Application of the DPSIR model in marine ecological environmental assessment Open Access

Global environmental issues, in particular, threaten the environmental safety of all countries (Zhang et al. 2021a, b). As a result, ecological environmental safety has gradually become an important aspect of national security for many countries (Vasseur & Cossu-Leguille 2003; Popkova et al. 2016; Ermakov et al. 2021; Kishor et al. 2021; Al-Tohamy et al. 2022). Environmental safety primarily refers to the condition of environmental elements and their structures within a region, as well as the functionality and regulatory capacity of these elements within the acceptable safety limits (Enyinna et al. 2022; Provodina et al. 2024). Environmental pollution and damage can harm the natural support systems that humanity depends on for survival and development, ultimately jeopardizing human health. According to the United Nations Convention on the Law of the Sea (UNCLOS), marine environmental pollution refers to harmful effects caused by the direct or indirect discharge of substances or energy into the marine environment due to human activities (UNCLOS 1982). These include damages to marine biological resources, harm to human health, interference with fishing and other legitimate activities, destruction of the seawater value, or reduction in marine environmental quality. Marine environmental pollution is the most urgent problem threatening marine environmental safety (Halpern et al. 2008; Kushwaha et al. 2024) and has led to the loss of self-purification capacity in some marine areas, frequent eutrophication in others, significant reductions in dissolved oxygen, and the emergence of hypoxic zones, causing large-scale fish deaths (Gutiérrez-Bravo et al. 2024; Lai et al. 2024). Insufficient dissolved oxygen in seawater has expanded the ‘dead zones’ (or marine desert areas) (Dybas 2005; Diaz & Rosenberg 2008; Walsh et al. 2009; Karstensen et al. 2015; Aniebone et al. 2024; Pontius & McIntosh 2024), posing a significant threat to the Earth and representing a major challenge for marine environmental protection. Over the past decade, the number of dead zones in the ocean has increased to nearly 400 (Rabalais et al. 2002; Dybas 2005; Diaz & Rosenberg 2008). Factors contributing to the destruction of the marine environment include human activity disturbances, overdevelopment, environmental pollution, habitat destruction, and threats to marine biodiversity. For example, the fish death zone in the Gulf of Mexico is due to large amounts of nitrogen and phosphorus present in agricultural fertilizers being discharged into the Gulf from the Mississippi River Basin. In 2002, the fish death zone in the Gulf of Mexico reached a record 22,000 km² (Shih 2015). The NOAA reported that the dead zone in the Gulf of Mexico in 2024 covers about 6,705 square miles, roughly the size of New Jersey (McKinney et al. 2021; NOAA 2024; Pontius & McIntosh 2024). Therefore, maintaining the safety of the marine ecosystem is a long-term complex process that requires finding a suitable and effective assessment method.

This study utilizes the ecological security assessment model that integrates human, economic, and social development with natural ecological environment protection – the ‘DPSIR (driving force–pressure–state–impact–response) model’ as the basis to integrate the ecological risk assessment, ecological health assessment, and policy analysis methods, to evaluate the ecological safety of the Anping Port in the Tainan City of Taiwan from 2015 to 2021. It also proposes an ecological security assessment framework for coastal areas.

In this study, the marine ecological security assessment framework was constructed based on the DPSIR model, which helps to systematically assess the ecological security status of the Anping Port in the Tainan City by analyzing the driving forces, pressures, states, impacts, and responses. In ecological safety assessment, the DPSIR method provides a widely applied framework that comprehensively analyses the interactions within ecosystems through the following five steps: ‘driving forces–pressures–states–impacts–responses.’ This approach is applicable not only to aquatic systems but also to terrestrial, atmospheric, and global-scale ecological issues (Gari et al. 2015; Zhang et al. 2021a, b; Carnohan et al. 2023; Benzina et al. 2024; Jing et al. 2024; Rochman et al. 2024). Maintaining ecological safety is not only the foundation and core of achieving sustainable development, but also an important way of providing scientific evidence for ecological environmental management and decision-making (Ruan et al. 2019).

The DPSIR model provides a clear theoretical basis for ecological safety assessment and has become an important reference for subsequent research works in this field. The DPSIR model is widely used in the selection and design of indicators for ecological safety research (Li et al. 2012; Kristiadi et al. 2022; Guo et al. 2023; Zhou et al. 2024), helping to more effectively quantify environmental changes and assess the safety status of ecosystems.

With the introduction of the 17 sustainable development goals (SDGs) by the United Nations in 2015, which was aimed at achieving a coordinated development in economic, social, and environmental dimensions to ensure a dynamic balance and sustainable use of Earth's ecosystems – single environmental pollution indicators are no longer sufficient to support sustainable development decisions (Li et al. 2024a, b). To provide a more comprehensive basis for supporting sustainable development, the Organization for Economic Co-operation and Development (OECD) and the UN Environment Programme developed the pressure–state–response (PSR) conceptual model, which includes economic, social, environmental, and organizational dimensions, and further evolved into the DPSIR model, encompassing indicators in five key areas (Tong et al. 2000; Brunhara et al. 2023; Xi et al. 2023; Li et al. 2024a, b).

When selecting appropriate ecosystem measurement methods, it is essential to balance the representativeness of the system with sufficient simplicity to achieve effective and efficient monitoring and modeling systems (Niemeijer & de Groot 2008; Maes et al. 2016; Halder et al. 2024; Leonhardt et al. 2024). The characteristics of the ecosystem and the availability of collected data must be considered when choosing these indicators (shown in Table 1).

The ecological security indicators selected in this study were based on the literature review and expert interviews and were determined in combination with the Delphi method and the entropy weighting method to ensure the consistency and credibility of the indicators. The construction of the indicator system follows the principles of rationality, measurability, and operability to ensure the accuracy of the results. The modification process may involve in-depth discussions with experts and the application of Delphi techniques to ensure the appropriateness of indicator selection and verification in the specific research context (Belton & Stewart 2002). Relevant literature data, as well as environmental survey data from port authorities and civil organizations regarding the Anping Port, were collected to organize the indicators affecting marine ecological safety and to construct a marine ecological safety assessment system for expert consultation.

This study uses time-series data from 2015 to 2021 as the primary research material. Analyzing data over this 7-year period allows for better observation of variations between different years and hence a better understanding of the origins and development of events or trends. Choosing this time frame also ensures that the study has a sufficient historical depth to conduct meaningful analysis of the evolution of the marine ecological environment. Additionally, using a longer time-series helps mitigate the impact of short-term fluctuations on the research results, leading to more stable and reliable outcomes. Overall, selecting the period from 2015 to 2021 enables a deeper understanding of the development trends in the marine environment and provides a comprehensive temporal context to support comparisons and analyses of changes over different periods (Table 2).

• The first level is the criterion layer or Level A, which includes five attributes of the safety index, labeled as A1–A5.

• The second level is the factor layer or Level B, which categorizes the indicators into several dimensions, labeled as B1–B13.

• The third level is the indicator layer or Level C, consisting of environmental impact indicators based on the actual environmental conditions of the Anping Port, labeled as C1–C67. According to this hierarchical structure, the environmental indicators at the indicator layer are compared and analyzed with the contents of the other two layers.

Table 2

Marine ecological security assessment index system.

Criterion layer .	Factor layer .	Average level .	Index layer .	Average .	Standard deviation .	Facets average .
D (A1)	Economic development (B1)	4.47	Number of ships registered (C1)	4.14	0.38	4.42
			Number of ships entering and leaving the port (C2)	4.71	0.49
			Vehicles entering and leaving the port area (C3)	4.43	0.53
			Cargo handling volume (C4)	4.71	0.49
			Cargo throughput (C5)	4.71	0.49
			Number of inbound and outbound passengers (C6)	4.29	0.49
			Free Trade Port Area (Trade Value) (C7)	4.14	0.38
			Annual oil consumption in the port area (C8)	4.57	0.53
			Fuel consumption (C9)	4.57	0.53
	Social development (B2)	4.29	Annual water consumption in the port area (C10)	4.43	0.53
			Reclaimed water use in the Anping Port operation area (C11)	4.14	0.69
	Demographic change (B3)	4.14	Population density (C12)	4.14	0.9
P (A2)	Anthropogenic (B5)	4.49	Greenhouse gas emissions of the Anping Port Operation Department (C13)	4.43	0.53	4.49
			Amount of waste oil and sewage generated from ships (C14)	4.57	0.53
			Ship waste oil and sewage ship volume (C15)	4.43	0.53
			Fuel consumption of ships in the port area (C16)	4.43	0.79
			Garbage/Port Waste (waste generated and treated) (C17)	4.57	0.53
S (A3)	Environmental quality and water quality (B7)	4.29	Perspective in water (C18)	4.29	0.49	4.36
			PH (C19)	4.14	0.69
			TSS (C20)	4.29	0.49
			Dissolved oxygen (DO) (C21)	4.29	0.49
			Orthophosphate (⁠⁠) (C22)	4.43	0.53
			Zinc (Zn) in seawater (C23)	4.29	0.76
			Arsenic in seawater (As) (C24)	4.29	0.76
	Ambient quality, air quality (B8)	4.43	TSP (C25).	4.57	0.53
			PM2.5 (C26)	4.57	0.53
			PM10 (C27)	4.14	0.9
			SO2 (C28)	4.43	0.53
			NO2 (C29)	4.43	0.53
			NO (C30)	4.43	0.53
			NOx (C31)	4.43	0.53
I (A4)	Economic activity (B9)	4.14	Dredging in the port area (C32)	4.14	0.69	4.19
			Number of accidents (C33)	4.14	0.69
	Land shock (B10)	4	Land-based mobile source control (number of penalties) (C34)	4	0.82
	Environmental impact (B11)	4.29	Water waste removal (C35)	4.43	0.53
			Number of water waste removal (C36)	4	0.82
			Household waste removal volume (C37)	4.43	0.53
R (A5)	Urgent treatment (B12)	4.36	Exhortation to improve (C38)	4.29	0.49	4.38
			Number of notifications (C39)	4.43	0.53
	Environmental policy (B13)	4.39	Backfill volume in the port area (C40)	4.29	0.76
			Proportion of closed storage (C41)	4.43	0.53
			Carbon reduction of pipeline operation (automated gate) (C42)	4.43	0.53
			Ships running on low-polluting fuels (C43)	4.57	0.53
			Number of loading and unloading control facilities (C44)	4.29	0.76
			Number of dust filters (C45)	4.43	0.53
			Shore power usage (C46)	4.29	0.49

Criterion layer .	Factor layer .	Average level .	Index layer .	Average .	Standard deviation .	Facets average .
D (A1)	Economic development (B1)	4.47	Number of ships registered (C1)	4.14	0.38	4.42
			Number of ships entering and leaving the port (C2)	4.71	0.49
			Vehicles entering and leaving the port area (C3)	4.43	0.53
			Cargo handling volume (C4)	4.71	0.49
			Cargo throughput (C5)	4.71	0.49
			Number of inbound and outbound passengers (C6)	4.29	0.49
			Free Trade Port Area (Trade Value) (C7)	4.14	0.38
			Annual oil consumption in the port area (C8)	4.57	0.53
			Fuel consumption (C9)	4.57	0.53
	Social development (B2)	4.29	Annual water consumption in the port area (C10)	4.43	0.53
			Reclaimed water use in the Anping Port operation area (C11)	4.14	0.69
	Demographic change (B3)	4.14	Population density (C12)	4.14	0.9
P (A2)	Anthropogenic (B5)	4.49	Greenhouse gas emissions of the Anping Port Operation Department (C13)	4.43	0.53	4.49
			Amount of waste oil and sewage generated from ships (C14)	4.57	0.53
			Ship waste oil and sewage ship volume (C15)	4.43	0.53
			Fuel consumption of ships in the port area (C16)	4.43	0.79
			Garbage/Port Waste (waste generated and treated) (C17)	4.57	0.53
S (A3)	Environmental quality and water quality (B7)	4.29	Perspective in water (C18)	4.29	0.49	4.36
			PH (C19)	4.14	0.69
			TSS (C20)	4.29	0.49
			Dissolved oxygen (DO) (C21)	4.29	0.49
			Orthophosphate (⁠⁠) (C22)	4.43	0.53
			Zinc (Zn) in seawater (C23)	4.29	0.76
			Arsenic in seawater (As) (C24)	4.29	0.76
	Ambient quality, air quality (B8)	4.43	TSP (C25).	4.57	0.53
			PM2.5 (C26)	4.57	0.53
			PM10 (C27)	4.14	0.9
			SO2 (C28)	4.43	0.53
			NO2 (C29)	4.43	0.53
			NO (C30)	4.43	0.53
			NOx (C31)	4.43	0.53
I (A4)	Economic activity (B9)	4.14	Dredging in the port area (C32)	4.14	0.69	4.19
			Number of accidents (C33)	4.14	0.69
	Land shock (B10)	4	Land-based mobile source control (number of penalties) (C34)	4	0.82
	Environmental impact (B11)	4.29	Water waste removal (C35)	4.43	0.53
			Number of water waste removal (C36)	4	0.82
			Household waste removal volume (C37)	4.43	0.53
R (A5)	Urgent treatment (B12)	4.36	Exhortation to improve (C38)	4.29	0.49	4.38
			Number of notifications (C39)	4.43	0.53
	Environmental policy (B13)	4.39	Backfill volume in the port area (C40)	4.29	0.76
			Proportion of closed storage (C41)	4.43	0.53
			Carbon reduction of pipeline operation (automated gate) (C42)	4.43	0.53
			Ships running on low-polluting fuels (C43)	4.57	0.53
			Number of loading and unloading control facilities (C44)	4.29	0.76
			Number of dust filters (C45)	4.43	0.53
			Shore power usage (C46)	4.29	0.49

Note: Security trend, + indicates that the indicator is positively correlated, the larger the value, the better the evaluation result; − indicates that the indicator is negatively correlated, the smaller the value, the better the evaluation result.

Data sources: ‘Tainan City Government Civil Affairs Bureau, 2015–2021,’ ‘Port Bureau of the Ministry of Transportation, 2015–2021,’ ‘Taiwan Port Co., Ltd, 2015–2021,’ ‘Port Technology Research Center, 2015–2021,’ ‘Harbor Environmental Information Network, 2015–2021,’ ‘Transportation Research Institute of the Ministry of Transport, 2015–2021,’ ‘Anping Port Environmental Report, 2015–2021,’ ‘Airport Statistical Annual Report, 2015–2021,’ ‘Ministry of Transport Central Meteorological Administration, 2015–2021,’ ‘Transportation Research Institute of the Ministry of Transportation, 2015–2021,’ ‘Tainan City Government Data Open Platform, 2015–2021,’ and ‘Building and Construction Administration of the Ministry of Interior, 2015–2021’ (compiled by this study).

View Large

Next is ‘Environmental quality (Water Quality Conditions),’ having a weight of 0.149, which ranks second in terms of weight. This indicates that water quality is crucial for the ecological environment, ranking just below policies. Water quality directly affects the marine life health and ecosystem stability, making it a priority for attention and improvement.

Conversely, ‘Land impact’ has the lowest weight, meaning that in the Anping Port area, the negative impact of land activities on the ecological environment is relatively small, or relevant management measures are already relatively mature, resulting in a lower weight for this dimension in the ecological environment assessment.

Overall, the ranking of these weights reflects the varying degrees of importance of each dimension to the ecological environment.

‘Population density’ reflects the concentration of the population in a region. High population density leads to increased resource consumption and puts pressure on the ecological environment (Motesharrei et al. 2016; Meirelles et al. 2020; Wilmoth et al. 2023). The Taichung Port has implemented measures such as entry and exit control, identity and destination checks for individuals and vehicles entering the port area, and increased fees for entry into the port area. The government can manage population influx through various ways to mitigate the impact of high population density on the environment.

‘Operating vehicles in and out of the port area’ reflects the intensity of transportation activities within the port area. A higher number of vehicles entering and leaving the port area indicates more active cargo transport and economic activities, which can impact traffic and the environment, and lead to a poor air quality due to the emission of harmful gases such as carbon dioxide, nitrogen oxides, and particulate matter.

‘Annual fuel consumption in the port area’ reflects the total amount of the fuel oil consumed in the port area each year. Higher fuel consumption indicates more traffic and industrial activities. The fuel consumption at the Anping Port has shown an overall upward trend from 2015 to 2021, but there was a decrease in 2020 due to the COVID-19 pandemic. The Anping Port, as a transportation hub, drives more economic development and cargo transport each year but also results in increased air pollution and continuous emission of greenhouse gases.

‘Fuel oil consumption’ is related to the energy consumption of transportation tools and facilities within the port area. The amount of the fuel oil consumed is an important indicator of the impact of economic activities on the environment. As Anping Port's economy grows, in 2015, Anping Port's fuel oil consumption was 1 million tons, and by 2021, it increased to 1.5 million tons, accounting for 70% of the energy consumption of transportation tools and facilities within the port area. Carbon dioxide emissions from the burning of fuel oil account for >60% of Anping Port's total greenhouse gas emissions.

Finally, ‘annual water consumption in the port area’ reflects the amount of water resources used in the port area each year. Water resource consumption is closely related to industrial, commercial, and residential water use. Data on annual water consumption at the Anping Port from 2015 to 2021 show that the average annual water use in the port area was 1 million cubic meters. Analyzing these data provides insights into the changes in water resource use efficiency in the port area. These interrelated issues require a comprehensive consideration of all factors to develop reasonable and effective plans to promote the sustainable development of the port area.

The analysis results (Table 4) indicate that the average scores of the pressure indicators reflect their relevance to ecological safety and priority indicators. For example, when evaluating environmental pollution pressure, experts focused on indicators such as ‘ship waste oil and wastewater generation.’ These indicators represent the pollution and harm caused to marine ecology by economic activities and other driving forces. Experts recommend prioritizing these indicators that have adverse effects on marine ecology. Figure 2 shows the analysis of weighted indicators for each aspect. Among the environmental pollution pressure indicators, the top two are ‘greenhouse gas emissions from Anping Port Operations’ (ranked 1) and ‘fuel oil consumption by ships in the port area’ (ranked 2).

‘Greenhouse gas emissions from Anping Port Operations’ refer to the total amount of greenhouse gases emitted from various activities managed by Anping Port Operations within the Anping Port area, including port machinery, ship exhaust, and air control. A list of greenhouse gases and air pollutants in the port area has been completed, allowing for better identification and monitoring of pollution sources and providing appropriate energy-saving and carbon reduction measures.

‘Fuel oil consumption by ships in the port area’ refers to the amount of the fuel oil used by ships operating within the Anping Port area. This indicator is related to the type of ships, operational activities, and the port's economic capacity. Higher fuel oil consumption may indicate more active economic activities. To reduce the environmental impact of ship fuel consumption, the government and the port have implemented measures in the Taichung Port, such as to improve the fuel efficiency, optimize navigation routes, and promote clean energy, thereby enhancing the regulation of fuel oil usage. In 2021, fuel oil consumption by ships at the Taichung Port decreased by >20% compared to 2015. These measures provide valuable experience in mitigating the environmental impact of ship fuel consumption and are worth emulating by the Anping Port.

According to the analysis results (Table 5), the average scores of the status indicators represent their relevance to ecological safety and priority indicators. For example, when evaluating the status of environmental pollution, experts focused on indicators such as ‘total suspended particulates (TSP),’ ‘fine particulate matter (PM_2.5),’ ‘orthophosphate (M₃PO₄),’ ‘sulfur dioxide (SO₂),’ ‘nitrogen dioxide (NO₂),’ ‘nitric oxide (NO),’ and ‘nitrogen oxides (NO_x).’ These indicators reflect the impact of pressure on marine ecology. Experts suggest that indicators with adverse effects on marine ecology should be prioritized. Figure 2 shows the analysis results of weighted indicators for each aspect. Among the environmental pollution status indicators, the top seven are ‘water transparency’ (ranked 1), ‘nitric oxide (NO)’ (ranked 2), ‘sulfur dioxide (SO₂)’ (ranked 3), ‘fine particulate matter (PM₁₀)’ (ranked 4), ‘total suspended particulates (TSP)’ (ranked 5), ‘orthophosphate (M₃PO₄)’ (ranked 6), and ‘nitrogen dioxide (NO₂)’ (ranked 7).

Table 5

Weights and average scores of status indicators.

Code .	Index .	Average score .	Indicator weight .	Level weight .
C18	Perspective in water	4.29	0.0614*	0.149
C19	Seawater pH	4.14	0.0107
C20	Total suspended solids (TSS)	4.29	0.0156
C21	Dissolved oxygen (DO)	4.29	0.0146
C22		4.43*	0.0189*
C23	Zinc in sea water	4.29	0.0109
C24	Arsenic in seawater	4.29	0.0169
C25	Total suspended particulates (TSP)	4.57*	0.0195*	0.0195
C26	Suspension of fine particles	4.57*	0.012
C27	Suspension of fine particles	4.14	0.0219*
C28		4.43*	0.0293*
C29		4.43*	0.0181*
C30	(NO)	4.43*	0.0315*
C31		4.43*	0.011

Code .	Index .	Average score .	Indicator weight .	Level weight .
C18	Perspective in water	4.29	0.0614*	0.149
C19	Seawater pH	4.14	0.0107
C20	Total suspended solids (TSS)	4.29	0.0156
C21	Dissolved oxygen (DO)	4.29	0.0146
C22		4.43*	0.0189*
C23	Zinc in sea water	4.29	0.0109
C24	Arsenic in seawater	4.29	0.0169
C25	Total suspended particulates (TSP)	4.57*	0.0195*	0.0195
C26	Suspension of fine particles	4.57*	0.012
C27	Suspension of fine particles	4.14	0.0219*
C28		4.43*	0.0293*
C29		4.43*	0.0181*
C30	(NO)	4.43*	0.0315*
C31		4.43*	0.011

Source: Created by this research. * p < 0.05.

View Large

‘Water transparency’ is one of the key indicators for measuring the clarity and quality of water. In the Anping Port, water transparency is affected by various factors, including different pressures exerted on the port, and the transparency of the water in the Anping Port ranges from 0.5 to 2.8 m. Water transparency at the Anping Port has fluctuated between 2015 and 2021, primarily influenced by factors such as rainfall, typhoon impacts, and human activities. However, a comprehensive understanding of the transparency conditions at the Anping Port requires further investigation and ongoing monitoring.

‘Nitric oxide (NO),’ ‘sulfur dioxide (SO₂),’ ‘fine particulate matter (PM₁₀),’ ‘total suspended particulates (TSP),’ and ‘nitrogen dioxide (NO₂)’ are indicators of air quality. From the air quality-monitoring data retrieved from the Tainan Environmental Protection Bureau, air quality in the Anping Commercial Port, and surrounding areas has improved from 2015 to 2021. Specifically, the annual average concentration of suspended particulates (PM₁₀) decreased by 20%, the annual average concentration of NO₂ decreased by 15%, and the annual average concentration of carbon monoxide (CO) decreased by 10%. The 2022 air-monitoring data revealed that Tainan city's air quality rate reached 90.8%, a historic high. The air quality in the Anping Commercial Port and its surrounding areas has also markedly improved. Since 2015, the Tainan City Government has promoted vehicle emission controls in the Anping Port area and completed an environmental report in 2021. The report indicates that after implementing control measures, pollutants from vehicle emissions in the Anping Port area have significantly decreased. For example, in 2021, the concentration of PM_2.5 in the Anping Port area decreased by 20% compared to 2015, and the concentration of NO_x decreased by 15%.

The analysis results (see Table 6) indicate that the average scores of the impact indicators represent their relevance to ecological safety and priority indicators. For example, when assessing the impact of environmental pollution, experts focused on indicators such as ‘ the amount of waste removed from water areas.’ These indicators reflect the current impact of marine ecological conditions on the environment. The study suggests prioritizing indicators that have adverse effects on marine ecology. Figure 2 shows the analysis results of weighted indicators for each aspect. Among the environmental pollution impact indicators, the top two are ‘number of accidental incidents’ (ranked 1) and ‘amount of waste removed from water areas’ (ranked 2).

The ‘number of accidental incidents’ includes events such as ship collisions, fires, and explosions. For instance, the ‘Heng-En’ vessel sinking incident in 2015 resulted in a heavy oil spill that polluted the waters around the Anping Port, causing widespread water discoloration and fish deaths. The explosion at Dock No. 30 in 2017 led to the release of large amounts of toxic gases, severely polluting the surrounding air quality. Safety issues at the Anping Port cannot be ignored, and relevant departments should implement effective measures to prevent future accidents and ensure port safety and the protection of personnel and property.

The ‘amount of waste removed from water areas’ involves waste disposal by shipping companies, the cleanup of waste by professional companies, environmental protection units, and private or public entities either conducting cleanup themselves or outsourcing it, as well as waste disposal by shipping companies. The amount of waste removed from Anping Port's commercial harbor waters from 2015 to 2021 was handled to outsourced contractors. These cleanup services were covered by various annual tender projects. For example, a 2016 tender for cleanup services amounted to NT$947,560.00. These tenders indicate Anping Commercial Port's commitment to environmental protection and its effective management of water area waste through outsourcing.

The analysis results (see Table 7) indicate that the average scores of the response indicators represent their relevance to ecological safety and priority indicators. For example, when assessing responses to environmental pollution, experts focused on indicators such as ‘use of low-pollution fuel by ships,’ ‘number of reports,’ ‘proportion of enclosed storage,’ ‘carbon reduction from pipeline operations (automated gate monitoring),’ and ‘number of dust suppression nets.’ These indicators reflect the current impact on marine ecology and the measures taken to address environmental damage. The study suggests prioritizing response measures that are beneficial to marine ecology. Figure 2 shows the analysis results of weighted indicators for each aspect. Among the environmental pollution response indicators, the top five are ‘advice for improvement’ (ranked 1), ‘number of reports’ (ranked 2), ‘use of low-pollution fuel by ships’ (ranked 3), ‘number of anti-pollution facilities for loading and unloading’ (ranked 4), and ‘carbon reduction from pipeline operations (automated gate monitoring)’ (ranked 5).

‘Advice for improvement’ involves initiatives such as collaborating with the Tainan City Government to promote investment, enhancing the port infrastructure to improve competitiveness and addressing traffic issues, mitigating pollution from sand and gravel handling to protect the port environment, and considering the opening of port roads with attention to safety and management issues.

‘Number of reports’ is an important indicator of port safety and management, reflecting the number of incidents at the Anping Port, including ship accidents and pollution events. A high number of reports may indicate potential safety risks or management issues, requiring strengthened oversight and improved safety measures. By reducing the number of reported incidents, the Anping Port can enhance its image and reputation, attracting more ships and cargo traffic.

‘Use of low-pollution fuel by ships’ refers to the use of low-sulfur or alternative fuels that comply with International Maritime Organization regulations to reduce sulfur oxides and other pollutants emitted by ships. Currently, the Anping Port does not supply marine gas oil with sulfur content <0.5%, unlike the five major international commercial ports of Keelung, Taichung, Kaohsiung, Suao, and Hualien. However, ships docking at shore power facilities can reduce the use of polluting fuels, which is a measure to maintain the urban air quality. Additionally, ongoing research efforts include developing ship emission standards and port reduction regulations to enhance energy efficiency and carbon reduction benefits. These measures are expected to improve air pollution issues.

‘Number of anti-pollution facilities for loading and unloading’ refers to measures taken by the port to prevent air pollution. According to the Taiwan Port Authority data, such measures include dust suppression nets and mobile spray towers, which help reduce the impact of loading and unloading operations on air quality. Effective anti-pollution facilities can reduce the influx of pollutants into the surrounding marine area, helping to protect marine ecology.

‘Carbon reduction from pipeline operations (automated gate monitoring)’ aims to reduce truck traffic, fuel consumption, and carbon emissions. Data show that the use of automated gate monitoring systems at Taiwan's commercial ports has achieved a carbon reduction of 286 metric tons, with decreasing fuel consumption and significant energy savings. Additionally, co-operation between the Chunghwa Telecom and Taiwan Port Authority to enhance the automation and intelligence of the Anping Port is estimated to result in a carbon reduction equivalent to approximately 18.8 Da'an Forest Parks annually.

Firstly, for 2015 to 2021, Figure 3 shows the indices for Driving forces (D), Pressures (P), and Impacts (I) exhibited a consistent downward trend. As negative indicators, declining values imply intensified ecological stress, suggesting that port activities increasingly contributed to environmental degradation. This may reflect growing commercial demand at Anping Port. Conversely, the State (S) and Response (R) indices showed a fluctuating pattern – initial decline followed by recovery – indicating that, despite anthropogenic and policy-related disturbances, targeted interventions have partially mitigated ecological impacts and supported environmental improvement.

Secondly, this study used 46 indicators to construct the marine ecological safety indicator system based on the DPSIR framework. This study's ecological safety assessment system uses publicly available government statistics, making the results more objective and comprehensive compared to other studies. From 2015 to 2017, the marine ecological safety index of the Anping Port gradually fell into an unsafe state. However, from 2018 to 2021, the index gradually recovered to a relatively unsafe state, showing a ‘U-shape’ trend, with the highest value in 2015 and the lowest in 2017. This indicates that emergency improvements made by government agencies can effectively enhance the ecological safety of the port. From 2015 to 2021, the safety indices for driving forces, pressures, and impacts showed a fluctuating downward trend, decreasing to 0, 0.003, and 0.008, respectively. From 2015 to 2017, the safety index for states showed a downward trend, while from 2017 to 2020, it showed a fluctuating upward trend, rising to 0.088. From 2015 to 2016, the response layer safety index showed a sharp decline to 0.003. Using the entropy method to organize the overall marine ecological safety and determine the weights of each DPSIR model layer, it is found that the response layer has the most significant weight, followed by the driving forces and impacts layers.

Thirdly, the analysis results indicate that ‘advice for improvement,’ ‘number of reports,’ ‘water transparency,’ ‘use of low-pollution fuel by ships,’ ‘nitric oxide,’ and ‘number of anti-pollution facilities for loading and unloading’ are the main indicators affecting marine ecological safety. ‘Advice for improvement’ and ‘number of reports’ have a significant impact on marine ecological safety, highlighting the importance of timely and frequent warnings and recommendations for protecting marine ecological safety. The main purpose of this study is to analyze various factor layers of the Anping Port and appropriately assess the key indicators of each factor layer. Emergency handling is highly beneficial for policy formulation concerning theAnping Port in the Tainan City.

In recent years, many countries have actively promoted the development of marine ecological safety, focusing on protecting the marine environment and ecosystem and promoting the sustainable use of marine resources. Therefore, the Ocean Conservation Administration, a dedicated unit responsible for marine conservation in Taiwan, was established. The ecological environment has deteriorated rapidly with the rapid development of the society and the economy. Coastal areas, in particular, are facing significant social and economic development and environmental pressures, and there is an urgent need to establish a universally applicable ecological safety assessment method. Case studies provide decision-making references for solving ecological safety problems.

With the assistance of every marine ecological safety experts, three rounds of surveys were conducted every three months. Based on the DPSIR model, a marine ecological safety indicator system was constructed from the following five aspects: driving forces, pressures, states, impacts, and responses. The system comprises five criteria layers, 11 factor layers, and 46 indicators. Summarizing the opinions of experts, it is generally concluded that the pressure aspect significantly affects marine ecological safety. Two indicators with an average value of >4.5, ‘greenhouse gas emissions from Anping Port Operations’ and ‘fuel oil consumption by ships in the port area’ are of particular concern, as their values indicate potential negative impacts on marine ecological safety that require special attention and management.

From 2015 to 2021, monitoring results of marine ecological safety at the Anping Port show that the water quality is generally good, but some indicators exceed the standard levels. The DPSIR method followed in this study has also been verified. The investigation results are consistent with the data from the on-site investigation, proving the feasibility of this method. To address the marine ecological safety issues at the Anping Port, the local government has implemented a series of measures, including strengthening pollution control and protecting marine biological resources. These measures have achieved certain results but still require further improvement and strengthening.

In Taiwan, with the enactment of the Marine Conservation Act in July 2024, the future establishment of an ecological environment assessment mechanism and indicators by the Marine Conservation Administration should serve as a reference for the future delineation of marine-protected areas, marine-protected area management, and conservation measures. In the future, strengthening marine ecological safety research, improving marine ecological safety policies, and strengthening marine ecological safety monitoring are even more crucial factors for maintaining the integrity of the marine ecosystem.

The authors thank the anonymous reviewers from this journal for their valuable comments and all the individuals who assisted in this study. Grants from the Taiwan National Science and Technology Council financed this study (NSTC 112-2410-H-006-099 and NSTC 112-2119-M-004-001).

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.