Keywords: Mangrove, monitoring, coastal zone, deforestation, remote sensing, GIS,
radar, optical, satellite images
AbstractMangrove forests form one of the primary coastal ecosystems in the tropical and subtropical regions
of the world. Mangrove ecosystems are very sensitive and fragile. The pressures of increasing population, and the
resulting expansion of agriculture and industrial and urban development, have caused a significant proportion of the
world’s mangroves to be destroyed. Reliable and timely information is therefore required in order to monitor and
manage the remaining mangrove resources. This paper describes research to compare the ability of three different
radar satellite imaging systems and three different optical satellite systems to detect mangrove deforestation in the
delta of the Mahakam River, East Kalimantan, Indonesia.
1. Introduction
Mangrove forests form one of the primary coastal ecosystems in the tropical and subtropical region of the world.
They are biologically diverse and have therefore traditionally been utilized for food resources, firewood, charcoal,
timber and other minor products. However, mangrove ecosystems are very sensitive and fragile. In recent years, the
pressures of increasing population, and the resulting expansion of agricultural land and industrial and urban
development, have caused a significant proportion of the world’s mangrove resource to be destroyed. In addition,
significant areas of mangrove swamps in Indonesia and other regions of Southeast Asia have been developed to
create ponds for the commercial production of fish and shrimps.
Information is crucial in order to assess mangrove deforestation, to monitor the state of the remaining mangrove
forests and ensure their sustainable management. In mangrove areas, the task of collecting information by ground
inventory is extremely difficult, time consuming and, therefore, expensive. For this reason, remote sensing is an
attractive means of obtaining data for defining deforested areas and updating management plans. In Indonesia, a
serious drawback to using optical satellite images for mangrove monitoring has been the unavailability of cloud-free
images at the time of interest. The advent of radar systems is expected to resolve this problem. Unlike optical
systems, radar systems can be operated day or night and have all weather capability. They can therefore penetrate
clouds cover, fog, rain and atmospheric dust. Furthermore, radar energy has the ability to penetrate tree canopies and
some surface features. This paper describes research to compare three different radar satellite imaging systems
(ERS-1, JERS-1, and Radarsat) and three types of optical satellite data (Landsat TM, MSS and Spot XS) for
detecting and monitoring mangrove deforestation in the delta of the Mahakam River on the coast of East Kalimantan.
2. Materials and Methods
2.1 Study Area
Because it is an archipelago, Indonesia has a large area of mangrove (4,250,000 ha are found around the four main
islands). There are basically six categories of land use in the Indonesian mangrove areas: 1) forest reserves; 2) nature
conservation; 3) protection forest; 4) production forest; 5) fish/shrimp ponds; and 6) salt production areas. Mangrove
forests are traditionally a source of livelihood for the local population. The main problems of managing these
resources therefore centre on delineating the area and resolving local community ownership issues.
In the study area, the delta of Mahakam River in East Kalimantan, the commercial production of shrimps for export
was established in 1974. Every year, the company needs about 1,000,000 Kg of shrimps to reach their export target.
For example, in 1996 it purchased about 1,015,000kg from local fishermen. At that time, shrimp ponds supplied
about 4% of the demand, and fishermen operating at sea supplied most of the shrimps. The high demand for shrimps
for export has resulted in substantial price increases. In 1997, the price of good quality of shrimps was about
Rp50,000, or about US$15 per kilogram. This has encouraged local fishermen to build large areas of shrimp ponds.
During the 1990s, they encroached into the mangrove forest, clear-cut the mangrove vegetation by slash and burn
and then constructed dykes to surround the ponds. This activity tends to be concentrated in the areas covered by
Nypa palm forest The system of production in the Mahakam delta is extensive. For example, the fishermen do not
use water mills for oxygen circulation in the ponds but, instead, rely on the current flowing between high and low
tides, large areas are required for extensive production. As a result, the ponds produce only about 2-3 tons per
hectare. This is low compared with the 10 tons per hectare produced from shrimp ponds on the northern coast of
Java. Because intensive production systems are expensive, the local fishermen assume that sufficient mangrove
forest will remain to allow the establishment of large areas of ponds.
2.2 Remote sensing data
Data acquired by both passive and active remote sensing systems were used in the study. The specifications are
summarized in Table 1.
Table 1 Image specifications
| Image |
Date
of
acquisition |
Wavelength |
Spatial
reso-
lution |
Polari-
zation |
Inci-
dence
angle |
| Landsat MSS |
15-04-1983 |
1) 0.5 - 0.6mm
2) 0.6 - 0.7mm
3) 0.7 - 0.1mm
4) 0.8 - 1.1mm |
70 m |
- |
- |
| SPOT-XS |
21-02-1987 |
1) 0.50-0.59mm
2) 0.61-0.68mm
3) 0.79-0.89mm |
20 m |
- |
- |
| Landsat TM |
1994 |
1) 0.52-0.60mm
2) 0.76-0.90mm
3) 1.55-1.75mm |
30 m |
- |
- |
| JERS-1 |
19-09-1996 |
1) 23.5 cm |
12.5 m |
HH |
35° |
| ERS-1 |
28-05-1996 |
1) 5.6 cm |
12.5 m |
VV |
23° |
| Radarsat-1 |
1-10-1997 |
1) 5.6 cm |
12.5 m |
HH |
40° |
