Mohd Hasmadi Ismail

Classification model for hotspot occurrences using a decision tree method

Forest fires in Indonesia mostly occur because of errors or bad intentions. This work demonstrates the application of a decision tree algorithm, namely the C4.5 algorithm, to develop a classification model from forest fire data in the Rokan Hilir district, Indonesia. The classification model used is a collection of IF-THEN rules that can be used to predict hotspot occurrences for forest fires. The spatial data consist of the location of hotspot occurrences and human activity factors including the location of city centres, road and river networks as well as land cover types. The results were a decision tree containing 18 leaves and 26 nodes with an accuracy of 63.17%. Each leaf node holds positive and negative examples of hotspot occurrences whereas the root and internal nodes contain attribute test conditions: the distance from the location of examples to the nearest road, river, city centre and the land cover types for the area where the examples are located. Positive examples are hotspot locations in the study area and negative are randomly generated points within the area at least 1 km away from any positive example. The classification model categorized whether the region was susceptible to hotspots occurrences or not. The model can be used to predict hotspot occurrences in new locations for fire prediction.

Land Use Trends Analysis Using SPOT 5 Images and Its Effect on the Landscape of Cameron Highlands, Malaysia

A large part of the mountain steep land in Peninsular Malaysia is covered by forests. Cameron Highland is a mountainous region with a climate favourable to the cultivation of tea, sub-tropical vegetable and flowers. However rapid economic growth and land use practices has altered the environment landscape of the area. This study was carried out to examine the rate of loss and pattern of fragmentation of the tropical mountain forests in Cameron Highlands. Temporal remotely sensed data (SPOT-5 images) from 2000, 2005 and 2010 were used in a GIS to calculate landscape indices. Results showed increases in the class area (15,384 ha to 15,691 ha), number of patches (499 to 545) and patch density (1.8 to 2.0 patches/100 ha). The largest patch index increase (34% to 40%) was associated with the decrease in the area of mean patch (30 ha to 28 ha). The observed landscape trends indicate slight increase of forest loss and fragmentation, particularly during the years 2005-2010 periods. Approximately 2 % of the forest cover in Cameron Highland had been lost in 10 years, and a proportion of the remaining forests had been degraded as a result of agricultural practices. Combining landscape ecology and remote sensing has the potential to provide a significant way in assessing the dynamic of highland landscapes. It is suggested that conservation efforts should be focused on the management of the natural system and the management of the external influences particularly restoration and sustainable forest exploitation in the highland.

Initial results of the spatial distribution of rubber trees in Peninsular Malaysia using remotely sensed data for biomass estimate

The preservation and sustainable management of forest and other land cover ecosystems such as rubber trees will help addressing two major recent issues: climate change and bio-resource energy. The rubber trees are dominantly distributed in the Negeri Sembilan and Kedah on the west coast side of Peninsular Malaysia. This study is aimed to analyse the spatial distribution and biomass of rubber trees in Peninsular Malaysia with special emphasis in Negeri Sembilan State. Geospatial data from remote sensors are used to tackle the time and labour consuming problem due to the large spatial coverage and the need of continuous temporal data. Remote sensing imagery used in this study is a Landsat 5 TM. The image from optical sensor was used to sense the rubber trees and further classified rubber tree by different age.

Mangrove Changes Analysis by Remote Sensing and Evaluation of Ecosystem Service Value in Sungai Merbok’s Mangrove Forest Reserve, Peninsular Malaysia

Mangrove forests are an important ecosystem which provides socioeconomic value to humankind. Despite their great value, mangroves have one of the highest rates of degradation of any global habitat, which is about 1% of the existing area per year. In fact, the socioeconomic value and ecosystem services of mangroves as a natural product are underestimated. The ecosystem services provided by mangroves are often ignored by the ongoing process of mangrove conversion. This is a major reason why conservation of this ecosystem is not a popular alternative. Thus, the main objective of this study is to evaluate the changes in mangrove forests and valuation of their ecosystem services. SPOT 5 imageries of years 2000 and 2010 have been used for change detection analysis. The vegetation index such as NDVI and AVI and unsupervised classification technique were employed in image processing. In order to obtain the value of socioeconomic impact from the mangrove changes and biodiversity disturbances, the ecosystem service valuation (ESV) model was applied. Results show that the total value of the existing mangrove forest ecosystem service was RM1,901,859.84. The value per unit area is about RM 1,650.92 /ha. The total values of others were RM161, 33.2 (crop land) and RM3,107,500 (water bodies), respectively. It is evident that Sungai Merbok’s Mangrove Forest Reserve is very important for coastal ecology, where the orientation of mangrove ecosystem is huge and serves to provide essential services for the community. It also plays a crucial role in providing ecological balance to the coastal environment.

Application of lidar and optical data for oil palm plantation management in Malaysia

Proper oil palm plantation management is crucial for Malaysia as the country depends heavily on palm oil as a major source of national income. Precision agriculture is considered as one of the approaches that can be adopted to improve plantation practices for plantation managers such as the government-owned FELDA. However, currently the implementation of precision agriculture based on remote sensing and GIS is still lacking. This study explores the potential of the use of LiDAR and optical remote sensing data for plantation road and terrain planning for planting purposes. Traditional approaches use land surveying techniques that are time consuming and costly for vast plantation areas. The first ever airborne LiDAR and multispectral survey for oil palm plantation was carried out in early 2012 to test its feasibility. Preliminary results show the efficiency of such technology in demanding engineering and agricultural requirements of oil palm plantation. The most significant advantage of the approach is that it allows plantation managers to accurately plan the plantation road and determine the planting positions of new oil palm seedlings. Furthermore, this creates for the first time, digital database of oil palm estate and the airborne imagery can also be used for related activities such as oil palm tree inventory and detection of palm diseases. This work serves as the pioneer towards a more frequent application of LiDAR and multispectral data for oil palm plantation in Malaysia.

Hotspot occurrences classification using decision tree method: Case study in the Rokan Hilir, Riau Province, Indonesia

Application of geospatial and data mining techniques in forest fires research have resulted interesting and useful information in decision making related to the forest fires management. This paper presents a result of the study in applying the C4.5 algorithm on a forest fire dataset in the Rokan Hilir district, Riau Province, Indonesia. The dataset consists of hotspot occurrence locations, human activity factors, and land cover types. Human activity factors include city center locations, roads network and rivers network. The results were a decision tree which contains 18 leaves and 26 nodes with accuracy about 63.17%. Most of positive examples (the area with hotspot occurrences) and negative examples (no hotspot occurrences in the area) that are incorrectly classified by the model are located near rivers and roads.

Assessing soil physical properties variability and their impact on vegetation using geospatial tools in Kebbi State, Nigeria

Geospatial distribution of soil physical properties using Geographic Information System (GIS) is essential and efficient to site-specific farming and environmental management processes. This study attempts to evaluate the spatial distribution of soil physical properties and their impact on vegetation status in Kebbi State, Nigeria using the geospatial technique. A total of one hundred and fifty-six (156) soil samples were collected and analysed for soil organic matter (SOM), porosity, texture, Bulk Density (BD) and pH. The data were analyzed both statistically and geospatially to describe the spatial distribution of soil physical properties in the area. The Normalized Difference Vegetation Index (NDVI) was examined on Sentinel 2 satellite imagery. Results show the normal distribution for all parameters, except for soil texture and SOM indicating a positively skewed distribution. The spatial distribution of soil parameters results revealed the existence low to moderate spatial distribution. In comparison to the NDVI, soil properties significantly correlate with vegetation status of the area, except soil porosity which shows an inverse correlation. The study revealed that the use of geospatial technique accurately generates the spatial distribution maps of soil properties in the area and therefore, present a recommendable tool for sustainable land management and environmental management.

Generic rule-sets for automated detection of urban tree species from very high-resolution satellite data

Abstract The sustainable management and monitoring of urban forests is an important activity in the urbanized world, and operational approaches require information about the status of urban trees to determine the best strategy. One limitation in urban forest studies is the detection and discrimination of tree species using limited training data. Thus, this study focuses on developing generic rule sets from high-resolution WorldView-2 imagery in conjunction with spectral, spatial, colour and textural information for automated urban tree species detection. The object-based image analysis and its combination with statistical analysis of object features is utilized for this purpose. Results of attribute selection indicated that from 55 attributes, only 26 were useful to discriminate urban tree species, namely Messua ferrea L., Samanea saman and Casuarina sumatrana. Finally, the high overall accuracy, approximately 86.87% with kappa of 0.75 confirmed the transferability of the generic model.

Classifying Forest Species Using Hyperspectral Data in Balah Forest Reserve, Kelantan, Peninsular Malaysia

This study attempts to classify forest species using hyperspectral data for supporting resources management. The primary dataset used was AISA sensor. The sensor was mounted onboard the NOMAD GAF-27 aircraft at 2,000 m altitude creating a 2 m spatial resolution on the ground. Pre-processing was carried out with CALIGEO software, which automatically corrects for both geometric and radiometric distortions of the raw image data. The radiance data set was then converted to at-sensor reflectance derived from the FODIS sensor. Spectral Angle Mapper (SAM) technique was used for image classification. The spectra libraries for tree species were established after confirming the appropriate match between field spectra and pixel spectra. Results showed that the highest spectral signature in NIR range were Kembang Semangkok (Scaphium macropodum), followed by Meranti Sarang Punai (Shorea parvifolia) and Chengal (Neobalanocarpus hemii). Meanwhile, the lowest spectral response were Kasai (Pometia pinnata), Kelat (Eugenia spp.) and Merawan (Hopea beccariana), respectively. The overall accuracy obtained was 79%. Although the accuracy of SAM techniques is below the expectation level, SAM classifier was able to classify tropical tree species. In future it is believe that the most effective way of ground data collection is to use the ground object that has the strongest response to sensor for more significant tree signatures.