Remote Sensing of Forest Condition in Tolerant Hardwoods: An Examination of Spatial Scale, Structure and Function

Sampson, P. H.; Treitz, Paul; Mohammed, Gina H.

doi:10.1080/07038992.2001.10854940

Cited by 17 publications

(17 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Given that older generation sensors have many known limitations with respect to their suitability for studying complex biophysical characteristics (De Jong et al, 2000;Salas et al, 2002;Sampson et al, 2001;Steininger, 2000), the need to evaluate the new generation of sensors is of critical importance. This is particularly true in the rainforests, where vegetation complexity, spatial variability, diversity, and dynamics are at their highest levels.…”

Section: Introductionmentioning

confidence: 99%

Hyperion, IKONOS, ALI, and ETM+ sensors in the study of African rainforests

Thenkabail

Enclona

Ashton

et al. 2004

Remote Sensing of Environment

274

198

View full text Add to dashboard Cite

Section: Introductionmentioning

confidence: 99%

Hyperion, IKONOS, ALI, and ETM+ sensors in the study of African rainforests

Thenkabail

Enclona

Ashton

et al. 2004

Remote Sensing of Environment

274

198

View full text Add to dashboard Cite

“…The results presented here were considered successful compared with similar RDA-type modelling studies using remote sensing, with the SCI explaining approximately 35% of the variance in the field-collected structural information. In comparison, Sampson et al (2001), using only spectral information extracted from hyperspectral airborne imagery in hardwood forests, did not produce useful relationships with respect to developing a gradient of complexity. Their modelling rested solely on the canopy openness variable and the relationship was lost altogether when clearcut plots were removed.…”

Section: Development Of An Scimentioning

confidence: 81%

“…In a similar type of exploratory analysis, Sampson et al (2001) used redundancy analysis (RDA), which is similar to a combined principal components analysis (PCA) and canonical correlation analysis and the focal method of this paper (see Modelling structural complexity using RDA section). They investigated relationships between forest structural variables (overstorey and understorey canopy openness, crown area, and basal area) and spectral indices calculated from hyperspectral airborne imagery to differentiate between three silviculture practices in a hardwood forest.…”

Section: Introductionmentioning

confidence: 99%

Development of a forest structural complexity index based on multispectral airborne remote sensing and topographic dataThis article is one of a selection of papers from Extending Forest Inventory and Monitoring over Space and Time.

Pasher

King

2011

Can. J. For. Res.

View full text Add to dashboard Cite

This paper presents development of a multivariate forest structural complexity index based on relationships between field-based structural variables and geospatial data. Remote sensing has been widely used to model individual forest structural attributes at many scales. As opposed to, or in addition to, individual structural parameters such as leaf area index or tree height, overall structural complexity information can enhance forest inventories and provide a variety of information to forest managers, including identifying damage and disturbance as well as indicators of habitat or biodiversity. In this study, a multivariate modelling technique, redundancy analysis, was implemented to derive a model incorporating both horizontal and vertical structural attributes as predicted by an ensemble of high-resolution multispectral airborne imagery and topographic variables. The first redundancy analysis axis of the final model explained 35% of the total variance of the field variables and was used as the complexity index. With a root mean squared error of 19.9%, the model was capable of differentiating four to five relative levels of complexity. This paper presents the forest ecological and modelling aspects of the research. A related paper presents the remote sensing aspects, including application of the model to map predicted structural complexity, map validation, and testing of the method at multiple scales.Résumé : Cet article présente le développement d'un indice multivarié de complexité de la structure de la forêt basé sur les relations entre des variables structurales mesurées sur le terrain et des données géospatiales. La télédétection a été largement utilisée pour modéliser les attributs structuraux de la forêt à plusieurs échelles. Contrairement à, ou en plus, des paramètres structuraux individuels comme l'indice de surface foliaire ou la hauteur des arbres, une information globale au sujet de la complexité structurale peut améliorer les inventaires forestiers et fournir une variété de renseignements aux aménagistes forestiers, incluant l'identification des dommages et des perturbations aussi bien que des indicateurs d'habitat ou de biodiversité. Dans cette étude, une technique de modélisation multivariée, l'analyse de redondance, a été appliquée pour dériver un modèle qui incorpore des attributs, tant horizontaux que verticaux de la structure, prédits par un ensemble d'images aériennes multispectrales à haute résolution et des variables topographiques. Le premier axe de l'analyse de redondance du modèle final expliquait 35 % de la variance totale des variables terrain et a été utilisé comme indice de complexité. Avec une erreur quadratique moyenne de 19,9 %, le modèle était capable de différencier quatre à cinq niveaux de complexité. Cet article présente les aspects de la recherche reliés à la modélisation et à l'écologie forestière. Un autre article présente les aspects reliés à la télédétection, incluant l'application du modèle pour cartographier la complexité structurale prédite, la validation de la cartogr...

show abstract

“…We transformed the digital counts (reflectance intensities) of the RGB bands in each season into three b Distribution of survey points in the basin principal components (PCs; Appendix 1). Such trans-formations are often used in remote sensing (e.g., Sampson et al 2001;Mullerova 2004) and were able to represent subtle color differences between leaves, litter, and soil, and to avoid the multicollinearity found The species has only recently been described (Watanabe and Serizawa 2009) among the original RGB data of the same season (r > 0.7). We call the ith PC of an RGB image in the jth month of a year jPCi (e.g., 4PC1 represents the first PC in April).…”

Section: Data Preparationmentioning

confidence: 99%

Comparison of effects of spatial autocorrelation on distribution predictions of four rare plant species in the Watarase wetland

Ishihama

Takeda

Oguma

et al. 2010

Ecological Research

View full text Add to dashboard Cite

We tested the effectiveness of distribution-prediction models for four rare herbaceous wetland species in the Watarase wetland, Japan, based on data obtained from aerial images. We used visible and near-infrared aerial images from three seasons, and elevations and vegetation heights derived from the images. Because spatial autocorrelation in species distribution data often biases the estimated effects of certain variables and reduces the prediction accuracy of distribution models, we compared the predictions of an intrinsic conditional autoregressive (CAR) model, which accounts for spatial autocorrelation, with those of a standard logistic regression model. The four study species had different distribution patterns: Ophioglossum namegatae and Impatiens ohwadae had aggregated distributions, whereas Galium tokyoense and Thalictrum simplex var. brevipes had scattered distributions. Predictions based on remote sensing images performed well for O. namegatae with the intrinsic CAR model and for I. ohwadae with both the logistic and CAR models; performance was poor for G. tokyoense and T. simplex var. brevipes with both models. Prediction accuracy improved by the CAR model in comparison to the logistic model most in O. namegatae and least in I. ohwadae. Impatiens ohwadae's distribution was explained well by ground height. In contrast, the apparent improvement in the prediction for O. namegatae resulted from a substantial spatial random effect, suggesting the presence of determinants that could not be detected by remote sensing. The number of explanatory variables with large effects decreased in the intrinsic CAR model in three species possibly by avoiding spatial pseudoreplication, but not for T. simplex var. brevipes.

show abstract

Remote Sensing of Forest Condition in Tolerant Hardwoods: An Examination of Spatial Scale, Structure and Function

Cited by 17 publications

References 36 publications

Hyperion, IKONOS, ALI, and ETM+ sensors in the study of African rainforests

Hyperion, IKONOS, ALI, and ETM+ sensors in the study of African rainforests

Development of a forest structural complexity index based on multispectral airborne remote sensing and topographic dataThis article is one of a selection of papers from Extending Forest Inventory and Monitoring over Space and Time.

Comparison of effects of spatial autocorrelation on distribution predictions of four rare plant species in the Watarase wetland

Contact Info

Product

Resources

About