Noordyana Hassan, Shinya Numata, Mazlan Hashim

Weekly Newsletter

Forest degradation and deforestation is now at alarming rates. This problem occurs due to anthropogenic pressures such as selective logging, illegal logging, forest harvesting, and over harvesting of wood fuel and timbers. Forest degradation and deforestation may give negative impacts to livelihood, biodiversity and climate. To achieve the sustainable forest management, estimation of tree species composition per unit compartment of forest reserve is crucial to be identified. Thus, forest management can be monitor by using remote sensing data. However, there are some limitations on coarse spatial resolution of satellite data due to the availability of mixing pixels in estimating tree species composition. Thus, spectral unmixing is one of the solution of mixing pixels.

Forest Degradation and Deforestation Issues

Tropical rainforest is rapidly decreasing and novel techniques are needed to ensure the ongoing conservation of this ecosystem (Mon et.al., 2012). As for example, according to Global Forest Resources Assessment (2010), mean of Malaysian deforestation rate is about 13% of 520,000 ha per year. Figure 1 shows the size of forest extent in million hectares which decreasing within ten years in Peninsular Malaysia. This problem occurs due to deforestation and forest degradation which are at progressive processed and advancing at an alarming rate (Panta et. al., 2008). Thus, forest degradation and deforestation in the tropics have importance to biodiversity conservation (Mon et. al., 2012).
Meanwhile, deforestation is the conversion of forest to other land uses such as agriculture area, oil palm plantation, roads and other infrastructure. Besides, selective logging is one of the main factors of forest degradation and deforestation (FAO, 2005). Heavy selective logging may cause deforestation. Hence, degradation may forerunner to deforestation because of logging reducing the perceived conservation value and increases access to the forest which leads to clearing.
Therefore, it is important to cope with this problem by identifying source of degradation and deforestation. Thus, to cope with deforestation and forest degradation problem, numerous of studies using remote sensing techniques has been widely analyzed. Remote sensing technology has an ability to identify forest degradation and deforestation at large areal extent (Panta, 2008). Identification of tree species distributions and estimation of tree species composition especially on timbers that have high commercial values can be one of the solutions to cope with forest degradation and deforestation problems by emphasizing remote sensing techniques. The useful information obtained can be used for planning and sustainable management of forest (Panta, 2008, Mon et. al., 2012).
In addition, by estimating tree species composition, the timber forest resource can be identified. Hence, selective and illegal logging can be monitored frequently due to the existence of temporally remote sensing data. Moreover, remote sensing data able to cover large areas and it suits with forest degradation and deforestation monitoring. Therefore, effective management activities can be developed to achieve sustainable forest management (SFM) (Mon et. al., 2012). However, monitoring forest degradation and deforestation using remote sensing techniques may have its own challenges especially when using medium and coarse spatial resolution.

Challenges in Identification of Tree Species using Remote Sensing Data

Remote sensing technology has been utilized in monitoring forest degradation and deforestation. Therefore, in order to cope with this problem, the sources of the problem need to be identified. Hence, estimations of tree species composition and its distribution are needed (Foody and Cutler, 2006). However, studies are limited by the spatial resolution of the remotely sensed data available expressed by the image’s pixel size that coarser than desired with target features. Therefore, this latter issue may relate to mixed pixel problem where the pixel not belongs to a single class but mixed with other classes. Nevertheless, medium and coarse spatial resolution may increase mixed pixels in the image (Foody et.al., 1994).
Besides, high heterogeneity of the features may lead to mixed pixels occurrence especially in medium and coarse spatial resolution data. Tropical rainforest on the other hand was heterogeneous, highly complex, with high density of emergent and large canopy tree. Hence, the occurrence of mixed pixels is high. Medium and coarse spatial resolutions may contribute to mixed pixels due to multiple spectral responses from many features in a pixel. According to Boardman and Kruse (2011), when target of interest is smaller or equal than the instantaneous of view (IFOV), therefore, mixed pixel is inevitable. High spectral resolution of satellite remotely sensed data is also appropriate for mapping individual tree species with high precision and accuracy due to similar characteristics within tree species especially tree species from the same genera (William and Hunt, 2002). Hence, spectral resolution with a narrow band is needed to differentiate similar tree from the same genera. Radiometric correction also important to ensure the ability to distinguish differences in reflectance values among pixel (Foody and Cox, 1994).
Thus, spectral unmixing approaches are necessary to overcome mixed pixels problem as spectral unmixing approaches has been used in numerous of previous studies (Hassan and Hashim, 2011; Ball et.al.,2004; William and Hunt, 2002). Further read on course spatial resolution data for SFM can be access trough: https://doi.org/10.1117/1.JRS.9.096046

References:
• R. Dehaan, J. Louis, A. Wilson, A. Hall, and R. Rumbachs, “Discrimination of blackberry (Rubus fruticosus sp. agg.) Using hyperspectral imagery in Kosciuszko National Park, NSW, Australia.”
• ISPRS Journal of Photogrammetry & Remote Sensing, 62, 13–24, 2007.
• FAO, F.,(2005). “Global Forest Resources Assessment 2005 country report: Myanmar”. FRA2005/107. Rome, Italy.
• Foody, G. M. and Cox., D. P. (1994). “Sub-pixel land cover composition estimation using a linear mixture model and fuzzy membership functions”. International Journal of Remote Sensing 15: 619–631.
• Foody, G. M., and Cutler, M. E. J., (2003). “Tree biodiversity in protected and logged Bornean tropical rain forests and its measurement by satellite remote sensing”. Journal of Biogeography, 30, 1053–1066
• Foody, G. M., and Cutler, M. E. J., (2006). “Mapping the species richness and composition of tropical forests from remotely sensed data with neural networks”. Ecological Modelling, 195, 37–42.
• Hassan, N. (2014) ‘Relative abundance estimations of chengal tree in a tropical rainforest by using modified Canopy Fractional Cover (mCFC)’, in IOP Conference Series: Earth and Environmental Science. doi: 10.1088/1755-1315/18/1/012189.
• Hassan, N. and Hashim, M. (2011) ‘Decomposition of mixed pixels of ASTER satellite data for mapping Chengal (Neobalanocarpus heimii sp) tree’, in Proceedings – 2011 IEEE International Conference on Control System, Computing and Engineering, ICCSCE 2011. doi: 10.1109/ICCSCE.2011.6190499.
• Keshava, N., and Mustard, J.F. (2002). “Spectral Unmixing”. IEEE Signal Process Mag. 19: 44-57.
• Somers, B., Asner, G.P., Tits, L., Coppin, P. (2011). “Endmember variability in Spectral Mixture Analysis: A review”. Remote Sensing of Environment 115: 1603 – 1616.
• Mon, M. S., Mizoue, N., Htun, N. Z., Kajisa, T., Yoshida, S.(2012). “Factors affecting deforestation and forest degradation in selectively logged production forest: A case study in Myanmar”. Forest Ecology and Management 267, 190–198.
• Panta, M., Kim, K., Joshi, C., (2008). “Temporal mapping of deforestation and forest degradation in Nepal: applications to forest conversation”. Forest Ecology Management 256, 1587–1595.
• Williams, A. P., and Jr, E. R. Hunt (2002). “Estimation of leafy spurge cover from hyperspectral imagery using mixture tuned matched filtering”. Remote Sensing of Environment 82: 446–456.

 

 

 

 

 

 

 

 

Name: Noordyana Hassan
Date: 13.07.2020
Email address: noordyana@utm.my

 

 

Research Ecosystem
Universiti Teknologi Malaysia UTM Nexus - Research & Innovation

Office of Deputy Vice Chancellor (Research & Innovation)

DVCRI Profile Johor Bahru Office Kuala Lumpur Office

Higher Institution Centre of Excellence (HI-COE)

Advance Membrane Technology Research Centre - AMTEC Institute of Noise & Vibration - INV Wireless Communication Centre - WCC

Research Institute

Centre of Excellence (COE)

Institute of High Voltage & High Current - IVAT UTM-MPRC Institue for Oil & Gas - IFOG Centre for Artificial Intelligence & Robotics - CAIRO Centre for Engineering Education - CEE Centre for Advanced Composite Materials - CACM Innovation Centre in Agritechnology for Advanced Bioprocessing - ICA Institute of Bioproduct Development - IBD

Service Entity

Research Management Centre - RMC Penerbit UTM Press Centre for Community & Industry Network - CCIN Innovation & Commercialisation Centre - ICC University Laboratory Management Centre - PPMU Institut Sultan Iskandar - UTM-ISI

Get the latest news & events

Customer Satisfaction Index