Enrico C. Paringit

Integrated Landsat Image Analysis and Hydrologic Modeling to Detect Impacts of 25-Year Land-Cover Change on Surface Runoff in a Philippine Watershed

Landsat MSS and ETM+ images were analyzed to detect 25-year land-cover change (1976–2001) in the critical Taguibo Watershed in Mindanao Island, Southern Philippines. This watershed has experienced historical modifications of its land-cover due to the presence of logging industries in the 1950s, and continuous deforestation due to illegal logging and slash-and-burn agriculture in the present time. To estimate the impacts of land-cover change on watershed runoff, land-cover information derived from the Landsat images was utilized to parameterize a GIS-based hydrologic model. The model was then calibrated with field-measured discharge data and used to simulate the responses of the watershed in its year 2001 and year 1976 land-cover conditions. The availability of land-cover information on the most recent state of the watershed from the Landsat ETM+ image made it possible to locate areas for rehabilitation such as barren and logged-over areas. We then created a “rehabilitated” land-cover condition map of the watershed (re-forestation of logged-over areas and agro-forestation of barren areas) and used it to parameterize the model and predict the runoff responses of the watershed. Model results showed that changes in land-cover from 1976 to 2001 were directly related to the significant increase in surface runoff. Runoff predictions showed that a full rehabilitation of the watershed, especially in barren and logged-over areas, will be likely to reduce the generation of a huge volume of runoff during rainfall events. The results of this study have demonstrated the usefulness of multi-temporal Landsat images in detecting land-cover change, in identifying areas for rehabilitation, and in evaluating rehabilitation strategies for management of tropical watersheds through its use in hydrologic modeling.

A novel approach for vegetation classification using UAV-based hyperspectral imaging

Abstract The use of unmanned aerial vehicle (UAV)-based spectral imaging offers considerable advantages in high-resolution remote-sensing applications. However, the number of sensors mountable on a UAV is limited, and selecting the optimal combination of spectral bands is complex but crucial for conventional UAV-based multispectral imaging systems. To overcome these limitations, we adopted a liquid crystal tunable filter (LCTF), which can transmit selected wavelengths without the need to exchange optical filters. For calibration and validation of the LCTF-based hyperspectral imaging system, a field campaign was conducted in the Philippines during March 28–April 3, 2016. In this campaign, UAV-based hyperspectral imaging was performed in several vegetated areas, and the spectral reflectances of 14 different ground objects were measured. Additionally, the machine learning (ML) approach using a support vector machine (SVM) model was applied to the obtained dataset, and a high-resolution classification map was then produced from the aerial hyperspectral images. The results clearly showed that a large amount of misclassification occurred in shaded areas due to the difference in spectral reflectance between sunlit and shaded areas. It was also found that the classification accuracy was drastically improved by training the SVM model with both sunlit and shaded spectral data. As a result, we achieved a classification accuracy of 94.5% in vegetated areas.

Energy potential metric for rapid macro-level resource assessment of tidal in-stream energy in the Philippines

This paper presents preliminary efforts in developing a rapid evaluation tool for tidal in-stream energy (TISE) in the Philippines. We study the possibility of using an energy density metric based on the sea surface elevation (SSE) or tide height difference at the boundaries of a site of interest as a gauge for the TISE potential of that site. Results show good correlation with high potential sites and the proposed metric. Verde Island Passage was assessed, through a combination of DELFT3D simulation and Matlab-based power computations, to have four potential TISE sites with a total energy density of 271.90 kW-h /m2 in a month.

Estimating the Gauss-Markov Random Field parameters for remote sensing image textures

Although there are some recent characterizations of Multivariate Gauss Markov-Random Field (MGMRF) models, these are limited to cases where the interaction matrix coefficients are modeled with some special form. We extend the modeling and parameter estimation for the interaction matrix coefficients for a general anisotropic MGMRF. Although the MGMRF is a natural generalization of its univariate counterpart, there are new problems which hold in the multivariate case but do not hold in the univariate case. The results show that in general, there is an improvement in the classification performance in the generalized model compared to competing MGMRF models.

Validation of the separability measure for Rhizophoraceae and Avicenniaceae using point density distribution from lidar

The extent at which mangrove forest characterization can be done through utilization of Light Detection and Ranging (LiDAR) data is investigated in this paper. Particularly, the ability of LiDAR parameters, such as its point density to provide height and structural information was explored to supplement manual field surveys which are time-consuming and requires great effort. Point cloud information was used to produce separability measure within a mangrove forest. The study aims to validate the point density distribution curves (PDDC) that were established to characterize the structural attributes between Rhizophoraceae and Avicenniaceae. The applicability of the PDDC was applied to fifteen (15) 5x5 sample plots of pure Rhizophoraceae and fifteen (15) 5x5 sample plots of pure Avicenniaceae in a one hectare (1ha) natural riverine mangrove forest. 15 out of 15 plots were correctly discriminated as Rhizophoraceae; however, Avicenniaceae plots were not correctly discriminated using the established separability measure. This study had determined that the two mangrove families are difficult to separate in terms of point density distribution alone. Enhancement of the PDDC as a separability measure should be improved to pave way for a more sensitive and robust way to separate the two families.

Addressing LiDAR overlap for Diameter at Breast Height estimation using a Point-Cloud Processing Software

LiDAR Overlap is the area that is common to two or more flight lines. This is essential to ensure the continuity of data as the acquisition moves from one flight line to another. Looking into overlaps is important when doing DBH Estimation using point cloud data because it doubles the density of points in the overlap region. To remove this effect when determining the DBH of a forest area, the LiDAR data was processed using a point-cloud processing software. The processes include separating flight lines using the GPS time when the points were acquired. After separating, the number of points in the overlap region were decreased by removing excess points within the area of twice the point spacing. The parameters needed for DBH estimation were then obtained. The absolute number of points in the whole overlap area was originally 4,960,726 after decreasing the number of points, it was reduced to 1,479,884. The number of points would have an effect on DBH estimation because the values obtained were significantly different at 95% level of confidence.

Diameter at breast height estimation in Mt. Makiling, Laguna, Philippines using metrics derived from airborne LiDAR data and Worldview-2 bands

Airborne LiDAR is fast becoming an innovation for forest inventory. It aids in obtaining forest characteristics in areas or cases where actual field inventory would be very tedious. This study aims to estimate diameter at breast height (DBH) using airborne LiDAR point-cloud parameters with Worldview-2 satellite images, and to validate these with actual measurements done in the field. The study site is a field plot with forest inventory at Mt. Makiling, Laguna, Philippines that was surveyed into 20m, 10m and 5m subplots or grids. The estimation of DBH was carried out by extracting the said parameters from the LiDAR point-cloud, and extracting different bands from the Worldview image and performing linear and log-linear regression of these values. The regressions were done in four different cases, namely: LiDAR parameters without intensity (case1), LiDAR parameters without intensity with Worldview bands (case 2), intensity of LiDAR points (case 3), and LiDAR parameters with intensity and Worldview bands (case 4). From these it was found that the best case for estimating DBH is with the use of LiDAR parameters with intensity and Worldview bands in a 10x10 grid, in Log-Linear regression with a root mean squared error of 1.96 cm and an adjusted R2 value of 0.65. This was further improved through stepwise regression, and adjusted R2 value was 0.71.

Estimation of the Separable MGMRF Parameters for Thematic Classification

Because of its ability to describe interdependence between neighboring sites, the Markov Random Field (MRF) is a very attractive model in characterizing correlated observations (Moura and Balram, 1993) and it has potential applications in areas of remote sensing, such as spatio-temporal modeling and machine vision. In this study, we model image random field conditional to the texture label as a Multivariate Gauss Markov Random Field (MGMRF); whereas; the thematic map is modeled as a discrete label MRF (Li, 1995). The observations in the Gauss Markov Random Field (GMRF) are distributed with the Gaussian distribution.