Quantifying the extent and degree of cloud-affected tropical environments with MODIS imagery. Two extreme environments: Lomas and cloud forests

Using a simple algorithm for extracting cloudiness from the visible MODIS imagery we have developed short-period (~ several years) cloudiness climatologies for different tropical regions at 250m spatial resolution. This presentation focuses on two environments that are difficult to accurately delineate with conventional climate data or even current satellite-based rainfall estimation techniques. One such environment, tropical cloud forest, is characterized by very high precipitation and also high cloudiness. However, cloud forest is mostly distinguished from surrounding lowland rain forest by the very high frequency of cloudiness and its small spatial extent. The second environment of our study is the coastal fog/low cloud zone along arid coastlines (known as “lomas” in Peru). These regions receive almost no rainfall, yet have vegetation supported by the frequent low clouds that intercept the topography along the coast. Such areas are even more difficult to identify from raingauges or satellite rainfall estimates than cloud forests.

We have developed a simple algorithm to use the cloudiness frequencies obtained from the MODIS imagery to classify the “intensity” of both the cloud forests and the lomas. The algorithm uses MODIS imagery from the NASA Terra and Aqua satellites to stratify cloudiness by annual amount, seasonality, and diurnal variability. Areas most favorable for vegetation growth are those with maximum annual frequency of cloudiness and minimum seasonality and diurnal variation, other factors being equal.