define and begin to implement an annual system of forest inventory which utilizes Landsat TM data to detect changes in forest cover.

Estimation of forest parameters through fuzzy classification of TM data

Article

Full-text available

Feb 1995

Several studies have investigated the utility of Landsat 5 TM imagery to estimate forest parameters such as stand composition and density. Regression equations have generally been used to relate these parameters to the radiance responses of the TM channels. Such a method is not feasible in highly complex landscapes, where forest mixtures and terrain irregularities may obscure the existence of simple relationships. A fuzzy approach to the problem is presented based on a multi-step procedure. First, some typical forest plots with known features are spectrally identified. A maximum likelihood fuzzy classification with nonparametric priors is then applied to the study images, so as to derive fuzzy membership grades for all pixels with respect to the typical plots. Finally, these grades are used to compute the estimates of the forest parameters by a weighted average strategy. The method was tested on a complex, rugged area in Tuscany mainly covered by deciduous and coniferous forests. Two TM scenes and accurate ground references taken in spring and summer 1991 were utilized for the testing. The first results, statistically evaluated in comparison with those of a more usual multivariate regression procedure, are quite encouraging. The possible application of the fuzzy approach to other cases of environmental monitoring is finally discussed

An Introduction to Statistical Methods and Data Analysis

Article

Aug 1994

Introduction to Linear Regression Analysis.

Article

Jan 1994

Regression Analysis

Book

Jan 1990

Some comments on CP

Article

Jan 1973
TECHNOMETRICS

Colin Mallows

An Introduction to Statistical Methods and Data Analysis

Article

Jan 1984

R.L. Ott

Numerical recipes in C

Book

Jan 1992

Applied Regression Analysis

Article

May 1999
AM STAT

Discrimination of subalpine forest species and canopy density using digital CASI, SPOT PLA, and Landsat TM data

Article

Oct 1994
PHOTOGRAMM ENG REM S

S.E. Franklin

The number of forest attributes which can be remotely sensed with sufficient precision and accuracy for stand inventory purposes is large, and as yet not fully identified, for different sensor systems and under a variety of forest conditions. In this paper, three digital image data sets were compared and used to discriminate a range of forest cover types in the Subalpine Forest Region west of Calgary, Alberta, Canada. Compact Airborne Spectrographic Imager (CASI) data, SPOT Panchromatic Linear Array (PLA) data, and Landsat Thematic Mapper (TM) data were collected over forest stands that different in species, density, and height. As in previous studies of mountainous areas, a digital elevation model (DEM) was used to increase classification accuracy (measured as the percent agreement with conventional aerial photointerpretation and field observations) to more acceptable levels. For example, the PLA data did not contain enough information to discriminate the training sample with better than about 50 percent of the sites correctly classified; but accuracies up to 74 percent were achieved when DEM data were added to the decision rules for this same sample. Various combinations of the TM and CASI image data both alone and augmented with elevation and first-order derivatives were tested, with best accuracies above 90 percent. Correlation matrices were also constructed and interpreted for these data to study the effect of differing spectral and spatial resolutions, and the topographic effect on discrimination of forest inventory parameters.

The importance of measurement error for certain procedures in remote sensing at optical wavelength

Article

Jan 1986
PHOTOGRAMM ENG REM S

Two problems are identified in the use of linear regression to relate remotely sensed data to ground variables: a specification problem and an errors problem. The extent of the errors is examined for commonly measured remotely sensed variables and ground variables. Three alternative methods of line fitting are examined: Wald's grouping methods, the reduced major axis, and a least- squares procedure. The least-squares method is recommended if the data are available.-Authors

Applied Regression Analysis: A Research Tool.

Article

Mar 1990

Trees as Crop Plants.

Article

Apr 1987

Regression Analysis: Theory, Methods, and Applications.

Article

Dec 1991

Regression Analysis: Theory, Methods, and, Applications.

Article

Sep 1991

Applied Regression Analysis: A Research Tool

Article

Dec 1990

John O. Rawlings

Factors affecting the remotely sensed response of coniferous forest plantations

Article

Jan 1993
REMOTE SENS ENVIRON

Remote sensing of forest biophysical properties has concentrated upon forest sites with a wide range of green vegetation amount and thereby leaf area index and canopy cover. However, coniferous forest plantations, an important forest type in Europe, are managed to maintain a large amount of green vegetation with little spatial variation. Therefore, the strength of the remotely sensed signal will, it is hypothesized, be determined more by the structure of this forest than by its cover. Airborne Thematic Mapper (ATM) and SPOT-1 HRV data were used to determine the effects of this structural variation on the remotely sensed response of a coniferous forest plantation in the United Kingdom. Red and near infrared radiance were strongly and negatively correlated with a range of structural properties and with the age of the stands but weakly correlated with canopy cover. A composite variable, related to the volume of the canopy, accounted for over 75% of the variation in near infrared radiance. A simple model that related forest structural variables to the remotely sensed response was used to understand and explain this response from a coniferous forest plantation.

Quantitative relationship between forest growth rates and Thematic Mapper reflectance measurements

Article

Mar 1991

Effective forest management requires reliable forecasts of timber growth. In New Brunswick, the growth of spruce-fir stands is greatly affected by annual and cumulative defoliation caused by the spruce budworm. This study was designed to determine whether the Thematic Mapper (TM) could detect defoliation in a way which was correlated with stand growth well enough to be used as a predictive tool. Regression analysis was used to explore relationships between stand reflectance in the Thematic Mapper bands and stand volume and growth measured in permanent sample plots. Good (R 2>0.8) relationships were found between net annual spruce-fir volume change and vegetation condition indices incorporating TM near-infrared and shortwave infrared reflectances. Good relationships were also found between live spruce-fir volume and TM band 4 (near-infrared) reflectance. These results have encouraged us to plan a province-wide calibration and measurement programme aimed at improved stand development forecasts for the 1992 New Brunswick timber inventory and allocation.

Estimation of biomass and other characteristics of boreal forest over extensive areas using NOAA AVHRR data

Article

Jan 1995
Proceedings of SPIE

Models using NOAA AVHRR data for estimating the areal distribution of biomass over the Boreal coniferous zone are developed. The method uses corresponding models that are estimated using Landsat TM data as input, and includes a mosaicing scheme for AVHRR images to cover very extensive areas. The estimation method may be adjusted to some other forest characteristics too.

Approach to forestry inventory and mapping by means of multi-spectral airborne data

Article

Jan 1995

In this study we evaluate the possibilities of the airborne sensor CASI (Compact Airborne Spectrographs Imager) when applied to mapping of variables characteristic of an ecologic and forestry inventory, i.e., wood biomass, leaf biomass, leaf area index (LAI), canopy cover, etc. The use of multiple regression techniques has made possible to obtain correlation values between 0.76 and 0.94 (p<0.05), between radiometrical and forestry data. The extrapolation ability of the models obtained has been verified starting from the comparison among different flight lines and several groups of sample areas.

Relaxation vegetation index in non-linear modelling of ground plant cover by satellite remote-sensing data

Article

Dec 1993

In order to obtain a model equation for the calculation of percentage plant cover by multi-spectral radiances remotely-sensed by satellites, a regression procedure is used to connect space remote-sensing data to ground plant cover measurement. A traditional linear regression model using the normalized difference vegetation index (NDVI) is examined by remote-sensing data of the SPOT satellite and ground measurement of LCTA project for a test site at Hohenfels. Germany. A relaxation vegetation index (RVI) is proposed in a non-linear regression modelling to replace the NDVI in linear regression modelling to get a better calculation of percentage plant cover. The definition of the RVI iswhere Xi is raw remote-sensing data in channel i. Using the RVI, the correlation coefficient between calculated and observed percentage plant cover for a test scene in 1989 reaches 0·9 while for the NDVI it is only 0·7; the coefficient of multiple determination R reaches 0·8 for the RVI while it is only 0·5 for the NDVI. Numerical testing shows that the ability of using the RVI to predict percentage plant cover by space remote-sensing data for the same scene or the scene in other years is much stronger than the NDVI.

Volume quantification of coniferous forest compartments using spectral radiance recorded by Landsat Thematic Mapper

Article

Jun 1992

Jonas Ardö

This study aims to (i) investigate the relationship between spectral radiance recorded by the Landsat Thematic Mapper and the volume (m ha) of forest compartments in a coniferous forest area in southern Sweden and (ii) establish and evaluate a regression for volume estimation of forest compartments. A regression of volume on spectral radiance of TM5 was established with Wald's method.Spectral radiance in Landsat TM band 5 had the largest correlation with field measurements of volume, r =−0·79. The correlation coefficient between observed volume and estimated volume was 0·83 and the standard error of estimate was 46·5mha. Accepting an error of ± 50mha in the estimation resulted in a classification accuracy of 77 per cent. It was concluded that there is a stronger relationship between spectral radiance and volume for compartments with small volumes than for compartments with large volumes.

Assessment of forest damage with Landsat TM: Correction for varying forest stand characteristics

Article

Mar 1994
REMOTE SENS ENVIRON

S. Ekstrand

The utility of satellite data in forest decline assessment is influenced by effects associated with variations in stand characteristics such as species composition, age, and density. In this study, the spectral effects of stand variations were determined for forest dominated by Norway spruce. The usefulness of digitized stand data for potentially reducing these effects were investigated. It is suggested that a damage estimation algorithm based solely on Landsat TM Band 4 is more appropriate than earlier proposed ratio algorithms in areas with moderate defoliation symptoms but no chlorosis. The two factors with the strongest negative effect on the defoliation assessment were varying hardwood and pine component. A hardwood component of 20% completely neutralized a defoliation of 20%. Age had a clear spectral effect up to 70 years, but above that the response was stable. There was no confusion between defoliation classes in stands with moderate to high density. A spruce defoliation model that used stand data from digitized forest maps to modify the intensity values of TM Band 4 prior to estimation of defoliation was developed. The resulting assessment of moderate defoliation in forest areas on level ground was of adequate accuracy when the spruce component was larger than 75%.

Choosing Subset Regression

Article

Jan 1973
TECHNOMETRICS

Colin Mallows

Plane Answers to Complex Questions: The Theory of Linear Models

Article

Dec 1997

R. Christensen

Providing a wide-ranging introduction to the use of linear models in analyzing data, this text presents a vector space and projections approach to the subject. The topics covered include ANOVA, estimation, hypothesis testing, multiple comparison, regression analysis, and experimental design. Also covered are: testing for lack of fit; models with singular covariance matrices; variance component estimation; best linear prediction; colinearity; and variable selection.

Coniferous forest leaf area index estimation along the Oregon transect using compact airborne spectrographic imager data

Article

Jan 1995

A new forest light interaction model in support of forest monitoring

Article

Oct 1992
REMOTE SENS ENVIRON

Landsat TM data may be of use for forest timber and vitality monitoring. Literature study reveals inconsistent results of techniques based on spectral indices and regression. Existing models appear incomplete because they do not account for shadowing between trees, or because they treat trees as opaque bodies. For this reason a new forest light interaction model has been developed. It accounts for both shadowing and crown transparency. The inverse model has been implemented as an image processing algorithm. Model results are studied and compared with ground data from the Kootwijk forest in The Netherlands. Results indicate that the model gives a fair representation of reality. In this respect, the utility of the NDVI is also discussed. More research is necessary for further validation.

Forest Ecosystem Processes at the Watershed Scale - Sensitivity to Remotely-Sensed Leaf-Area Index Estimates

Article

Sep 1993
INT J REMOTE SENS

Recent research has shown that general trends in forest leaf area index along regional climatic gradients can be adequately characterized by using ratios of near-infrared and red reflectances. However it has proven difficult to represent properly the spatial distribution of Leaf Area Index (LAI) at subregional scales such as small catchments. The key problem at Thematic Mapper scale is the variation in canopy closure and understorey contribution, which dramatically influences near-infrared reflectance from conifer forests. In this paper, a new spectral index is presented to estimate LAI of conifer forests using a combination of Red, NIR and mid-IR reflectances from the Landsat Thematic Mapper (TM). A simulation system (RHESSys) was used first, to generate potential vegetation patterns around a watershed in order to test them against remotely-sensed vegetation patterns, and secondly, to test the sensitivity of forest ecosystem processes to LAI estimated from combinations of the Thematic Mapper data. The relation between Normalised Difference Vegetation Index (NDVI) and LAI is poorly defined at TM scale because of the outsized contribution of understorey vegetation and background materials to the NIR reflectance in open canopies. The mid-IR correction factor acting as a scalar for canopy closure scaled down the inflated NDVI in the open canopies, resulting in an improved relation between NDVI and LAI. LAI estimates from the MIR corrected NDVI better represented the vegetation patterns in Soup Creek watershed than those from uncorrected NDVI both in terms of magnitude and spatial patterns. Simulations using LAIs derived from corrected NDVI showed lower rates evapotranspiration and net photosynthesis. Differences in mean responses of evapotranspiration and photosynthesis were as large as 8 cm and 2 ton C ha-1 yr-1 respectively between simulation runs using LAIs from corrected and uncorrected NDVI.

Introduction to Linear Regression Analysis

Book

Jan 1992

Variability of Landsat MSS spectral responses of forests in relation to stand and site characteristics

Article

Sep 1987
INT J REMOTE SENS

Stephen J. Walsh

The use of field measures of slope angle, slope aspect, cover type, crown size and crown density is evaluated in appraising the variability of Landsat Multispectral Scanner (MSS) spectral responses for 182 sample sites within Crater Lake National Park, Oregon. Multiple linear regression models indicate that 73, 72, 71 and 57 percent of the variation in the mean response of MSS bands 4, 5, 6 and 7, respectively, was explained by the environmental variables entered into the models. In general, crown size and crown density are less important in altering spectral response than terrain orientation. This type of analysis is useful in guiding field work for remote sensing studies into areas that are environmentally diverse and which are, therefore, capable of significantly altering the spectral response of cover types.

Linear regressions for canopy cover estimation in acacia woodlands using Landsat-TM, -MSS and SPOT HRV XS data

Article

Jul 1993

H. LARSSON

The aim of the study was to establish remote sensing models for the estimation of canopy cover in Acacia woodlands. The models were established using Landsat-TM and MSS data and SPOT HRV XS data and based on field data from eastern Sudan. The models were derived using the Reduced Major Axis (RMA) method. Correlation coefficients between NDVI and canopy cover are for Landsat-TM 0-552, for Landsat-MSS 0-698 and for SPOT HRV XS 0-718. The confidence intervals of predicted canopy cover are also presented.

An Introduction To Statistical Methods and Data Analysis

Chapter

Sep 1983

Lyman Ott

An assessment of canopy chemistry with AVIRIS-A case study in the Landes Forest, South West France

Conference Paper

Sep 1993

Studies the capability of airborne (AVIRIS) and laboratory high spectral resolution information for assessing the chemical composition (lignin, nitrogen, cellulose, ...) of a pine forest (Les Landes, SW France). Simultaneously with AVIRIS acquisition, an atmospheric profile and a forest vegetation sampling for chemical and laboratory spectral analyses, were collected. Predictive relationships between concentrations of nitrogen (r=97%), ligin (r=89%), cellulose (r=83%), and reflectances of pre-treated pine needles were determined through stepwise regression analyses. A methodology was designed to assess their extrapolation to remotely acquired spectrometric data: (1) geometric and atmospheric corrections, (2) registration within a biophysical data base (LAI, biomass, ...), and (3) comparative statistical analysis of laboratory and airborne spectrometric information. Nitrogen and cellulose concentrations predicted with canopy reflectances were relatively correlated with actual concentrations (74% and 79% respectively); poorer results were obtained for lignin (55%). Atmospheric corrections did not improve correlations. It was attempted to improve these results while taking into account the influence of the canopy structure and total quantity of chemical compounds. Predictive equations base on laboratory measurements were applied to reflectances of pine needles that were computed through the inversion of two reflectance models. This approach only improved correlations for lignin (74%). Finally, chemical concentrations in the studied area were mapped in order to provide spatial information suitable for ecosystem models

Remote sensing and the measurement of geographical entities in a forested environment. 2. The optimal spatial resolution

Article

Aug 1994
REMOTE SENS ENVIRON

The prime objective of this study was to propose and test a method to identify the optimal spatial resolutions for detection and discrimination of coniferous classes in a temperate forested environment. The approach is based on the paradigm that there is an intricate relationship between the definition and the measurement of geographical entities and implies the following steps: 1) a priori define the geographical entities under investigation, 2)determine an optimization criterion for the choice of a sampling system, 3) progressively aggregate data acquired from a fine spatial sampling grid, 4) apply the optimization criterion on the series of spatially aggregated data, and 5) verify the validity of the results obtained in relation to the goal of the study. Airborne MEIS-II data, acquired at 0.5 m in eight spectral bands of the visible spectrum, were used for the study. Fourteen forest classes, at the stand level, were defined on the basis of four attributes: species, density, height, and organization of the trees. Representative sites for each forest class were selected. From the center of each site, the spatial resolution of the original data was degraded to 29.5 m, with an increment of 1 m, using an averaging window algorithm. The intraclass variance was calculated for each forest class, at every spatial resolution and for the eight spectral bands. The minimal variance was used as the indicator of the optimal spatial resolution. To evaluate the importance of the optimal resolution for class discrimination, a bivariate test of variance was performed for each pair of forest class considered at their optimal spatial resolution. Profiles of spectral separability were also established in relation to the whole series of spatial resolutions. The results show that, for all coniferous classes and for the eight spectral bands considered in the study, there is a minimal value in intraclass variance that indicates the optimal spatial resolution for each class, varying between 2.5 m and 21.5 m. The optimal spatial resolution is primarily affected by the spatial and structural parameters of the forest stands. The analysis of variance between each pair of forest classes considered at their respective optimal spatial resolution reveals that all classes are significantly different in at least two spectral bands, except for 10 pairs. The spectral separability of the forest classes is at a maximum at, or very close to, their optimal spatial resolution. The study confirms the validity of the concept of optimal spatial resolution and proposes an original solution to the problem of the adequate scale of measurement for geographical entities.

A simple radiometric correction model to improve automatic mapping of vegetation from multispectral satellite data

Article

May 1994
REMOTE SENS ENVIRON

A simplified model for radiometric corrections has been used to improve nonsupervised classification of vegetation cover in a hilly area near Barcelona, Spain. A digital elevation model and standard parameters for exoatmospheric solar irradiance, atmospheric optical depth, and sensor calibration are the only inputs required. Radiometric classes obtained by cluster classification of Landsat TM images from nonradiometrically corrected images include several classes related to terrain illumination, but not to vegetation or thematic cover differences. The use of radiometric correction allows identifying all radiometric classes obtained as vegetation or thematic classes with 83.3% global accuracy. Classes obtained include Pinus halepensis, Quercus ilex, and Quercus cerrioides forests, shrublands, grasslands, urban areas with vegetation, urban areas without vegetation, and denuded areas. Radiometric correction helps in estimating surfaces and spectral features of these classes. The results are discussed considering botanical composition, date (phenology), and vegetation dynamics.

Seasonal LAI in slash pine estimated with Landsat TM

Article

Jan 1992
REMOTE SENS ENVIRON

The leaf area index (LAI, total area of leaves per unit area of ground) of most forest canopies varies throughout the year, yet for logistical reasons it is difficult to estimate anything more detailed than an annual average LAI. To determine if remotely sensed data can be used to estimate LAI at times throughout the year (herein termed seasonal LAI), field measurements of LAI were compared to normalized difference vegetation index (NDVI) values, derived using Landsat Thematic Mapper (TM) data, for 16 fertilized and control slash pine plots on three dates. Linear relationships existed between NDVI and LAI with R2 values of 0.35, 0.75, and 0.86 for February 1988, September 1988and March 1989, respectively. Predictive relationships based on data from eight of the plots were used to estimate the LAI of the other eight plots with a root-mean-square error of 0.74 LAI, which is 15.6% of the mean LAI. This demonstrates the potential use of Landsat TM data for studying seasonal dynamics in forest canopies.

Theory and Applications of Optical Remote Sensing

Article

Jan 1989

G. Asrar

Gran Geografia Comarcal de Catalunya, el Bages, el Berguedà i el Solsonès

Jan 1981
352-355

M Gasol
A Pladevall
A Bach

M. Gasol, A. Pladevall, and A. Bach, Gran Geografia Comarcal de Catalunya, el Bages, el Berguedà i el Solsonès. Barcelona, Spain: Enciclopèdia Catalana, 1981, pp. 352-355.

Gran Geografia Comarcal de Catalunya, el Bages, el Bergueda` i el Solsone`s.

M Gasol
A Pladevall
A Bach

Te&acute,le&acute,de&acute,tection Applique&acute,e

M C Girard
C M Girard

An assessment of canopy chemistry with AVIRIS-A case study in the Landes Forest

Jan 1995
487-501

J P Gastellu-Etchegorry
F Zagolski
E Mougin
G Marty
G Giordano

J. P. Gastellu-Etchegorry, F. Zagolski, E. Mougin, G. Marty, and G. Giordano, "An assessment of canopy chemistry with AVIRIS-A case study in the Landes Forest, South-West France," Int. J. Remote Sensing, vol. 16, pp. 487-501, 1995.

On the reliability of Landsat TM for estimating forest variables by regression techniques: a methodological analysis

Abstract

No full-text available

Recommended publications

Cultivated area estimation combining moderate spatial resolution remote sensing data and MODIS time...

Classification of Spectrally-Similar Land Cover Using Multi-Spectral Neural Image Fusion and the Fuz...

Study on Automatic Extraction of Corn Fields Information on Remotely Sensed Imagery Based on Multi-c...

Global vegetation - Assessing Landsat 7/ETM+ coverage of tropical rainforest and global agricultural...