Problem Awareness and Project Objectives

The Touwian Creek is the mother river of Hsinchu, nourishing the life on this land for a long time. However, with industrial development and urban expansion, Touwian Creek is facing dual threats of invisible pollution (such as industrial and domestic wastewater) and visible pollution (such as river garbage and waste). These pollutants not only poison the water quality and sediment but also pose a serious threat to the health and quality of life of local residents. In response to this challenge, this project aims to protect and restore the environment of Touwian Creek through the introduction of advanced technology and deep collaboration with the community, combining empirical research and competency education to promote the sustainability of environmental protection.

This project is committed not only to current environmental issues but also hopes to establish a sustainable watershed governance model through technological innovation and community participation, ensuring the long-term health and prosperity of the Touwian Creek watershed.

Project Implementation Strategies and Approaches

1. Technology Adoption:

This project will employ various technologies for pollution monitoring and analysis, including real-time water pollution detection, sediment bio-toxicity testing, river waste image recognition systems, drone monitoring, and big data analytics. These technologies will help us timely monitor and analyze pollution sources and provide accurate pollution warning systems by integrating multiple data sources. Technical details will include the design principles of water quality and sediment monitoring systems and the selection and application of technologies based on different pollutant characteristics.

2. Expanding Participation:

The project will establish the “Touwian Creek Industry-Government-Academia Governance Platform” to promote cross-departmental collaboration and public-private cooperation. The platform will regularly hold seminars, community meetings, and workshops, allowing local residents, businesses, and government agencies to jointly participate in the governance and protection of Touwian Creek. The project will outline how to drive community involvement, ensure the sustainability of participation, and address potential conflicts and challenges.

3. Environmental Guardianship:

The long-term goal is to protect and restore the freshwater ecosystem of the Touwian Creek watershed, including mountains, forests, water channels, farmland, and underground aquifers. The project will promote concrete policy measures, such as separating irrigation systems from wastewater discharge systems to reduce the impact of pollution sources on Touwian Creek. This part will include specific policy promotion plans, expected environmental improvement outcomes, and how to achieve overall watershed ecological protection through the collaborative efforts of technology and community.

4. Sustainable Education:

During the execution of the project, a “Greater Hsinchu Watershed School” will be developed concurrently, integrating knowledge related to the Touwian Creek watershed into the education system through curriculum design, teaching material development, and science exhibitions. The project team will support curriculum development through empirical research. The educational content will specifically cover areas such as environmental governance, watershed management, and technology application.

Expected Benefits and Key Performance Indicators

This project is expected to generate the following benefits, assessed through key performance indicators:

(1) Pollutant Data Collection and Analysis: Eight coastal sampling points will be selected during both wet and dry seasons to collect twelve sediment samples for heavy metal and organic compound analysis, forming a complete pollutant database.

(2) Development and Application of Water Quality Monitoring Technology: Develop and deploy miniaturized water quality analysis technology for rapid on-site measurements with minimal water sample volume, establishing a standardized process for water quality analysis.

(3) Establishment of River Waste Image Database: Build a real-time image monitoring and AI recognition system to quickly detect abnormal pollution situations and immediately notify relevant units for processing, forming an effective environmental management tool.

(4) Development and Application of Water Quality Early Warning System: Integrate monitoring system data to develop a water quality index prediction system, combined with sediment correlation analysis to provide accurate warnings for water quality indicators.

(5) Establish the “Touwian Creek Industry-Government-Academia Governance Platform” to promote cross-departmental collaboration and public-private cooperation.

(6) Promote concrete policy measures, such as the separation of drinking water systems, irrigation systems, and wastewater discharge systems, to reduce the impact of pollution sources on Touwian Creek.

(7) Continue river sustainability education: promote watershed issues into schools and classrooms.

(8) Through the “Greater Hsinchu Watershed School” forum and watershed learning成果展示, expand educational impact and enhance environmental literacy among students and community residents.

Sub-Project Execution Status

"Correlation Analysis of Water Pollution in the Touqian River," was conducted by the Data Analysis & Interpretation Lab at the Department of Engineering & System Science, National Tsing Hua University. It aims to deeply investigate the pollution characteristics and water quality variation patterns of the Touqian River, a vital water source in the Hsinchu area. The project's content first involves integrating diverse, cross-disciplinary data sources, including: 1) Long-term, manual monitoring data from the Ministry of Environment's "National Environmental Water Quality Monitoring Information Network," covering key pollution indicators such as Dissolved Oxygen (DO), Biochemical Oxygen Demand (BOD), Chemical Oxygen Demand (COD), and Ammonia Nitrogen (NH3-N); 2) Climate observation data from the Central Weather Administration, used to analyze the potential impact of climatic factors like rainfall; 3) High-frequency, real-time monitoring data from the Hsinchu City Environmental Protection Bureau's "River Water Quality Automatic Monitoring Network."

Project Results:

The core achievements of this project are reflected in the systematic data analysis performed using the aforementioned databases. Key results include:

1. Pollution Index Trend Analysis: The team calculated the River Pollution Index (RPI) and conducted an in-depth analysis of its annual and quarterly correlations. This successfully depicted the long-term overall trends in Touqian River's water quality and identified pollution hotspots during specific seasons (e.g., wet and dry seasons).
2. Real-time Dynamic Grasp: By analyzing real-time automatic monitoring data, the project captured instantaneous changes and short-term trends in water quality, helping to clarify the occurrence patterns of specific pollution events.
3. Correlation Mining: The central objective of this research was to establish correlation models between various pollutant indicators (e.g., the relationship between DO, BOD, and Ammonia Nitrogen) and between pollution indicators and climatic factors. This lays the groundwork for subsequent pollution source tracing and water quality prediction.

"Touqian River Waste AI Image Recognition System," led by Professor Neng-Fu Huang's team from the Department of Computer Science at National Tsing Hua University, aims to address the challenge of monitoring waste pollution in the Touqian River, Hsinchu's vital waterway. Facing the difficulty of capturing real-time, comprehensive data on river waste, the project's core content involves developing and deploying an advanced AI image recognition system.

The project began by installing real-time cameras at key points in the Touqian River basin (such as Matagudao) to automatically collect images. To train a high-accuracy recognition model, the team integrated a diverse image database, including: 1) Actual images captured by on-site cameras (over 500 collected); 2) Public open-source waste datasets like TrashNet (over 4,700 collected); 3) Samples photographed by the team (over 300 collected). Furthermore, the project introduced Generative AI technology to create images of specific or rare types of waste, significantly enhancing the model's training effectiveness and recognition breadth.

Project Results:

This project successfully constructed an automated, intelligent platform for monitoring river waste. Key results include:

1. AI Recognition Core: An advanced AI recognition model was successfully trained and deployed. It can accurately identify various types of waste in images, such as plastic bottles, paper, and metal cans.
2. Real-time Monitoring and Alert System: The system automatically analyzes camera-feed images 24/7. Upon detecting waste, it immediately triggers an alert mechanism (e.g., via Line Notify) to inform relevant authorities for cleanup, drastically reducing response time.
3. Data Dashboard and Trend Analysis: The project developed a visual data dashboard. This platform not only displays real-time waste detection results but also features statistical analysis functions. It reveals waste hotspots, common types, and trend periods, providing a strong scientific basis for allocating cleaning resources and formulating pollution prevention strategies.

"Rapid and Miniaturized Water Quality Analysis Technology," is led by Professor Ruey-An Doong's team from the Institute of Analytical and Environmental Sciences at National Tsing Hua University. The project aims to solve fundamental challenges in timeliness and accuracy within current water quality monitoring systems. Traditional monitoring relies heavily on laboratory analysis, which suffers from significant drawbacks: the inability to immediately identify pollutant types and concentrations after sampling, time-consuming analysis (by which time pollutants have already been diluted and spread downstream), and limited analytical capacity. While automatic monitoring stations exist, they are expensive, difficult to maintain, and primarily designed for regular monitoring, lacking the capability to respond effectively to sudden pollution events (like illegal industrial dumping).

Therefore, the core content of this project is the research and development of advanced "rapid and miniaturized" detection technologies, aspiring to achieve the goal of "knowing pollution from a single drop of water." The team focuses on creating novel nanomaterials and sensing technologies, such as using graphene quantum dots or other specific nanomaterials, to build highly sensitive and selective chemical sensors for on-site, real-time analysis of specific heavy metals (like mercury ions) or harmful substances in water samples.

Project Results:

This project has successfully developed several innovative rapid water quality detection technologies that overcome the limitations of traditional methods. Key results include:

1. High-Sensitivity Nanosensors: The team successfully developed new sensing platforms based on nanomaterials (like graphene quantum dots). These sensors demonstrate excellent performance, such as high selectivity (not interfered with by other metal ions) and a limit of detection (LOD) reaching the nanomolar (nM) level when analyzing mercury ions (Hg2+), far surpassing traditional methods.
2. Real-time, On-Site Analysis Capability: Compared to conventional laboratory analyses that require hours or even days, the technology developed in this project drastically reduces detection time to within minutes (e.g., reaction completed in 10 minutes). This makes accurate "on-site analysis" possible at the very time and location a pollution event occurs.
3. Bridging the Gap in Sudden Pollution Monitoring: The real-time and miniaturized nature of this technology makes it ideal for tracking sudden, illegal discharge events. It overcomes the difficulty of tracing pollutants after they have been diluted or have flowed downstream, providing environmental agencies with a more agile and accurate tool for enforcement and response.