Mapping Forest Cover Change during Cuba's Special Period & Beyond
Conference Year
January 2019
Abstract
During the second half of the 20th Century, Cuba received Soviet Union imports such as fertilizer and fuel in support of agriculture. The Soviet Union further bolstered the Cuban economy as the main purchaser of Cuban exports. The collapse of the Soviet Union in 1991 had profound and immediate impacts on agriculture in Cuba, starting a period of hunger and resource scarcity known as the “Special Period.” The impacts of the Special Period on human wellbeing and agricultural production are well known and there are generalized reports of agricultural abandonment. The extent and location of abandonment and subsequent forest regrowth, however, have not previously been documented.
We use Google Earth Engine to create a time series of land cover Cuba from 1985 to 2010, in five year intervals. Cuba is an extremely cloud-prone region, so assessing forest cover can be difficult. We use a multi-sensor composite to obtain cloud free images using a combination of Landsat 4, 5, 7, and 8, depending on the availability of satellites for particular years. However, this method becomes challenging when creating a time series because early years have fewer available satellites to utilize than later years. It is unclear to what extent images created using one satellite can be fairly compared to images composited with multiple satellites. Here we present a robustness check to determine how multisensor compositing effects uncertainty in forest cover detections. Overall forest cover increased due to agricultural abandonment. In some regions, nearly 30% increase in forest cover is observed between 1985-2010, much greater than expected.
Primary Faculty Mentor Name
Professor Gillian Galford
Graduate Student Mentors
Eva Kinnebrew
Status
Undergraduate
Student College
College of Arts and Sciences
Program/Major
Environmental Sciences
Primary Research Category
Engineering & Physical Sciences
Mapping Forest Cover Change during Cuba's Special Period & Beyond
During the second half of the 20th Century, Cuba received Soviet Union imports such as fertilizer and fuel in support of agriculture. The Soviet Union further bolstered the Cuban economy as the main purchaser of Cuban exports. The collapse of the Soviet Union in 1991 had profound and immediate impacts on agriculture in Cuba, starting a period of hunger and resource scarcity known as the “Special Period.” The impacts of the Special Period on human wellbeing and agricultural production are well known and there are generalized reports of agricultural abandonment. The extent and location of abandonment and subsequent forest regrowth, however, have not previously been documented.
We use Google Earth Engine to create a time series of land cover Cuba from 1985 to 2010, in five year intervals. Cuba is an extremely cloud-prone region, so assessing forest cover can be difficult. We use a multi-sensor composite to obtain cloud free images using a combination of Landsat 4, 5, 7, and 8, depending on the availability of satellites for particular years. However, this method becomes challenging when creating a time series because early years have fewer available satellites to utilize than later years. It is unclear to what extent images created using one satellite can be fairly compared to images composited with multiple satellites. Here we present a robustness check to determine how multisensor compositing effects uncertainty in forest cover detections. Overall forest cover increased due to agricultural abandonment. In some regions, nearly 30% increase in forest cover is observed between 1985-2010, much greater than expected.