GEMI: a non-linear index to monitor global vegetation from satellites

Abstract

Knowledge about the state, spatial distribution and temporal evolution of the vegetation cover is of great scientific and economic value. Satellite platforms provide a most convenient tool to observe the biosphere globally and repetitively, but the quantitative interpretation of the observations may be difficult. Reflectance measurements in the visible and near-infrared regions have been analyzed with simple but powerful indices designed to enhance the contrast between the vegetation and other surface types, however, these indices are rather sensitive to atmospheric effects. The 'correction' of satellite data for atmospheric effects is possible but requires large data sets on the composition of the atmosphere. Instead, we propose a new vegetation index which has been designed specifically to reduce the relative effects of these undesirable atmospheric perturbations, while maintaining the information about the vegetation cover.Terrestrial vegetation, which constitutes the bulk of the continental biomass (Ajtay et al. 1979), affects the climate system over a wide range of space and time scales by modifying the surface energy balance, and by influencing the exchanges of water and carbon between the surface and the atmosphere. The identification of green plant material is of great interest since leaves and needles are the site of photosynthesis and the prime link between the biosphere and the atmosphere.

Development of a chemiluminescence-based quantitative lateral flow immunoassay for on-field detection of 2,4,6-trinitrotoluene

“…Estas variables miden las diferencias de las medianas de los índices GEMI (Pinty et al, 1992), BBFI e infrarrojo medio (NIR) entre el periodo post-fuego del año del incendio y el año anterior, así como el valor máximo del índice BBFI del año anterior al incendio.…”

“…Umbral relativo de la diferencia entre la mediana del GEMI (Pinty et al, 1992) Una vez elegido el mejor modelo de clasificación atendiendo al error que se comete en la clasificación del propio conjunto de entrenamiento (mediante un proceso de refinamiento de casos de entrenamiento y/o inserción/eliminación de variables), éste se aplica sobre el conjunto de datos en su totalidad, obteniéndose un mapa de probabilidades: cada variable tiene una probabilidad asociada dependiendo de la clase a la que pertenece (quemado, no-quemado). El mapa de probabilidades se obtiene al normalizar la probabilidad conjunta de todas las variables (tomando como positivas las que pertenezcan a la clase quemado y como negativas a la clase no quemado).…”

Sentinel-2’s Potential for Sub-Pixel Landscape Feature Detection

“…The foreground classes have been chosen within a set of key structuring landscape objects described in Section 2.1.1. The separability analysis, described in Section 2.1.2, has been performed for the spectral bands, for common spectral indices used to enhance the discrimination as well as for new indices based on the less common spectral bands available from Sentinel-2 (Table 2 [ [39][40][41][42][43][44][45][46][47][48][49][50][51][52][53][54][55][56][57][58]). For this purpose, pure spectral signatures of objects have been extracted from the images at well known locations (see Section 3).…”

Global terrestrial biogeochemistry: Perturbations, interactions, and time scales

“…Although the relationship between NDVI and fAPAR is generally linear, it is well-known to be sensitive to the status of many soil-vegetation-atmosphere system parameters that exhibit high spatial and/or temporal variability (Goward and Huemmrich 1992). Thus, a one-to-one relationship between a VI and a biophysical variable may not be generally applicable, except through careful consideration of complicating factors; by inventing indices that are less sensitive to atmospheric or soil influences (Huete 1988, Pinty andVerstraete 1992), by normalizing the index (Roujean and Breon 1995), or by selecting pixels with preferred view angles (Cihlar et al 1994). Such considerations (especially the latter two) potentially involve the iterative use of radiative transfer (RT) models which is similar, in principle, to the method of inversion.…”