OES-Special Session TENCON 2025
Track B6F6 OES 1: OES Session
Room: F6. 505 Sepilok (Level 5)
14:35 – 15:15: Setting the Theme
Session Leader and Co-Chair: Melanie Olsen, Reefworks Project Director, Australian Institute of Marine Science, Townsville, Australia; Chair, OES Australia Chapter.
Co-Chair: Mal Heron, Adjunct Professor, James Cook University, Australia; Executive Vice President, OES.
14:35 – 15:15: Invited Paper: Intelligent Robotic Technologies for Sustainable Coastal Futures
Presenter: Prof. Mohd Rizal Arshad, Dean, School of Robotics, Xi’an Jiaotong–Liverpool University (XJTLU), Suzhou, China; OES Distinguished Lecturer.
15:15 Contributed Paper: Snail Parasite Detection via Convolutional Neural Network
Glenn C. Virrey, Gregorio L. Martin I, Alexander Clarkk Castro, Jullian Kristof Escubil, Aeon Fran Galvez, Jeremie Langreo, Melaijah Johannz Manansala and Arman Mapoy (University of Santo Tomas, Philippines)
Agriculture plays a crucial role in the Philippine economy, but its proximity to freshwater environments increases the risk of snail-borne diseases such as Fascioliasis, which severely affects both human and livestock health. Detecting parasitic infections in snails is essential to mitigating the spread of such diseases, yet traditional laboratory-based diagnostic methods are time-consuming, invasive, and resource-intensive. While recent studies have applied image recognition techniques to classify snail species, few have focused on detecting parasitic infections directly. This study proposes the use of Artificial Intelligence, specifically a Convolutional Neural Network (CNN), to facilitate mass detection of parasite-infected snails. Snail samples were collected by the Faculty of Pharmacy and imaged using a digital microscope at 175x magnification. These images were pre-processed and used to train and test a CNN model using a 70/30 data separation for training and implementation. The CNN achieved an average testing accuracy of 90.11% with an F1 score of 58.46 in detecting parasite manifestations. During live trials, the model maintained an accuracy rate of 91.21%, and a statistical comparison with expert parasitologist evaluations showed no significant difference (t = -2.39, p = 0.39), indicating the model's reliability as a diagnostic tool. The results demonstrate that CNN-based detection offers a faster, non-invasive alternative to traditional methods, which can significantly aid parasitologists in fieldwork and improve disease monitoring and control in agricultural areas. Future research is recommended to integrate automated imaging systems into diagnostic prototypes and develop real-time detection capabilities for broader deployment.
15:30 Contributed Paper: Decadal Land Cover Change Analysis and Forecasting in the Mahiga River Watershed Using GIS and CA-ANN Modeling
Leo Nhel D. Abao (Cebu Technological University, Philippines); Ricardo L. Fornis and Jonah Lee I. Bas (University of San Carlos, Philippines)
This study thoroughly examined land cover changes in the Mahiga River watershed using advanced remote sensing and geographic information system (GIS) techniques. Land cover data for the years 2000, 2023, and a future projection for the year 2030 were systematically mapped, processed, and analyzed in detail. The 2030 land cover was accurately predicted using the MOLUSCE plugin in QGIS, a modeling tool that incorporates both cellular automata and artificial neural network models for spatial prediction. Results revealed a significant and measurable increase in built-up areas from 9.33 km² in the year 2000 to 13.19 km² projected for 2030. In contrast, wooded land, covered land, and bare regions showed a consistent and notable decline, decreasing from 5.34 km², 3.51 km², and 0.17 km² in 2000 to 3.98 km², 1.16 km², and 0.02 km², respectively, by 2030. These observable trends underscore the ongoing transformation of the watershed due to urban expansion and land use change, emphasizing the importance of land cover monitoring for effective and sustainable environmental planning.
15:45 Contributed Paper: Evaluation of Water Hyacinth-Polypropylene Hybrid Adsorbents for Oil Spill Remediation in the San Cristobal River Station of Laguna de Bay
John Aries P Cruz (National University, Philippines); Ma. Kathleen Duran (National University, Philippines & Mapua University, Philippines); John Michael Dellova, Abby Joyce U Oquendo and Raiza H. Viernes (National University, Philippines)
Oil spills in freshwater ecosystems continue to pose critical environmental and economic threats, particularly in highly urbanized watersheds like Laguna de Bay. In this study, we developed and evaluated a hybrid adsorbent composed of water hyacinth (Eichhornia crassipes) and polypropylene for oil spill remediation. Three blend ratios (25:75, 50:50, 75:25) were fabricated and tested in both laboratory and field settings to determine their oil adsorption capacity (OAC) and oil recovery efficiency (ORE). Laboratory analysis followed ASTM F726-12 protocols using gravimetric methods and Soxhlet extraction. The 75:25 blend (PP: WH) achieved the highest OAC (1.631 g/g) and ORE (90.81%), confirmed through statistical tests (ANOVA, Kruskal-Wallis, p < 0.05). Field deployment in the San Cristobal River validated its performance under real conditions. Though more costly per gram of oil adsorbed than commercial polypropylene pads, the hybrid adsorbent offers key advantages in biodegradability and local material sourcing. These results demonstrate the feasibility of developing eco-efficient hybrid materials for oil remediation in tropical aquatic systems.
16:00 – 16:30: Refreshment
16:30 Contributed Paper: GIS-Based Flood Vulnerability Map Using the Fuzzy Analytical Hierarchy Process and Response Surface Method: Flood Risk Reduction Program for Valenzuela City
Mark B. Ondac and Dante L Silva (Mapúa University, Philippines); Sheina Pallega (National University, Philippines); Jimmy G. Catanes (Commission on Higher Education, Philippines)
Flooding has been a persistent issue in Valenzuela city due to the overflowing of two major channels such as Meycauayan River and Tullahan River. This study aimed to understand the complex interaction of various factors to flood vulnerability in priority catchments identified using integrated approach. Two priority subbasins were practically selected and clipped from the watershed based on the peak flow simulated using HEC-HMS. Normalized weights computed from FAHP revealed that elevation having a weight of 30.70% is the most influential factor that contributes to flooding while the least is population density with a weight of 2.50%. Final overlay map generated using GIS highlighted that Barangays Marulas, Gen. T. De Leon, and Ugong were classified as highly to very highly vulnerable to flooding while Barangays Malinta, Parada, and Karuhatan were under low to very low vulnerability. RSM provided a better understanding in the effect of the individual parameters and combined effect of parameters on flood vulnerability. 2FI-model emerged as the best model with adjusted R2 and predicted R2 values of 0.8783 and 0.5772 respectively. The interaction between elevation and flow discharge was observed to be significant based on the regression model that was used to generate the contour plots. The illustration of the combined effect of elevation and flow discharge to flood vulnerability helped in the development of flood mitigation strategies. Declogging of canals and retrofitting of drainage system was suggested to Barangays Marulas and Gen. T. de Leon due to low elevation and poorly, clogged drainage. Elevated areas such as Barangay Ugong remained vulnerable to flooding because of large volume of water that must be controlled by installing retention basins and ensuring that upstream drainage is connected to a properly sized downstream drainage.
16:45 Contributed Paper: Utilization of Mask Region-Based Convolutional Neural Network (RCNN) for Fish Body Length and Height Measurement
John Peter Austria, Kierra Manuel Francisco, Lucas Andrew Lumotan, Charlos Dhanniel Macalinao and Catherine Montalbo (National University Philippines, Philippines); Herbert V Villaruel (De La Salle University, Philippines)
This project leverages computer vision and deep learning technologies to enhance fish sampling processes, providing valuable support to various sectors, including the academe, agriculture, aquaculture, and scientific research. It is specifically developed to assist the manual sampling procedures of the National Freshwater Fisheries Technology Center (NFFTC), focusing on the precise measurement of fish length, width, and weight. Accurate morphometric data are essential for monitoring fish growth, assessing health, and improving yield projections-critical components in both research and aquaculture operations. Traditional manual measurements are often prone to inconsistencies and human error, which can affect data reliability. By integrating intelligent technologies, the system automates data collection, resulting in improved accuracy, efficiency, and consistency. This innovation reduces human intervention while accelerating the sampling process, making it more scalable and repeatable. Overall, the project introduces a data-driven, technology-enhanced solution that strengthens the reliability of fish sampling and supports evidence-based decision-making in fisheries management, education, and sustainable aquaculture practices.
17:00 Contributed Paper: Assessing Tidal Energy Potential in the Visayas: Viability of the San Bernardino, San Juanico, and Cebu Straits
Justin Ricafort and King Harold A Recto (Ateneo de Manila University, Philippines)
By laying the groundwork for sustainable tidal energy infrastructure, this study contributes to advancing the Philippines' renewable energy portfolio and supports the global transition to clean energy solutions. With the country's extensive coastline and rising energy demands, tidal energy presents a largely underutilized yet promising resource that can address both local and global energy challenges. Tidal energy is highly predictable, stable, and environmentally friendly, offering a reliable alternative to conventional energy sources like coal and natural gas, which are often subject to price volatility and environmental concerns. The study focuses on the Visayas region, a prime candidate for tidal energy development due to its dense population centers, major seaports, and high energy consumption. Integrating tidal energy into the national grid could reduce the Philippines' reliance on fossil fuels and lower carbon emissions. This research explores tidal stream turbines' potential, addressing challenges like maritime traffic and environmental considerations, while suggesting strategies for overcoming these obstacles. Ultimately, this study supports the transition to a cleaner, sustainable energy future.
17:15 – 18:00: OES Networking
OES members and Invited Guests: Socialising, collaborating, and mentoring
18:00 Adjourn