Fig 1.
Examples of the categories of change identified within the color composite imagery.
Changes in mangrove extent were identified through a combination of their color, their shape and the context of the surrounding environment in the composite imagery. a) intact mangroves in Papua, Indonesia b) prior disturbance (aquaculture) at Guayaquil, Ecuador c) loss of mangrove along the coastline of French Guiana d) colonization of mangrove along the French Guiana coastline e) extensive aquaculture at the Mahakam Delta, East Kalimantan, Indonesia f) mangrove dieback in West Papua, Indonesia g) logging within the managed Matang forest reserve, Perak, Malaysia h) prior and on-going agriculture in Sumatra. Imagery copyright of JAXA.
Fig 2.
Distribution of different drivers of change in mangrove forest extent across the tropics.
A) Advance and regrowth of mangrove extent (1996-2010) B) Degradation from anthropogenic drivers of change including evidence of prior disturbance C) Hotspots where substantial changes in mangrove forest extent were observed (1996-2010) D) Tiles that contained intact mangrove (1996-2010). The total distribution of mangrove tiles is provided in gray.
Table 1.
Global mangrove forest change distribution and frequency of mangrove forest change (gain and loss) 1996-2010 by region, as a percentage of the total number of occurrences of change observed globally.
NA = North America (including Caribbean), SA = South America, MEI = Middle East and India, SE Asia = Southeast Asia.
Table 2.
Regional mangrove forest change distribution and frequency of mangrove forest change (gain and loss) 1996-2010 as a percentage of the regional occurrences of change, highlighting that lower occurrences in comparison to the global observations still have substantial localized impacts.
NA = North America (including Caribbean), SA = South America, MEI = Middle East and India, SE Asia = Southeast Asia.
Fig 3.
The conversion of mangroves to aquaculture at the Mahakam delta, Kalimantan, Indonesia.
Mangrove degradation in the region was observed in the JERS-1/ALOS PALSAR color composite imagery (Red = 1996 JERS-1, Green = 2007 PALSAR, Blue = 2010 PALSAR) and verified using Landsat imagery. a) Color composite SAR image, b) 1996 Landsat 5 TM image, c) 2010 Landsat 5 TM image. The mangrove loss was identified using the distinct color in the radar composite imagery and geometric shape of the change feature. The distinct red color is a consequence of a decrease in radar backscatter in 2007 and 2010 from 1996 due to the replacement of a rough mangrove environment with the smooth surface of an aquaculture pond. Radar imagery copyright of JAXA. Landsat data available from the U.S. Geological Survey.
Fig 4.
Mangrove advance along the French Guiana coastline.
Mangrove advance in the region was observed in the JERS-1/ALOS PALSAR color composite imagery (R = 1996 JERS-1, G = 2007 PALSAR, B = 2010 PALSAR) and verified using Landsat imagery. A) JERS-1/PALSAR color composite image, B) 1997 Landsat 5 TM image, C) 2010 Landsat 5 TM image. The advance was identified due to the coloration of the feature in the radar composite image and its context along the coastline. The distinct blue color is a consequence of enhanced radar backscatter in 2010 due to the rough texture of a mangrove environment over that of the smoother surface of the ocean in 1996 and 2007. Radar imagery copyright of JAXA. Landsat data available from the U.S. Geological Survey.
Table 3.
Proportional mangrove area contained within change tiles.
Area of mangrove forest contained within tiles where change processes were observed. Forest area is provided as a percentage of the global quantity of mangrove forest.
Table 4.
Accuracy of identifying a change process within mangrove forest extent 1996-2010.