Forest fragmentation and its correlation to human land use change in the state of Selangor, peninsular Malaysia

https://doi.org/10.1016/j.foreco.2006.12.016Get rights and content

Abstract

In many tropical countries forest fragmentation caused by changes in human land use activities is of primary concern for sustainability. In Malaysia this relationship is usually described without any quantitative measurement to show the implications for land use planning and management. To understand this relationship, this study developed a single forest fragmentation index based on a combination of three landscape metrics, i.e., non-forest area, forest edge bordered by human land use and patch size coefficient of variation, using the state of Selangor, Malaysia as a case study. Two defined natural landscapes; wetland and forest landscapes were used as a basic unit of analyses of pattern change of forest fragmentation and its relationship to human land use in three temporal years of 1966, 1981 and 1995. Results showed that forest fragmentation in the state of Selangor generally increased during the time periods studied. Multiple regression analysis showed that human land use is an important determinant of forest fragmentation, but the significance of several land uses as explanatory variables to forest fragmentation changed temporally. Oil palm and rubber plantations were apparently the major contributors to forest fragmentation. Differences in fragmentation pattern occurred between the two natural landscapes. Forest fragmentation increased in the wetland landscape, whereas not much change was observed in the forest landscape. Oil palm plantation appeared to be the major contributor to forest fragmentation in the wetland landscape, whereas rubber plantation was the major factor in the forest landscape. This study also revealed that knowledge of forest fragmentation through a single index in different natural landscapes provides a pathway for identifying which forested areas are highly threatened and must be given priority in strategic planning of nature conservation. In conclusion, it is vital to understand the relationship between changes in human land use and degree of forest degradation measured by a single index in order to describe the implications for land use planning and management.

Introduction

In developing tropical regions, forest fragmentation is occurring at an alarming rate (Rudel and Roper, 1997, Laurance, 1999, Sánchez-Azofeifa et al., 2001). Since the last few decades of the 20th century, spatial and temporal change of land utilization by human activities has become more prominent as a proximate factor that catalyses forest fragmentation in the region (e.g., McMorrow and Talip, 2001, Ochoa-Gaona, 2001). Rapid economic development, as well as pressure on development from increasing human populations, are among the major underlying factors that created these circumstances (Tole, 1998, Koop and Tole, 2001, Uusivuori et al., 2002).

Malaysia is one of the developing tropical countries where both proximate and underlying causes of land use play a major role promoting forest fragmentation. From the 1950s to the 1970s, most of the natural forests were converted into agricultural land, mainly for rubber and oil palm plantations (Wong, 1974). However, prosperity from economic development due to emphasizing on the manufacturing sector since the 1980s, catalysed land development for other land uses, such as housing, new urban areas and industrial estates. But, at the same time, land development for these land-uses also intruded into forested areas. Increasing pressure on natural forest areas by human land use activities has led to diverse environmental problems, including landslides (Wan et al., 2000) and river sedimentation (Yusuf and Nordin, 2003), which are naturally related to the degradation of forest areas. However, the relationship between human land use change and forest fragmentation has always been described without any quantitative measurement to show the implications for the existing land use planning and management with regards to forest conservation.

In tropical regions, many studies have shown that human intervention in land utilization has changed forest cover over time (e.g., Kammerbauer and Ardon, 1999, Millington et al., 2003, Van Laake and Sánchez-Azofeifa, 2004). Forest fragmentation has also been measured to show the condition or trends of forest loss in a particular area or region (Dirzo and Garzia, 1992, Sierra, 2000, Trejo and Dirzo, 2000, Imbernon and Branthomme, 2001). However, these studies have not examined the correlation between the degree of forest fragmentation and human land use. In these studies, forest fragmentation was measured using multiple landscape metrics or indices.

Landscape metrics, such as mean patch size, edge density, mean shape index, contagion and mean nearest-neighbour distance, are usually used to describe forest fragmentation (O’Neill et al., 1988, O’Neill et al., 1999, McGarigal and Marks, 1995, He et al., 2000, Jaeger, 2000). However, not all the fragmentation metrics can capture the entire extant of forest fragmentation in a particular landscape (Cain et al., 1997). One of the major reasons is that most of the metrics are highly correlated to each other and some of them are redundant (McGarigal and Marks, 1995, Turner et al., 2001). In addition, landscape metrics are also sensitive to changes in levels of forest loss (Trani and Giles, 1999), which highlight the importance of awareness of their application for assessing and monitoring forest fragmentation. Thus, multiple landscape metrics cannot feasibly be correlated to human land use statistically. Therefore, an explicit single forest fragmentation index that can be correlated to human land use must be developed.

Recently, the landscape context has received much attention in land use planning and management (Brandt et al., 2002, Romero-Calcerrada and Perry, 2004), especially concerning forest and biodiversity conservation (Fukamachi et al., 2000, Nakamura and Short, 2001). Nevertheless, to augment the effectiveness in land use planning and management for the sustainability of land development, the concept of landscape type has been developed (Aubrecht et al., 2001). Landscape type can be used principally as a spatial reference unit for any further step of land evaluation with respect to nature conservation (Banko et al., 2003). Therefore, the concept provides other options to develop an approach for assessing levels of forest fragmentation through a single index and its relationship to human land use change. This can subsequently be integrated into land use planning and management for forest conservation.

Therefore, the objectives of this study are to develop a single forest fragmentation index to describe the degree of forest fragmentation and to understand the relationship between forest fragmentation and human land use change, using the state of Selangor, Malaysia as a case study. Forest fragmentation as defined here is caused directly by changes in human land use and excludes natural fragmentation and temporary human land use activity, such as logging.

Section snippets

Study area

The state of Selangor, which extends along the west coast of peninsular Malaysia, is located between 2°35′ and 3°60′N and between 100°45′ and 102°00′E. Selangor, the most rapidly developing state in Malaysia, has an area of approximately 800,000 ha and borders the state of Perak to the north, the state of Negeri Sembilan to the south, and the state of Pahang to the east (Fig. 1).

Having the highest gross domestic product per capita, particularly since the late 1980s, Selangor has become the

Data set

The pattern change of forest fragmentation and its relationship to human land use were analysed in 1966, 1981 and 1995. In this study the classification scheme of land use/cover categories in the state of Selangor as developed by Abdullah and Nakagoshi (2006) were used (Table 1). This classification was based on digitised (vector version) land use/cover maps of these years, produced by the Institute for Environment and Development (LESTARI), Universiti Kebangsaan Malaysia. These digitised maps

Land use change in natural landscape

Generally, in natural landscapes there were five land use/cover categories that showed a tendency to increase over the study period (Fig. 2a). These were barrenland/bareland, mixed horticulture, oil palm, other agriculture and urban and built-up area. The largest increase was in oil palm, followed by urban and built-up areas, while the proportion of barrenland/bareland, mixed horticulture and other agriculture increased only slightly (Fig. 2a). For oil palm, the percentage of increase was

Discussion

The environmental and biodiversity impact of forest fragmentation is a growing concern in tropical regions (Stevens and Husband, 1998, Laurance and Laurance, 1999, Sodhi, 2002). The forest fragmentation index is a useful tool for measuring the level of disturbance and vulnerability of these forests. The index developed in this study is potentially applicable not only for making comparisons of forest fragmentation across the state over time, but also for measuring the differences between types

Acknowledgements

We would like to thank the Ministry of Science, Technology and Innovation Malaysia, for providing financial support that enable us to conduct this research. The comments and suggestions provided by the two anonymous reviewers, which helped us to improve the presentation of this paper, were greatly appreciated.

References (58)

  • S.G. Laurance et al.

    Tropical wildlife corridors: use of linear rainforest remnants by arboreal mammals

    Biol. Conserv.

    (1999)
  • W.F. Laurance

    Reflections on the tropical deforestation crisis

    Biol. Conserv.

    (1999)
  • D.J. Mahar

    Policies affecting land use in the Brazilian Amazon—impact on the rainforest

    Land Use Policy

    (1990)
  • J. McMorrow et al.

    Decline of forest area in Sabah, Malaysia: relationship to state policies, land code and land capability

    Global Environ. Change

    (2001)
  • A.C. Millington et al.

    Scale dependence in multitemporal mapping of forest fragmentation in Bolivia: implication for explaining temporal trends in landscape ecology and applications to biodiversity conservation

    J. Photo. Rem. Sens.

    (2003)
  • T. Nakamura et al.

    Land-use planning and distribution of threatened wildlife in a city of Japan

    Landsc. Urban Plann.

    (2001)
  • H. Ohta

    A phenomenological approach to natural landscape cognition

    J. Environ. Psychol.

    (2001)
  • G.V.N. Powell et al.

    Assessing representativeness of protected natural areas in Costa Rica for conserving biodiversity: a preliminary gap analysis

    Biol. Conserv.

    (2000)
  • R. Romero-Calcerrada et al.

    The role of land abandonment in landscape dynamics in the SPA ‘Encinares del río Alberche y Cofio’, Central Spain, 1984–1999

    Landsc. Urban Plann.

    (2004)
  • R. Sierra

    Dynamics of patterns of deforestation in the western Amazon: the Napo deforestation front, 1986–1996

    Appl. Geogr.

    (2000)
  • R. Sierra et al.

    Assessing biodiversity conservation priorities: ecosystem risk and representativeness in continental Ecuador

    Landsc. Urban Plann.

    (2002)
  • S.M. Stevens et al.

    The influence of edge on small mammals: evidence from Brazilian Atlantic forest fragments

    Biol. Conserv.

    (1998)
  • M.K. Trani et al.

    An analysis of deforestation: metrics used to describe pattern change

    For. Ecol. Manage.

    (1999)
  • I. Trejo et al.

    Deforestation of seasonally dry forest: a national and local analysis in Mexico

    Biol. Conserv.

    (2000)
  • J. Uusivuori et al.

    Population, income and ecological conditions as determinants of forest area variation in the tropics

    Global Environ. Change

    (2002)
  • P.E. Van Laake et al.

    Focus on deforestation: zooming in on hot spots in highly fragmented ecosystems in Costa Rica

    Agric. Ecosyst. Environ.

    (2004)
  • S.A. Abdullah

    Fragmented forest in tropical landscape: the case of the state of Selangor, Peninsular Malaysia

    J. Environ. Sci.

    (2003)
  • C.L. Arnold et al.

    Remote-sensing-enhanced outreach education as a decision support system for local land-use officials

    Photo. Eng. Rem. Sens.

    (2000)
  • Aubrecht, P., Götz, B., Zethner, G., 2001. Constraints in land use, nature protection issues, rural development and...
  • Cited by (120)

    • Application of graph-based indices to map and develop a connectivity importance index for large mammal conservation in a tropical region: A case study in Selangor State, Peninsular Malaysia

      2022, Ecological Indicators
      Citation Excerpt :

      Therefore, Kuala Lumpur and Putrajaya were included in our analysis. The total size of the study area is about 800,000 ha (Abdullah and Nakagoshi, 2007) and extends along the west coast of Peninsular Malaysia. The climate is characterised by warm, sunny days and cool nights throughout the year with temperature ranges from 23 °C to 33 °C while humidity usually exceeds 80%.

    • Ecosystem-Based Adaptation: Approaches to Sustainable Management of Aquatic Resources

      2022, Ecosystem-Based Adaptation: Approaches to Sustainable Management of Aquatic Resources
    View all citing articles on Scopus
    View full text