G.A. Sanchez-Azofeifa

Satellite Remote Sensing of Land Use Changes in and around San José, Costa Rica

Abstract Using AVHRR imagery, the effects of urbanization on surface microclimate during a 7-year period in San Jose, Costa Rica and its surroundings are expressed as changes in fractional vegetation cover, scaled surface temperature, surface wetness, and surface evapotranspiration. Scatterplots of fractional vegetation cover versus scaled surface temperature depict elongated clusters along which pixels undergoing urbanization migrate toward lower vegetation cover, higher surface temperature, less evapotranspiration, and generally dryer surfaces. This urbanization occurred largely in response to pasture and low-density urban areas changing to high-density urban areas. The evolution in surface microclimate accompanying these changes differs considerably for the various types of urban surfaces, but ultimately the pixels become virtually indistinguishable from one another as urbanization becomes advanced. On the other hand deforestation, which occurred as trees were replaced by crops, did not result in large changes in surface temperature and wetness and may have produced increases with time in evapotranspiration.

Estimating leaf area index from satellite imagery using Bayesian networks

In this study, we investigated the use of Bayesian networks for inferring tropical dry forest leaf area index (LAI) from satellite imagery in dry and wet seasons. LAI was chosen as the variable of interest because leaf area is the exchange surface between the photosynthetically active component of the canopy and the atmosphere. Initial network estimates were obtained from ground truth plot data with known forest structure, LAI, and satellite reflectance in the red and near-infrared bands (as observed by the Landsat 7 Enhanced Thematic Mapper Plus sensor). We tested the performance of the Bayesian networks with scoring rules and also with confidence and surprise scores. We evaluated the networks on a per-pixel basis and created both LAI maps of the study area as well predicted the probability maps for the highest LAI states. Results not only demonstrate the predictive power of a Bayesian network but also its explanatory power which is far beyond what is typically available with current pixel classifier approaches such as spectral vegetation indices or other approaches such as neural networks.