Geospatial Approach for Ecosystem Change Study of Lombok Island under the Influence of Climate Change

Abstract

Lombok Island is categorized as a small island with an area o f 4738.7 km2 which makes it susceptible to climate change. Climate change is thought to have reduced the ability of forest biophysical and can cause damage to the ecosystem or a shifting of forest ecosystems. The purpose of this study is to identify the ecosystem changes in Lombok Island based on historical climate data and also to identify the vegetation domination based on ground truth result. Research study area covers the entire area of Lombok Island, West Nusa Tenggara. Data analysis was performed using Geographic Information Systems (GIS) to analyze the spatial distribution of ecosystem zones. Spatial data processing was performed by interpolation method to estimate values at unknown locations or adjacent points. Ecosystem types in this study were determined by Holdridge Life Zones which are generated from historical climate data from 1975 to 2012 and described by the time series map of 1975, 1995, and 2012. The result indicates that ecosystem change occurred in Lombok Island which has shown in time series map of ecosystem zone. There were only five life zones in 2012 which were tropical moist forest, tropical dry forest, tropical very dry forest, subtropical wet forest and subtropical moist forest. The tropical dry forest was seen to dominate the area of Lombok Island in 2012. The ground truth result shown that the vegetation domination matches each ecosystem zone in the recent time.

Full text

Procedia Environmental Sciences 24 ( 2015 ) 165 – 173
1878-0296 © 2015 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Selection and peer-review under responsibility of the LISAT-FSEM Symposium Committee
doi: 10.1016/j.proenv.2015.03.022
Available online at www.sciencedirect.com
ScienceDirect
The 1st International Symposium on LAPAN-IPB Satellite for Food Security and Environmental
Monitoring
Geospatial approach for ecosystem change study of Lombok Island
under the influence of climate change
Saputri Saptaa,*, Bambang Sulistyantarab, Indung Sitti Fatimahb, Akhmad Faqihc
*
aDepartment of Landscape Architecture, Bogor Agricultural University Graduate School, Kampus IPB Darmaga, Bogor 16680,
Indonesia
bDepartment of Landscape Architecture, Bogor Agricultural University, Kampus IPB Darmaga, Bogor 16680, Indonesia cDepartment of
Geophysics and Meteorology, Bogor Agricultural University, Kampus IPB Darmaga, Bogor 16680, Indonesia
Abstract
Lombok Island is categorized as a small island with an area o f 4738.7 km2 which makes it susceptible to climate change.
Climate change is thought to have reduced the ability of forest biophysical and can cause damage to the ecosystem or a shifting
of forest ecosystems. The purpose of this study is to identify the ecosystem changes in Lombok Island based on historical climate
data and also to identify the vegetation domination based on ground truth result. Research study area covers the entire area of
Lombok Island, West Nusa Tenggara. Data analysis was performed using Geographic Information Systems (GIS) to analyze the
spatial distribution of ecosystem zones. Spatial data processing was performed by interpolation method to estimate values at
unknown locations or adjacent points. Ecosystem types in this study were determined by Holdridge Life Zones which are
generated from historical climate data from 1975 to 2012 and described by the time series map of 1975, 1995, and 2012. The
result indicates that ecosystem change occurred in Lombok Island which has shown in time series map of ecosystem zone. There
were only five life zones in 2012 which were tropical moist forest, tropical dry forest, tropical very dry forest, subtropical wet
forest and subtropical moist forest. The tropical dry forest was seen to dominate the area of Lombok Island in 2012. The ground
truth result shown that the vegetation domination matches each ecosystem zone in the recent time.
© 2015 The Authors. Published by Elsevier B.V.
Selection and peer-review under responsibility of the LISAT-FSEM Symposium Committee.
Keywords: climate change; ecosystem change; Geographic Information Systems (GIS); Holdridge Life Zones
* Corresponding author. Tel.: +62 899 954 4559; fax: +62 251 863.
E-mail address: saputri.sapta@gmail.com.
© 2015 The Authors. Published by Elsevier B.V This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Selection and peer-review under responsibility of the LISAT-FSEM Symposium Committee

166 Saputri Sapta et al. / Procedia Environmental Sciences 24 ( 2015 ) 165 – 173
1. Introduction
1.1. Background
Ecosystems of the Lombok Island are very interesting to study considering the beauty of coastal ecosystems in
the South and mountain ecosystems with the Mount Rinjani dominates the West of Lombok Island. In addition,
Lombok Island is a border region between the flora and fauna of Asia and Australia and form unique ecosystem
diversity. However, Lombok Island is categorized as a small island with an area of 4738.7 km2 which makes it
susceptible to climate change. Climate change is thought to have reduced the ability of forest biophysical and can
cause damage to the ecosystem or a shift of forest ecosystems.
There has been a research on climate change on ecosystem zones in Lombok Island which has shown the shifted
climate types in the period 1961-2008 [1]. However, in that study, the climate data obtained only from
meteorological stations of Selaparang and Kediri, thus the distribution of climate zones and ecosystems is not very
representative of actual conditions. Therefore, further study is needed using a geospatial approach to see the
distribution of climate zones and ecosystems of Lombok Island.
This previous study showed that there have been some changes in Lombok Island on its climate which is
characterized by the changing trend of rainfall, temperature and climate types. The impacts of climate change on
forest ecosystems include mangrove forest ecosystem devastation, loss of endemic species, decreased in land cover,
as well as reduced quality and quantity of springs. Therefore, climate change information and the resulting impacts
need to be updated to support the optimization of adaptation and mitigation of climate change so as to reduce the risk
of ecological damages.
1.2. Objective
The purpose of this study is to determine the ecosystem zones spatially and identify the ecosystem change in
historical data in Lombok Island based on the climate historical data. The results of this study are expected to be the
basis of the data that is important to formulate mitigation and adaptation strategies to deal with the clim ate change,
particularly on the Lombok Island landscape.
1.3. Problem Statement
This research was done by analyze the climate data and classify the ecosystem zones. To determine the result of
the research there will be some problems which are stated below:
1. How to classify the ecosystem of an area?
2. How are the changes of the ecosystem zone in the Lombok Island?
3. How was the climate change affect to the ecosystem zone in the Lombok Island?
2. Methodology
2.1. Study Area
Research study area covers the entire area of the island of Lombok, West Nusa Tenggara (Figure 1). Field
observations and ground truthing will be conducted on Lombok Island at some points in the region in accordance
with the results of the determination of the sampling area will be undertaken ahead of the field observations. While
the analysis and synthesis will be performed in Bogor, West Java.

167
Saputri Sapta et al. / Procedia Environmental Sciences 24 ( 2015 ) 165 – 173
Fig. 1. Areas of research, Lombok - NTB (Source: Google Earth)
2.2. Research Framework
The scope of this study is limited to landscape change of Lombok Island using historical climate data from 1975
to 2012. Figure 2 shows that data collecting process covers observation tabular data from climate station and also
base map to be used in data processing and analyzing with GIS approach. GIS approach is used to obtain bio-
temperature and precipitation data which are required for the ecosystem zones classification process. Ecosystem
zones to be classified using Holdridge Life Zones method as it correlates to climate factor which would be explained
in the next section (subchapter 2.3). Ground truthing is conducted to validate the result derived from this study to
verify the current existing condition with the description obtained from analysis result.
Fig. 2. Workflow diagram of the research
2.3. Ecosystem Zones Classification
Ecosystem zones were determined using Holdridge Life Zones classification system to assess the impact of
climate change on ecosystem zones. The Holdridge Life Zones system correlates climatic indices with 37 life zones
ranging from polar desert to wet tropical rainforest. It uses two main variables in determining classification, average
biotemperature and average annual precipitation [2].
The average rainfall annually (mm) is used for precipitation data . Biotemperature (BT) is a unit of measurement
of energy used in the life zone chart where this unit shows the average value of the air temperature in Celsius is used
for growing crops. This temperature range is between 0°C as the minimum point and 30°C as the maximum point
(0°C <T <30°C). The ratio of annual potential evapotranspiration as the third variable is a function of biotemperature
and precipitation. Therefore, this is not required as input for the life-zone classification scheme [3].
Holdridge Life Zones diagram (Figure 3) is a graphical classification of zone ecosystems on earth that shows the
relationship of the mountains and lowland vegetation based on latitude, elevation, precipitation and air temperature
[2]. Classification zones are a rough model that can predict potential forest types that can grow optimally in regions
with certain climatic conditions. This diagram is formed using two identical axes for average annual precipitation to

168 Saputri Sapta et al. / Procedia Environmental Sciences 24 ( 2015 ) 165 – 173
make up two sides of an equilateral triangle. The third side of the triangle is a logarithmic axis for potential
evapotranspiration (PET) ratio measured in millimeters per year (mm/yr). Axes for mean annual biotemperature are
set to the base of the triangle [4].
Since the number of surface observation station for temperature database availability is inadequate, temperature
data should be derived from DEM (satellite image) and will be references to existing data point from surface
observation of Meteorology Climatology and Geophysics Agency (BMKG). Meanwhile, the precipitation data
collected from 33 rainfall observation station of Sub Directorate of Hydrology and Water Quality.
Fig. 3. Holdridge Life Zones classification system [2]
3. Results and Discussion
3.1. Data Analysis
Data analysis was performed using Geographic Information Systems (GIS) to analyze the spatial distribution of
climate zones and ecosystems to facilitate interpretation from these data. Spatial data processing was done by
interpolation to estimate values at unknown locations or adjacent points [5]. Ecosystem zones are determined using
Holdridge Life Zones method with historical climate data from 1975 to 2010. Climate data used in this study
consists of biotemperature and rainfall data. Temperature data derived from DEM data analysis to create a
biotemperature map, while areal rainfall map obtained from the analysis of Thiessen Polygon.
Holdridge [2] proposed a life zone classification to predict the potential vegetation of a region for values of
climatic indices. Life zones are delimited by bio-temperature, precipitation, and potential evaporation ratio. In
particular, the system is based on two factors: mean annual biotemperature and mean annual precipitation [6]. Based
upon study of several ecosystems, Holdridge assumes that the potential evaporation ratio (PET) is proportional to
biotemperature with a proportionality constant of 58.93. PET is therefore not an independent variable but simply
derived from the two primary variables of precipitation and biotemperature [4].

169
Saputri Sapta et al. / Procedia Environmental Sciences 24 ( 2015 ) 165 – 173
Spatial data processing on this study, as described on figure 4, consists of climate data analysis to create the
climatic map which then can be used to perform the ecosystem zone analysis based on Holdridge Life Zones
classification system. Result was validated with ground truthing to verify the recent existing ecosystem.
Fig. 4. Spatial data processing scheme
3.1.1. Base Map
Data used for spatial processing analysis on this study include the administration map, SRTM data, and climate
observation station map of Lombok Island. Administration map is used to provide the area identification, and also to
clip the other spatial maps according to the boundary of Lombok Island. SRTM data used are in ARC GRID, ARC
ASCII and Geotiff format with 6000 x 6000 pixels, in decimal degrees and datum WGS84. Climate observation
stations are shown on the map according to its coordinate which then can be included the historical climate data
information attribute for each station point.
3.1.2. Biotemperature
Biotemperature defined as the mean of unit-period temperatures with substitution of zero for all temperature
values below 0°C and above 30°C. On this study, air temperature was estimated by using Braak equation. The
higher the elevation, the lower is the air temperature [7]:
T = 26,3 °C - (0,01 x elevation in meter asl x 0,6 C)
The average air temperature at zero elevation (coast) ranges from 25 to 27°C [8].
DEM analysis was used to create a biotemperature map based on the temperature data from observation stations
combined with elevation data of the area. Digital Elevation Models (DEM) are 3-dimensional representations of the

170 Saputri Sapta et al. / Procedia Environmental Sciences 24 ( 2015 ) 165 – 173
topography of landscapes and can be used to generate information relevant to hydrological, geomorphological,
biological and other environmental applications [9] [10]. Topography often overlooked because of its influence on
climate. DEM data was derived from the USGS/NASA SRTM data [11], while the historical temperature data
derived from four climatology and meteorology stations of Meteorology Climatology and Geophysics Agency
(BMKG). As the climate data from those stations were inadequate to describing the whole temperature condition,
therefore this study fetched the data from other area by using Braak formula.
Evaluating the fetched data to the existing one were done by linear regression method, the main purpose of this
step is to enhance the scope of data Regression analysis is a statistical method that investigates the relationship
between a response variable Y and a set of other variables named as independent variables or predictors X [12]..
Linear regression is conducted by comparing between surface observations data with data derived from Braak
equation. Regression equation result is used to correct biotemperature data which has been generated from DEM
data. The map below (Figure 5) is the example of biotemperature map which uses the surface observation data of
2000 since its data availability is the most sufficient for the area coverage.
Fig. 5. Biotemperature map of Lombok Island in 2000
3.1.3. Precipitation
Data limitation both from the surface observation station and satellite imagery enforces this study to use Thiessen
Polygon method although the accuracy of this method has not been adequate. However, Thiessen Polygon are best
used at an early stage in a spatial study to sort out visually confusing site distributions into patterns that are str ong
enough to warrant further study.
Rain gauges generally measure rainfall at individual points but can be compared directly with runoff from that
area to obtain the average depth of rainfall occurring over an area which is called as the areal rainfall. The Thiessen
polygon method accounts for the variability in spatial distribution of gauges and the consequent variable area which
teach gauge represents. The area of a polygon for an individual station as a proportion of the total basin area
represents the Thiessen weight for that station [13]. The Thiessen polygon method is objective and readily
computerized which makes it as a widely-used method.
Interpolation refers to the process of estimating the unknown data values for specific locations using the known
data values for other points. Thiessen polygons are constructed by around each sample point. All points within a
polygon are assumed to have the same data value as the sample point around which polygon is constructed [14].
Thiessen polygon analysis for this study was generated on GIS using rainfall observation point of study area which
Legend
Biotemperature range (in °C)

171
Saputri Sapta et al. / Procedia Environmental Sciences 24 ( 2015 ) 165 – 173
then created the areal rainfall map of Lombok Island (Figure 6). This example of rainfall map used the surface
observation data of 2000 since it has the most complete number of surface observation stations which provide the
rainfall data to show the distribution of the total observation stations located in study area.
Fig. 6. Areal rainfall map of Lombok Island in 2000
3.1.4. Ecosystem Zone Classification
Combination between biotemperature and precipitation index value create 37 ecosystem zones which called as
life-zones. Biotemperature index and precipitation index are determined based on value category of each life-zone
which is referred to Holdridge Life Zones classification scheme. Although regional section of humidity and
altitudinal belts cutting across regions have some natural relations, the latitudinal biotemperature regions are seem to
be natural geographical units [2]. The preliminary result of ecosystem zone map which displays the life zone
classification over an area is showed through three years (1975, 1995, and 2012). This spatial information shows
that Lombok Island has 10 types of life zones, such as subtropical thorn steppe/woodland, subtropical dry forest,
subtropical moist forest, subtropical wet forest, subtropical rain forest, tropical thorn wood, tropical very dry forest,
tropical dry forest, tropical moist forest, and tropical wet forest (Figure 7).
In 1975, there are seven life zones such as tropical wet forest, tropical moist forest, tropical dry forest, tropical
very dry forest, subtropical rain forest, subtropical wet forest, and subtropical moist forest. In 1995, it changed to
eight life zones which are tropical moist forest, tropical dry forest, tropical very dry forest, tropical thorn woodland,
subtropical wet forest, subtropical moist forest, subtropical dry forest, and subtropical thorn steppe/woodland. The
tropical wet forest and subtropical rain forest life zones were no longer exist in 1995. There were only five life zones
in 2012 which are tropical moist forest, tropical dry forest, tropical very dry forest, subtropical wet forest and
subtropical moist forest. The tropical dry forest was seen to dominate the area of Lombok Island in 2012. The
ground truth check result verified that the vegetation domination in Lombok area are dominated by grasses,
groundcovers, shrubs, scrubs, and short trees which could indicate a dry forest. There are plenty of grasslands and
shrublands found within this area. Nevertheless, some rain forest ecosystems are found in the center of the island.
The significant changes from 1975 to 2012 caused by the changes of the precipitation and biotemperature values.
But, there were also heterogeneity of the climate data attributes from each observation station which make the
Legend

172 Saputri Sapta et al. / Procedia Environmental Sciences 24 ( 2015 ) 165 – 173
visible changes become biased. It is necessary to improve the data availability from each observation station by
conducting a further data inventory.
Fig. 7. Ecosystem zones map of Lombok Island in 1975 (a), 1995 (b), and 2012 (c) based on Holdridge Life Zones classification.
3.2. Discussion
Former research of ecosystem change study in Lombok Island by Rieke Nandini and Budi Narendra [1] has
shown the shifted climate types in the period 1961-2008. The climate data of the study obtained only from
meteorological stations of Selaparang and Kediri, which makes distribution of climate zones and ecosystems is less
representative to describe actual conditions.
Yates et al. [3] compared the Holdridge Life Zones model as a climate-vegetation model to three mechanistic
simulation models (BIOME2, Dynamic Global Phytogeography Model (DOLY), and Mapped Atmosphere-Plant-
Soil System (MAPSS)) for the conterminous United States under contemporary climate and a set of future climates
prescribed by three Global Circulation Model experiments. It showed that under circumstances of limited data
availability, computation resources, and access to mechanistic models and model expertise, simple correlation
models such as Holdridge may be the only method that can be applied.
Lasco et al. [15] figured out that temperature increase had minimal effect on life zones in the Philippines, because
all parts of the country already fall within the tropical belt under the Holdridge system (>24°C). However, specific
forest type could change dramatically with increases in precipitation levels.
4. Conclusions
The preliminary result shows that the ecosystem zone of Lombok Island has been varied from 1975 to 2012
according to its climatic data which are subjects to climate change. The range of ecosystem zone of Lombok Island
from 1975-2012 includes tropical wet forest, tropical moist forest, tropical dry forest, tropical very dry fore st,

173
Saputri Sapta et al. / Procedia Environmental Sciences 24 ( 2015 ) 165 – 173
tropical thorn woodland, subtropical rain forest, subtropical wet forest, subtropical moist forest, subtropical dry
forest, and subtropical thorn steppe/woodland. The tropical dry forest is seen to dominate the area of Lombok Island
in 2012.
It is recommended for the future studies will look at how climate change affect the ecosystem in the future using
climate scenarios and also to show trends of climate change that has occurred from precipitation and biotemperature
values.
References
1.
R. Nandini and B. Narendra , "Kajian Perubahan Curah Hujan, Suhu, dan Tipe Iklim pada Zona Ekosistem di Pulau Lombok," Jurnal
Analisis Kebijakan Kehutanan, 2011; 3(8): 228-244,.
2.
L. Holdridge, Life Zone Ecology, San Jose, Costa Rica: Tropical Science Center; 1967.
3.
D. Yates, T. Kittel and Cannon RF, "Comparing the correlative Holdridge model to mechanistic biogeographical models for assessing
vegetation distribution response to climatic change. Climatic Change 2000;44:59-87.,"
4.
C. Thomas, "Global Warming and World Ecosystem Distribution: Toward Quantifying Ecosystem Change," Massachusetts Institute of
Technology, Massachusetts; 1993.
5.
A. As-Syakur, "Evaluasi zona agroklimat dari klasifikasi Schmidt-Ferguson menggunakan aplikasi Sistem Informasi Geografi (SIG)," Jurnal
Pijar MIPA, 2009; 3(1):17-22,.
6.
T. C. N. K. ML Parry, The Impact of Climatic Variations on Agriculture, Volume 2. Assessments in Semi-Arid Areas, Kluwer: Dordrecht;
1988.
7.
C. Braak, The Climate of The Netherlands Indies. Proc. Royal Mogn. Meteor. Observ. Batavia; 1928.
8.
S. Ritung, W. F. Agus and H. Hidayat, Land Suitability Evaluation with a Case Map of Aceh Barat district, Bogor, Indonesia: Indonesian Soil
Research Institute and World Agroforestry Centre; 2007.
9.
S. Jenson and J. Domingue, "Extracting topographic structure from digital elevation data for geographic information system analysis,"
Photogrammetric Engineering and Remote Sensing, 1988;54:1593-1600,.
10.
M. Hutchinson and J. Gallant, "Representation of terrain," in Geographical Information Systems: Principles, Technical Issues, Management
Issues and Applications. Second Edition, New York, Wiley, 1999:105-124.
11.
A. Jarvis, H. Reuter, A. Nelson and E. Guevara, Hole-filled SRTM data Version 4, available from CGIAR-CSI SRTM 90m Database
(http://srtm.csi.cgiar.org), International Centre for Tropical Agriculture (CIAT); 2008.
12.
A. A. Smadi and N. H. Abu-Afouna, "On Least Squares Estimation in a Simple Linear Regression Model with Periodically Correlated Errors:
A Cautionary Note," Austrian Journal of Statistics,. 2012;41( 3): 211-226,.
13.
Hydrology Project Training Module, How to compile rainfall data, New Delhi: Hydrology Project Technical Assistance; 2002.
14.
P. A. Burrough and R. A. McDonnell, Principles of Geographical Information Systems, New York: Oxford University Press; 1998.
15.
R. Lasco, S. Roy, P. Sanchez and K. Garcia, Modeling of Climate Change on Transition of Forest types in the Philippines using the
Holdridge Life Zones and GCM Projections for Southeast Asia., in press; 2007.
16.
W. Taesombat and N. Sriwongsitanon, "Areal rainfall estimation using spatial interpolation," ScienceAsia, 2009; 35: 268-275,.
17.
N. Estwick, "GIS as a Tool for Rainfall Analysis," in Esri UC, San Diego; 2011.
18.
M. Hall, Archaeology Africa, Oxford: James Currey Publishers; 1996.