Search

2003 - International Geoscience and Remote Sensing Symposium, IGARSS'03. Toulouse (France), 21-25/7/2004, This investigation quantitatively links physiologically based estimators of forest stand condition, such as chlorophyll concentration, to hyperspectral observations of Jack Pine (Pinus Banksiana), a dominant Boreal Forest species. Between June and September of 2001, four Intensive Field Campaigns (IFC) of data collection were conducted over the forested areas near Sudbury, Ontario, Canada. Using the CASI sensor, data were collected, in the visible and near infrared domain, over eight selected Jack Pine sites. Supplementing the airborne campaigns, was simultaneous onsite collection of foliage samples for laboratory spectral and chemical measurements. The study first linked needle-level reflectance and pigment content through the inversions of leaf level radiative transfer models such as PROSPECT. Next, the red-edge index (R750/710), was scaled up to the canopy level through the use of canopy models and infinite reflectance calculations, which simulate the canopy as an optically thick vegetation medium. However, for the relatively open and clumped jack pine stands such a simple approach requires careful validation due to the confounding effects of the open canopy structure. Accordingly, the analysis has focused on high spatial resolution CASI imagery (1 meter) for which tree crowns, shadows, and open (sun-lit) understory can be identified visually and approaches can be examined for validity and effects. Effectively eliminating these confounding variables will permit the generation of predictive needle pigment content maps for forest condition assessment.

To develop practical and objective measures of forest condition, the Bioindicators of Forest Sustainability Project has used a physiological, remote sensing approach that emphasizes identifying early warning measures of stress effects in forests. While stress indicators exist at the leaf level (e.g., chlorophyll fluorescence, pigment levels), developing reliable indicators at the canopy level is a challenge. Hyperspectral sensors, such as the Compact Airborne Spectrographic Imager (CASI), may be useful in remotely detecting vegetation stress effects. In this study, an inverse modeling approach demonstrated that the CASI could be used to map chlorophyll content (root mean square errors ranging from 12.6 to 13.0 mg/cm2) following different silvicultural treatments in a tolerant hardwood (sugar maple [Acer saccharum M.]) forest. This capability could be readily applied to operationally assessing forest physiological strain and in classifying forest condition based on chlorophyll content. A change analysis study was also conducted to evaluate chlorophyll estimation across seasons for a range of sites. The implications of these findings and recommendations for a prototype system to monitor forest condition are presented., The authors gratefully acknowledge the financial support provided for this research through the Ontario Ministry of Natural Resources, the Centre for Research in Earth and Space Technology (CRESTech), and Geomatics for Informed Decisions (GEOIDE)—a Canadian Centre of Excellence.

XII World Forestry Congress, Quebec, Canada, 21-28th September, 2003, The Bioindicators of Forest Sustainability Project has applied a physiological, remote sensing approach to develop practical and objective measures of forest condition,. This project was designed to address ecosystem condition and productivity (C&I, Criteria Number 2), by producing an indicator for disturbance and stress. While stress indicators like chlorophyll fluorescence and pigment content exist at the leaf level, developing reliable indicators at the canopy level is a challenge. In this study, an inverse modelling approach has demonstrated the capability of hyperspectral sensor (compact airborne spectrographic imager (CASI)) reflectance images to map chlorophyll content in 12 stands of a tolerant hardwood sugar maple (Acer saccharum) forest in the Algoma region of Ontario, Canada that vary in condition from healthy to chronically stressed. The practical significance of developing spectral features related to chlorophyll concentration is in identifying whether forests are healthy or stressed. Temporal variations in chlorophyll concentrations could provide an objective, early-warning indicator of stress. The Bioindicators Project has expanded to include the boreal forest species jack pine (Pinus banksiana), black spruce (Picea mariana), trembling aspen (Populus tremuloides), and white birch (Betula papyrifera). Preliminary results suggest that chlorophyll concentrations of jack pine, aspen, and white birch canopies can be estimated from CASI images. Current efforts are focussing on scaling this technique up from the high spatial resolution of CASI images to lower spatial resolution, but larger image size, of satellite images of the hyperion sensor of NASA’s Earth Observing-1 satellite. Once developed, this technique could be used as an efficient means to operationally assess both acute and chronic forest physiological stress and classify forest condition based on chlorophyll content. Forest condition maps developed using this technology could be used for state of the resource reporting, assessing the effects of silvicultural operations, and as indicators of incipient insect and disease outbreaks.

Presented at the 2001 AVIRIS Workshop, JPL-NASA, Pasadena, California, Feb. 27th - March 2nd, 2001., This paper reports on progress made within the Bioindicators of Forest Sustainability Project (Mohammed et al., 1997; Sampson et al., 1998) to develop links between physiologically-based bio-indicators (e.g., pigment concentrations, chlorophyll fluorescence) from field and laboratory data and optical indices from hyperspectral remote sensing data for assessing forest condition.

Presented at the Third North American Forest Ecology Workshop, Duluth, Minnesota (USA), 24th-27th June, 2001, To develop practical and objective measures of forest condition, the Bioindicators of Forest Condition Project has applied a physiological, remote sensing approach. While stress indicators at the leaf-level exist (e.g., chlorophyll fluorescence, and pigment content), developing reliable indicators at the canopy level is a challenge. Hyperspectral sensors, such as the Compact Airborne Spectrographic Imager (CASI), may be useful in remote detection of vegetation stress effects. In this study, an inverse modeling approach demonstrated the capability of the CASI to map chlorophyll content (root mean square errors ranging from 12.6 to 13.0 mg/cm2) following different silvicultural practices in a tolerant hardwood (Acer saccharum M.) forest. This capability could be readily applicable to operational assessment of forest physiological strain, and in classification of forest condition based on chlorophyll content. The practical significance of developing spectral features related to chlorophyll or other pigments is in identifying whether forests are healthy or stressed (to the point where productivity of the resource may be constrained). A change analysis study was also conducted to evaluate chlorophyll estimation across seasons for a range of sites. Temporal variations in pigment concentrations (e.g. chlorophyll) could provide an objective, earlywarning indicator. The implications of these findings and recommendations for a prototype system to monitor forest condition are presented.

Presented at the IX Congreso Nacional de Teledetección, Asociación Espanola de Teledetección, Lerida (Spain), 19-21th September, 2001, [ES] El presente trabajo describe los resultados obtenidos en la estimación de contenido clorofílico mediante inversión de modelos de transferencia radiativa. El estudio se desarrolló en 17 zonas de estudio de dos especies forestales con grandes diferencias estructurales, Acer saccharum M. y Pinus strobus L. entre 1997 y 2000 en Ontario, Canadá, estimando chla+b mediante inversión de PROSPECT, SAILH y SPRINT. El sensor hiperespectral aerotransportado CASI se utilizó para obtener datos de reflectancia a 0.5 y 2 m de resolución espacial, con 72 y 288 bandas en el intervalo 400-950 nm. Datos de LAI, contenido clorofílico, carotenoides, reflectancia y transmitancia fueron obtenidos en las muestras foliares de acículas y caducifolias en cada una de las zonas de estudio., [EN] This paper reports on progress made to estimate leaf chlorophyll content by radiative transfer model inversion techniques. The study was carried out at seventeen sites of Pinus strobus L. and Acer saccharum M. in Ontario, Canada, where field measurements and hyperspectral CASI imagery were collected between 1997 and 2000 deployments. PROSPECT, SAILH, and SPRINT models were used by numerical model inversion. Individual tree samples were collected at each site for biochemical analysis and measurement of leaf chlorophyll, carotenoid concentrations, as well as reflectance and transmittance of broad leaves and needles. The model inversion techniques were tested using CASI data of 72 and 288 channels and 0.5 and 2 m spatial resolution.

[ES] El presente trabajo demuestra la utilización de modelos de simulación de la cobertura forestal mediante su aplicación a datos hiperespectrales del sensor aerotransportado CASI. Los modelos SAIL y Kuusk permiten ser utilizados como nexo de unión entre los niveles de hoja y de cobertura: las relaciones a nivel de hoja obtenidas entre índices ópticos y bioindicadores de estrés, como contenido clorofílico o fluorescencia clorofílica, pueden ser transformadas a un nivel superior de cobertura mediante la utilización de dichos modelos. Finalmente se realiza una demostración de la utilización de modelos de cobertura a través de los resultados obtenidos en el proyecto Bioindicators of Forest Sustainability, desarrollado en 12 zonas de Acer saccharum M. localizadas en Ontario (Canadá) donde se obtuvieron medidas de campo de muestras foliares, así como datos hiperespectrales del sensor aerotransportado CASI en 1997, 1998 y 1999. Los indices ópticos desarrollados a nivel de hoja fueron aplicados, a través de modelos de cobertura, a los datos de reflectancia obtenidos por CASI de 2 m de resolución espacial y 72 bandas, [EN] This paper demonstrates the use and applications of Canopy Reflectance Models (CR) with airborne hyperspectral CASI data. SAIL and Kuusk canopy reflectance models are the link between the leaf and canopy levels: leaf-level relationships obtained between optical indices and stress bioindicators, such as chlorophyll content and chlorophyll fluorescence can be scaled-up to the canopy level using canopy reflectance models. The application of canopy reflectance models is demonstrated with the results obtained in the Bioindicators of Forest Sustainability Project. The work was carried out in 12 study areas of Acer saccharum M. in the Algoma Region, Ontario (Canada), where field measurements and hyperspectral CASI imagery have been collected in 1997, 1998 and 1999 deployments. Single leaf reflectance and transmittance, chlorophyll and carotenoid content, and chlorophyll fluorescence of broad leaves were measured. The physiological indices and derivative analysis indices extracted from leaf spectral reflectance were tested at canopy level using CASI data of 72 channels and 2 m spatial resolution.

This study explores some of the biological factors influencing leaf-based spectral reflectance, and endeavours to answer the questions: (1) Can Ontario plant species be satisfactorily distinguished by their leaf spectra? (2) Can spectral properties be used to identify physiological strain in plants before visual symptoms appear? and (3) How strongly do factors other than species and stress affect spectral properties? Forty-four species were examined for species effects. Some general patterns were revealed, including: low reflectance in the near infrared region for conifers compared to broadleaved species; higher reflectance in the blue wavebands for species with blue hues in their foliage; and more rapid decay of the green spectral peak in deciduous tree and shrub species. However, the influence of species was easily superceded by other factors, such as leaf age, leaf side, and stress status. The effects of stresses such as senescence and herbicide on leaf reflectance were evident prior to the appearance of visual symptoms such as chlorosis, browning, and reduced growth. The application of spectral indices was a useful means to quantify previsual changes in spectral reflectance. The red edge inflection point was well correlated with chlorophyll content, confirming previous studies. Pooling a range of coniferous species and stock types, the Photochemical Reflectance Index (PRI) was a promising early indicator of strain which may have broad application; within species, additional indices were useful. Normalized difference vegetation indices (NDVI) had limited correlation to pigments and other physiological indicators of stress status. Altogether, 11 indices reported in the literature were tested. These findings suggest that leaf reflectance may be an effective early indicator of physiological strain, and underpin our efforts to develop monitoring tools at remote scales to aid sustainable forest management.

Presented at the VIII Congreso Nacional de Teledetección, Asociación Espanola de Teledetección, Albacete (Spain), 22-24th September, 1999., [ES] El presente trabajo es un avance del proyecto Bioindicators of Forest Sustainability que persigue el desarrollo de índices ópticos relacionados con indicadores fisiológicos mediante teledetección hiperespectral. El estudio se desarrolla en 17 zonas de Pinus strobus L., Pinus banksiana Lamb. y Acer saccharum M. localizadas en Ontario (Canadá) donde se obtuvieron medidas de campo de muestras foliares así como datos hiperespectrales del sensor aerotransportado CASI en 1997, 1998 y 1999. Datos de índice de área foliar e índice de cobertura así como medidas de contenido clorofílico, fluorescencia clorofílica, carotenoides, reflectancia y transmitancia fueron obtenidos de cada muestra foliar en cada una de las zonas de estudio. Indices ópticos desarrollados a nivel de hoja fueron aplicados a los datos de reflectancia obtenidos por CASI de 2 m de resolución espacial y 72 bandas., [EN] This paper reports on progress made to link physiologically-based indicators to optical indices from hyperspectral remote sensing. The study is being carried out at seventeen sites of Pinus strobus L., Pinus banksiana Lamb. and Acer saccharum M. in the Algoma Region, Ontario (Canada), where field measurements and hyperspectral CASI imagery have been collected in 1997, 1998 and 1999 deployments. Crown cover, LAI and individual tree samples were collected at each site for biochemical analysis and measurement of leaf chlorophyll, chlorophyll fluorescence and carotenoid concentrations, as well as reflectance and transmittance of broad leaves and needles. The physiological indices and derivative analysis indices extracted from leaf spectral reflectance have been tested at canopy level using CASI data of 72 channels and 2 m spatial resolution.

Measurable indicators are needed to gauge the effects of management activities and natural phenomena on forest sustainability. To meet this need, the Bioindicators of Forest Sustainability Project is testing physiological approaches to develop a Forest Condition Rating (FCR) system. An FCR system would directly support provincial policy (e.g., Forest Resource Assessment Policy) and national initiatives (e.g., Criteria and Indicators) by providing an understanding of ecosystem condition. Furthermore, this project addresses a pressing need for indicators that can support operational forest management and possible concerns of sustainability at the local level. Development of an FCR system involves interpreting remotely sensed spectral information with the aid of ground-based assessments of structural and functional (i.e., physiological) aspects of forest condition. Analysis of this spectral information may reveal indicators of health across a wide range of tree species and ages. Current research activities include controlled laboratory studies, ground-based field assessments and acquisition of hyperspectral airborne data to develop gradients in key spectral features that correspond to actual differences in physiology. This report provides first-year progress results. Preliminary correlations in controlled experiments have been made between leaf-based spectral reflectance and physiological status. Compilation of a leaf-based spectral database has been initiated. The database shows the influence of species, leaf age, stress status, season and other factors on spectral features. Other physiological and structural measures, such as foliar biochemistry, stem electrical resistance, and leaf area index, have also been linked to forest decline status. Finally, a brief summary of collaborative projects and proposed research activities is provided., Funding for the project is being provided by the Ontario Ministry of Natural Resources, with in-kind contributions from the Centre for Research in Earth and Space Technology (CRESTech), the Canadian Forest Service and the Ministry of Environment and Energy.