HR: 0800h
AN: B41B-05    [Abstracts]
TI: Mapping the Distribution of Cloud Forests Using MODIS Imagery
AU: * Douglas, M W
EM: Michael.Douglas@noaa.gov
AF: National Severe Storms Laboratory/NOAA, 120 David L. Boren Blvd, Norman, OK 73072, United States
AU: Mejia, J
EM: John.Mejia@noaa.gov
AF: CIMMS/University of Oklahoma, 120 David L. Boren Blvd, Norman, OK 73072, United States
AU: Murillo, J
EM: Javier.Murillo@noaa.gov
AF: CIMMS/University of Oklahoma, 120 David L. Boren Blvd, Norman, OK 73072, United States
AU: Orozco, R
EM: Raquel.Orozco@noaa.gov
AF: CIMMS/University of Oklahoma, 120 David L. Boren Blvd, Norman, OK 73072, United States
AB: Tropical cloud forests - those forests that are frequently immersed in clouds or otherwise very humid, are extremely difficult to map from the ground, and are not easily distinguished in satellite imagery from other forest types, but they have a very different flora and fauna than lowland rainforest. Cloud forests, although found in many parts of the tropics, have a very restricted vertical extent and thus are also restricted horizontally. As a result, they are subject to both human disturbance (coffee growing for example) and the effects of possible climate change. Motivated by a desire to seek meteorological explanations for the distribution of cloud forests, we have begun to map cloudiness using MODIS Terra and Aqua visible imagery. This imagery, at ~1030 LT and 1330 LT, is an approximation for mid-day cloudiness. In tropical regions the amount of mid-day cloudiness strongly controls the shortwave radiation and thus the potential for evaporation (and aridity). We have mapped cloudiness using a simple algorithm that distinguishes between the cloud-free background brightness and the generally more reflective clouds to separate clouds from the underlying background. A major advantage of MODIS imagery over many other sources of satellite imagery is its high spatial resolution (~250m). This, coupled with precisely navigated images, means that detailed maps of cloudiness can be produced. The cloudiness maps can then be related to the underlying topography to further refine the location of the cloud forests. An advantage of this technique is that we are mapping the potential cloud forest, based on cloudiness, rather than the actual cloud forest, which are commonly based on forest estimates from satellite and digital elevation data. We do not derive precipitation, only estimates of daytime cloudiness. Although only a few years of MODIS imagery has been used in our studies, we will show that this is sufficient to describe the climatology of cloudiness with acceptable accuracy for its intended purposes. Even periods as short as one month are sufficient for depicting the location of most cloud forest environments. However, we are proceeding to distinguish different characteristics of cloud forests, depending on the overall frequency of cloudiness, the seasonality of cloudiness, and the interannual variability of cloudiness. These results should be useful to those seeking to describe relationships between the physical characteristics of the cloud forests and their biological environment.
UR: http:www.nssl.noaa.gov/projects/pacs/web/MODIS/
DE: 0410 Biodiversity
DE: 0416 Biogeophysics
DE: 0426 Biosphere/atmosphere interactions (0315)
DE: 1640 Remote sensing (1855)
DE: 3374 Tropical meteorology
SC: Biogeosciences [B]
MN: 2007 Joint Assembly