A logistic model was constructed to assess the risk of forest fire and tested over the central region of Mexico. The model incorporates both static and dynamic predictive variables: elevation, aspect, slope, vegetation type, precipitation, Normalized Difference Vegetation Index (NDVI), land surface temperature (LST), and cloud cover. The latter three variables were derived from National Oceanic and Atmospheric Administration (NOAA) Advanced Very High Resolution Radiometer (AVHRR) images from the four months (November to February) before the fire seasons. Actual forest fires were detected on NOAA-AVHRR images from the four fire seasons (March to May) from 1997 to 2000. Variables included in the model were chosen following a stepwise strategy. Statistically, the January NDVI, the February LST, vegetation type and slope had the greatest influence on the distribution of forest fires; however, elevation and precipitation were also included in the final model. The probability of forest fire occurrence for each fire season from 1997 to 1999 was mapped. The accuracy of the model was estimated to be 79.8% with reference to sensitivity, specificity and receiver operating characteristic (ROC) curves. Model predictions were validated against data from the 2000 fire season. The fire occurrence probability map is useful for designing large-scale management strategies for wildfire prevention not only in the test area of this study but also in regions where the static and dynamic variables can be similarly defined.

Advanced Very High Resolution Radiometer (AVHRR) National Oceanic and Atmospheric Administration (NOAA)-14 imagery was used to analyse changes in land surface temperature in an area of Central Mexico during the course of each dry season (November–April) for the period 1996–2000. Daily surface temperature was obtained by the split-window method and cloud-free monthly composites were subsequently built. This value was related to maximum air temperatures recorded at meteorological stations and to forest fires detected from night-time images. During 1996–1997 and 1997–1998 (El Niño) dry seasons, monthly surface temperature ranged from 35°C to 46°C and from 33°C to 51°C, respectively; during 1998–1999 (La Niña) and 1999–2000 it was lower, ranging from 28°C to 47°C, and from 28°C to 41°C, respectively. At the end of El Niño, land surface temperatures higher than 50°C were registered, and 730 forest fires were detected, suggesting that this temperature increment also contributed to the vulnerability of vegetation to fire. It is concluded that land surface temperature during the first four months of the dry season can be used as a variable for modelling the probability of forest fire occurrence, in combination with other environmental variables. Similarities between land surface temperature and maximum air temperature suggest the potential use of NOAA-AVHRR imagery for evaluating El Niño/La Niña effects on the continental surface.

An inventory of land-cover conditions throughout Mexico was performed using North American Landscape Characterization (NALC) Landsat Multi-Spectral Scanner (MSS) 'triplicate' images, corresponding to the 1970s, 1980s and 1990s epoch periods. The equivalent of 300 image scenes were analysed using an unsupervised classification approach by a consortium of 13 universities and institutes across Mexico. Accuracy assessments were conducted to validate the 1970s and 1990s results using independent land-cover classifications (reference data) developed from the interpretation of 1:100 000-scale aerial photography collected in 1973, and Landsat Thematic Mapper (TM) imagery collected between 1990 and 1993. The 1980s epoch classifications were compared with both reference datasets, collectively. The relative accuracy of the classifications results were 60% for both the 1970s and 1990s epoch and 67% for the 1980s epoch. The significantly ( p =0.05) higher accuracy for the 1980s epoch (67%) was thought to be an aberration resulting from the combined application of two reference datasets, resulting in a random compensation of reference data error. Significantly different ( p =0.05) results were documented for a subset of Mexico's major habitat regions. Desert and xeric shrublands were most accurate (74%), followed by conifer and xeric dominated habitats (64%) and other mixed habitats (54%). Scenes representing the highest accuracies (15 percentile) almost exclusively represented desert and xeric shrub habitat regions, and the lowest (17 percentile) represented predominantly mixed habitat regions. Significant differences among the 13 member consortium universities and institutes were attributed to habitat region assignments. Results indicated that large area spectral based land-cover categorizations should be stratified and processed on a habitat or ecoregion basis. Results also suggested that any future land-cover conversion analysis for Mexico would probably best be accomplished using a post-classification approach, based on major habitat regions, rather than on a scene-by-scene or pixel-wise basis.

A humid forest in the neotropical area of Los Tuxtlas, in southeastern Mexico has been used as a test area (900km²) for classification of landscape and vegetation by means of Landsat Thematic Mapper (TM) data, aerial photography and 103 ground samples. The area presents altitudinal variations from sea level to 1640m, providing a wide variety of vegetation types. A hybrid (supervised/unsupervised) classification approach was used, defining spectral signatures for 14 clustering areas with data from the reflective bands of the TM. The selected clustering areas ranged from vegetation of the highlands and the rain forest to grassland, barren soil, crops and secondary vegetation. The digital classification compared favourably with results from aerial photography and with those from a multivariate analysis of the 103 ground data. The statistical evaluation (error matrix) of the classified image indicated an overall 84·4 per cent accuracy with a kappa coefficient of agreement of 0·83. A geographical information system (GIS) was used to compile a land unit and a vegetation map. The TM data allowed for delineation of boundaries in the land unit map, and for a finer differentiation of vegetation types than those identified during field work. Digital value patterns of several information classes are shown and discussed as an indirect guide of the spectral behaviour of vegetation of highlands, rain forest, secondary vegetation and crops. The method is considered applicable to the inventory of other forested areas, especially those with significant variations in vegetation.