SCAVI: A Sunlit Canopy Adjusted Vegetation Index

Abstract. The new Sunlit Canopy Adjusted Vegetation Index (SCAVI) uses subpixel scale spectral mixture analysis (SMA) principles for improved biophysical parameter estimation. SCAVI and a new NDVI-modified extension (SCAVI+N) were formulated after soil-adjusted vegetation index (VI) equations and tested using NASA COVER airborne multispectral data for boreal forest black spruce stands in Superior National Forest, Minnesota, USA. Fifteen VIs and 3 SMA fractions were compared. SCAVI was the top-ranked VI for each of leaf area index (LAI: r2 = 0.72), net primary productivity (NPP: 0.72), and biomass (BIO: 0.63), with an overall r2 = 0.69 being >10% higher than the next-ranked VI. Shadow-adjusted vegetation indices (SHAVI, SHAVI+N) were also formulated but had low predictive capabilities. The best result of all variables was from SMA shadow fraction with overall r2 = 0.78 (LAI 0.79, NPP 0.80, BIO 0.74). The importance of endmember-based analysis was clear, because these occupied the top 6 of the 18 rankings. In the absence of SMA or physically based canopy reflectance modeling (CRM), SCAVI might represent a preferred VI and offers advantages of computational simplicity and involving only 1 endmember. Recommendations included testing SCAVI for other areas, refinement of equations, developing other Endmember-Adjusted Vegetation Indices (EAVIs), and the integration of VI, SMA, and CRM concepts. Résumé. Le nouvel indice de végétation ajusté pour le couvert forestier ensoleillé «Sunlit Canopy Adjusted Vegetation Index» (SCAVI) utilise des principes de l’analyse de mélange spectral «spectral mixture analysis» (SMA) à l’échelle du sous-pixel pour une meilleure estimation des paramètres biophysiques. SCAVI et une nouvelle extension modifiée du NDVI (SCAVI+N) ont été formulés à partir des équations de l'indice de végétation «vegetation index» (VI) ajusté pour le sol et testés à l'aide des données multispectrales aéroportées de NASA COVER pour des peuplements d'épinettes noires dans la forêt boréale, dans Superior National Forest, Minnesota, États-Unis. Quinze VIs et 3 fractions de SMA ont été comparés. Le SCAVI a été classé au premier rang des VIs pour l'indice de surface foliaire «leaf area index» (LAI: r2 = 0,72), la productivité primaire nette «net primary productivity» (NPP: 0,72) et la biomasse (BIO: 0,63), avec un r2 global de 0.69 qui était >10 % plus élevé que le deuxième VI. Des indices de végétation ajustés pour les ombres (SHAVI, SHAVI+N) ont également été formulés mais ont montré des capacités prédictives faibles. Le meilleur résultat de toutes les variables était pour la fraction d'ombre SMA avec un r2 global de 0.78 (LAI 0,79; NPP 0,80; BIO 0,74). L'importance d’une analyse fondée sur des endmembers fut claire, car les analyses qui en incluaient ont obtenu les 6 premiers rangs sur 18. En l'absence de la SMA ou de modélisation de la réflectance du couvert végétal à base physique «canopy reflectance modeling» (CRM), SCAVI peut représenter un VI préféré et offre les avantages de la simplicité de calcul et de l’utilisation d’un seul endmember. Les recommandations incluent de tester SCAVI pour d’autres régions, le raffinement d'équations, le développement d'autres indices de végétation ajustés pour les endmembers «Endmember-Adjusted Vegetation Indices» (EAVIs), et l'intégration des concepts de VI, SMA, et CRM.

[1]  J. Roujean,et al.  Estimating PAR absorbed by vegetation from bidirectional reflectance measurements , 1995 .

[2]  Nicholas C. Coops,et al.  Generation of a novel 1 km NDVI data set over Canada, the northern United States, and Greenland based on historical AVHRR data , 2012 .

[3]  A. Huete,et al.  Overview of the radiometric and biophysical performance of the MODIS vegetation indices , 2002 .

[4]  Scott J. Goetz,et al.  Biophysical, morphological, canopy optical property, and productivity data from the Superior National Forest , 1992 .

[5]  Derek R. Peddle,et al.  Forest structure without ground data: Adaptive Full-Blind Multiple Forward-Mode reflectance model inversion in a mountain pine beetle damaged forest , 2010 .

[6]  A. Huete,et al.  A review of vegetation indices , 1995 .

[7]  S. T. Gower,et al.  Leaf area index of boreal forests: theory, techniques, and measurements , 1997 .

[8]  C. Jordan.  Derivation of leaf-area index from quality of light on the forest floor , 1969 .

[9]  R. Fournier,et al.  Estimating stand attributes of boreal forests using digital aerial photography and a shadow fraction method , 2013 .

[10]  Derek R. Peddle,et al.  A Comparison of Spectral Mixture Analysis and Ten Vegetation Indices for Estimating Boreal Forest Biophysical Information from Airborne Data , 2001 .

[11]  J. Chen.  Evaluation of Vegetation Indices and a Modified Simple Ratio for Boreal Applications , 1996 .

[12]  D. Peddle.  Spectral Mixture Analysis and Geometric-Optical Reflectance Modeling of Boreal Forest Biophysical Structure , 1999 .

[13]  B. Pinty,et al.  GEMI: a non-linear index to monitor global vegetation from satellites , 1992, Vegetatio.

[14]  C. Tucker.  Red and photographic infrared linear combinations for monitoring vegetation , 1979 .

[15]  Adrian V. Rocha,et al.  Advantages of a two band EVI calculated from solar and photosynthetically active radiation fluxes , 2009 .

[16]  F. Baret,et al.  About the soil line concept in remote sensing , 1993 .

[17]  J. Clevers.  Application of a weighted infrared-red vegetation index for estimating leaf Area Index by Correcting for Soil Moisture , 1989 .

[18]  Xuexia Chen,et al.  Using lidar and effective LAI data to evaluate IKONOS and Landsat 7 ETM+ vegetation cover estimates in a ponderosa pine forest , 2004 .

[19]  Karl Fred Huemmrich,et al.  Remote Sensing of Forest Biophysical Structure Using Mixture Decomposition and Geometric Reflectance Models , 1995 .

[20]  Richard A. Fournier,et al.  Application of shadow fraction models for estimating attributes of northern boreal forests , 2012 .

[21]  A. Huete.  A soil-adjusted vegetation index (SAVI) , 1988 .

[22]  Ian Olthof,et al.  Modelling and Mapping Damage to Forests from an Ice Storm Using Remote Sensing and Environmental Data , 2005 .

[23]  A. Huete,et al.  Development of a two-band enhanced vegetation index without a blue band , 2008 .

[24]  D. Peddle,et al.  Spectral mixture analysis of agricultural crops: Endmember validation and biophysical estimation in potato plots , 2005 .

[25]  N. Goel,et al.  Influences of canopy architecture on relationships between various vegetation indices and LAI and Fpar: A computer simulation , 1994 .

[26]  Yosio Edemir Shimabukuro,et al.  The least-squares mixing models to generate fraction images derived from remote sensing multispectral data , 1991, IEEE Trans. Geosci. Remote. Sens..

[27]  D. Roberts,et al.  Endmember selection for multiple endmember spectral mixture analysis using endmember average RMSE , 2003 .

[28]  Douglas J. King,et al.  Shadow brightness and shadow fraction relations with effective leaf area index: importance of canopy closure and view angle in mixedwood boreal forest , 2003 .

[29]  R. Latifovic,et al.  Large area forest classification and biophysical parameter estimation using the 5-Scale canopy reflectance model in Multiple-Forward-Mode , 2004 .

[30]  J. A. Schell,et al.  Monitoring the Vernal Advancement and Retrogradation (Green Wave Effect) of Natural Vegetation. [Great Plains Corridor] , 1973 .

[31]  C. Perry,et al.  Functional equivalence of spectral vegetation indices , 1984 .

[32]  A. Huete,et al.  A Modified Soil Adjusted Vegetation Index , 1994 .

[33]  Ranga B. Myneni,et al.  The interpretation of spectral vegetation indexes , 1995, IEEE Transactions on Geoscience and Remote Sensing.

[34]  Shunlin Liang,et al.  Advances in Land Remote Sensing , 2008 .