Multi-scale/multi-level geographic object-based image analysis (MS-GEOBIA) methods are becoming widely-used in remote sensing because single-scale/single-level (SS-GEOBIA) methods are often unable to obtain an accurate segmentation and classification of all land use/land cover (LULC) types in an image. However, there have been few comparisons between SS-GEOBIA and MS-GEOBIA approaches for the purpose of mapping a specific LULC type, so it is not well understood which is more appropriate for this task. In addition, there are few methods for automating the selection of segmentation parameters for MS-GEOBIA, while manual selection (i.e., trial-and-error approach) of parameters can be quite challenging and time-consuming. In this study, we examined SS-GEOBIA and MS-GEOBIA approaches for extracting residential areas in Landsat 8 imagery, and compared naï ve and parameter-optimized segmentation approaches to assess whether unsupervised segmentation parameter optimization (USPO) could improve the extraction of residential areas. Our main findings were: (i) the MS-GEOBIA approaches achieved higher classification accuracies than the SS-GEOBIA approach, and (ii) USPO
This study uses a spatially-explicit land-use/land-cover (LULC) modeling approach to model and map the future (2016-2030) LULC of the area surrounding the Laguna de Bay of Philippines under three different scenarios: 'business-as-usual', 'compact development', and 'high sprawl' scenarios. The Laguna de Bay is the largest lake in the Philippines and an important natural resource for the population in/around Metro Manila. The LULC around the lake is rapidly changing due to urban sprawl, so local and national government agencies situated in the area need an understanding of the future (likely) LULC changes and their associated hydrological impacts. The spatial modeling approach involved three main steps: (1) mapping the locations of past LULC changes; (2) identifying the drivers of these past changes; and (3) identifying where and when future LULC changes are likely to occur. Utilizing various publically-available spatial datasets representing potential drivers of LULC changes, a LULC change model was calibrated using the Multilayer Perceptron (MLP) neural network algorithm. After calibrating the model, future LULC changes were modeled and mapped up to the year 2030. Our modeling results showed that the 'built-up' LULC class is likely to experience the greatest increase in land area due to losses in 'crop/grass' (and to a lesser degree 'tree') LULC, and this is attributed to continued urban sprawl.
Vegetation analysis of “muyong” was done to determine the species composition and diversity of tree species in Barangays (Brgy.) Amganad and Poitan. Trees growing within the established plots (10 m x 10 m) were identifiedand described. Tree diameter at breast height (DBH), merchantable height, totalheight, and diameter of the crown were measured. Brgy. Amganad has flat tosteep slopes and elevations ranging from 1120 to 1240 m above sea level (masl),with the “muyong” located at 1240 masl. On the other hand, Brgy. Poitan hasflat to steep configuration and elevations ranging from 990 to 1200 masl, withthe “muyong” found at 1200 masl. Vegetation was comprised of 67 speciesbelonging to 22 families. The tree species common in the dominant layers offorest stands in both barangays (local communities) were pine (Pinus kesiya),tabangawon/tabangawen (Weinmannia luzoniensis) and palayon (Lithocarpussubmonticolus). Height of dominant trees in Brgy. Amganad ranged from 14to 20 m while that in Brgy. Poitan ranged from 17 to 30 m. “Muyong” forest inBrgy. Poitan showed higher diversity compared with the “muyong” forest in Brgy.Amganad. However, the latter had higher species evenness value. In general,species diversity in the two barangays was not very dissimilar.Keywords: Biodiversity, indigenous tree species, muyong forest, descriptive design, tropical forest, Philippines
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.