The feedback model and stock-flow model were first conceptualised and visualised in Vensim [35], and all of the equations within the model and analyses were conducted in R using the packages “deSolve” [36] and “ggplot2” [37].

Mangrove sector

The mangrove forest is modelled as a renewable resource. The mangrove sector contains the stock Mangrove, which represents the mangrove area. The model captures the mangrove planting and loss rates to determine the change in mangrove area over time. This feedback loop allows us to better understand the effects of mangrove restoration and degradation on the overall system dynamics.

The net change in mangrove area over time is determined by subtracting the area of mangroves lost from the area of mangroves added. This equation reflects the balance between mangrove restoration and degradation within the system. If the net change is positive, it indicates a growth in the mangrove area, while a negative net change implies a decline in the mangrove area. Furthermore, we assume some abandoned shrimp ponds will eventually revert to mangrove habitats [38]. Landsat imagery from 1999 to 2022 showed a shift in land use from shrimp ponds to mangroves [10].

Policymakers and authorities often overlook the dependence of the shrimp industry on ecological services and fail to implement measures to limit its expansion and preserve natural capital. As a result, shrimp ponds may expand beyond ecological carrying capacity, especially within mangrove areas, with substantial environmental repercussions and potential system collapse [38].

Figure 1.

Proposed dynamic model for shrimp aquaculture expansion and mangrove conservation

The variable Effect of land availability on Mangrove represents the impact of land availability on the mangrove group. It estimates the extent to which land availability influences the growth or decline of mangrove areas over time. This variable captures the competition for land between mangroves and ponds within the system. As the pond area expands, it reduces the available land for mangroves, potentially constraining their growth. Conversely, if the pond area decreases, it frees up more land for mangrove expansion. Competition between anthropogenic activities and mangrove areas was studied not only in Vietnam but also in international coastal areas such as Indonesia [39], India [40] or Mexico [41].

Incorporating the Effect of land availability on Mangrove into the model accounts for the feedback between mangrove and pond dynamics. The availability of land for mangrove growth influences the net change in mangrove areas by adjusting the planting and loss rates based on the proportion of available land. This feedback mechanism allows researchers and policymakers to explore the interplay between mangrove and pond areas and understand how changes in land availability can impact the growth or decline of mangroves within the system.

Understanding the dynamics of the mangrove sector is critical since mangroves are essential to coastal ecosystems. They provide habitat for various animals, aid in carbon sequestration, help stabilise shorelines, and support local livelihoods. The model allows researchers and policymakers to examine the impacts of mangrove restoration or degradation on overall system dynamics and explore options for sustainable mangrove management by evaluating the feedback loops connected with the mangrove sector.

The model was estimated using historical data from 2007 to 2021 and predicted from 2022 to 2050 with a time step of 1 year. Two scenarios were analysed regarding the balance between mangrove conservation and shrimp pond expansion in Ca Mau (Table 2).

The first scenario, known as the “business-as-usual” (BAU) or baseline scenario, extrapolates the trends of all variables observed from 2007 to 2021 to predict future developments until 2050 without imposing any policy constraints on the expansion of shrimp aquaculture. Besides, the auxiliary variable “land capacity” constrained expanding shrimp ponds and mangrove areas. This growth limitation was 464,105 ha, representing the total area designated for agriculture, forestry and aquaculture in Ca Mau province [11].

The second scenario termed the “Policy scenario”, incorporates the policy objectives stated in the “Enhancing Efficiency and Sustainable Development of the Shrimp Industry in Ca Mau Province by 2025 and Vision towards 2030 Project” and the “Sustainable Agricultural and Rural Development Strategy in Ca Mau Province for the Period 2021-2030, with a Vision Towards 2050 Plan” into the model. Per the policy measures, the area designated for shrimp aquaculture in Ca Mau is capped at 280,000 ha. This limit is enforced by adjusting the net change in pond area flow within the model, ensuring that the total pond area does not exceed this limit once reached. Consequently, by stabilising the shrimp farming area at 280,000 ha, the model predicts a reduction in the conversion of mangrove areas to shrimp ponds and anticipates a rehabilitation of abandoned or inefficient ponds back to mangroves. Additionally, the objective of decreasing the percentage of the workforce in agriculture and fisheries acts as a benchmark for evaluating the projected employment generated by the shrimp industry. Projections for Ca Mau’s population and labour force growth are collected from the projection data of the General Statistics Office. All other variables are projected to follow the patterns established in the baseline scenario.

Data were collected from Ca Mau’s Statistical Yearbooks from 2007 to 2021 and the Rural, Agricultural and Fishery Census of 2006, 2011, and 2016. Spatial data from Clark Labs [10] were utilised for visualising and mapping changes in mangroves in Ca Mau. The land cover mapping process utilised Landsat 5 imagery from 1999 and Landsat 8 imagery from 2014 to 2022. All Landsat data were resampled to a spatial resolution of 15 m and, when necessary, substituted with PAN-merged bands. In addition to the Landsat data, supplementary data were incorporated into the analysis, including SRTM elevations, Tasseled Cap transformed images, and various convolutions of reflectance data. Classification techniques involved using a multilayer perceptron neural network for classifying pond aquaculture and open water, while Mahalanobis typicality was employed to classify the remaining land cover classes [10].

The model’s validation depends on its structural and behavioural validity [50]. In this study, the model was calibrated and validated using key variables from 2007 to 2021. The process began in 2007 when trial and error simulations were conducted by adjusting individual parameter values. The goal was to minimise the error between the simulation results and the actual values until an acceptable level was reached without any significant increase in the error. The historical test involved comparing the simulation values with the real values. If the error between the two fell within an acceptable range, it indicated that the model had high credibility and was suitable for subsequent analysis [48]. Additionally, behaviour tests were performed using extreme-condition testing based on instructions from Duggan [51] and the “Runit” package [52]. Table 3 presents each simulation value’s relative errors, confirming the model’s capability to predict and analyse future scenarios.

From 1999 to 2022, Ca Mau’s aquaculture landscape dramatically expanded (Figure 2), predominantly at the expense of rice fields and mangrove areas, supported by Directive No. 09/CT/TU in 2000, promoting the transformation of low-yield rice fields into shrimp farms [53]. This policy-driven change mirrored trends in coastal Bangladesh, with both regions seeing rice fields converted into shrimp farming [54]. Although mangrove areas initially increased until 2014, they subsequently declined, with the majority of the loss attributable to their conversion into aquaculture zones. Nonetheless, some former pond areas have reverted to mangrove forests [10]. Furthermore, silt deposition in Ca Mau’s northern and northwestern districts has encouraged fresh mangrove colonisation [55].

Figure 2.

Land covers in Ca Mau 1999-2022 [10]

In Ca Mau, there are currently five main shrimp farming models, including industrial shrimp farming, improved extensive shrimp farming, shrimp-rice farming, shrimp-mangrove farming, and a combination of extensive farming methods. With increasing cultivation areas and yields over the years, shrimp production in Ca Mau Province has increased from 35,000 t in 1997 to 158,887 t in 2016 and 205,290 t in 2021 [11].

In 2022, the combined extensive shrimp farming area was 96,264.68 ha, including 20,486 ha of giant river prawn with a yield of 240-250 kg/ha and a production of 5,050 t. The area of improved extensive shrimp farming was 176,276 ha, exceeding the anticipated aim by 2.49% and increasing by 8.53% over 2021. The area of intensive and super-intensive shrimp farming reached 6,317 ha; super-intensive farming accounted for 4,352 ha, reaching 117.6% of the planned target and increasing by 18.2% compared to 2021, with a yield ranging from 40 to 50 tons per hectare per crop, intensive farming accounted for 1,965 ha, with an average yield of 5 tons per hectare per year for tiger shrimp and 8 tons per hectare per year for whiteleg shrimp [20].

In the baseline period 2007-2021, the shrimp pond area gradually increased between 2007 and 2016 (Figure 3). The most significant growth in aquaculture was recorded between 2014 and 2016 when the area jumped from 268,600 ha to 278,642 ha. These figures indicate a period of aggressive expansion driven by promoting the large-scale transformation of low-yield rice cultivation into shrimp farming [53]. In this period, “aquaculture development” was documented as a favourable policy, and “economic opportunities” were the most relevant factor in the expansion of shrimp aquaculture [56]. From 2016 to 2021, the area of shrimp ponds experienced fluctuations, alternating between reductions and growth, indicating variability in the expansion of shrimp farming.

Figure 3.

Comparison of Shrimp pond area (a) and Mangrove area (b)

After 2021, distinct trends emerged between the two scenarios. Throughout the simulation period, the BAU Scenario showed that the shrimp area ranged from 262,177 ha in 2007 to a maximum of 317,038 ha in 2050. The BAU shrimp areas exhibited a growth pattern consistent with past trends, with fluctuations in shrimp pond expansion reflecting historical variability. A similar expansion trajectory was forecasted for Malaysian shrimp farms over the next three decades [32]. In contrast, the Policy Scenario showed that annual changes in shrimp pond area were consistently smaller and decreased over time, implying a controlled and deliberate expansion rate that diverged from the more erratic pattern in the BAU Scenario. This shift aligns with policy goals, aiming to cap the shrimp farming area at 280,000 ha from 2028 onwards, fostering sustainable development through regulated land-use changes.

A comparison of the two scenarios clearly illustrates that the policy-driven approach would alter the trajectory of the shrimp pond area from those projected in the BAU scenario. These diverging paths will likely have different environmental, social, and economic implications in the analyses.

Regarding the mangrove area, from 2007 to 2014, Ca Mau’s mangrove area continuously increased from 70,072 ha to 79,310 ha before decreasing to 72,524 ha in 2021 [10]. The consistent growth from 2007 to 2014 may be attributed to the efforts to meet the objectives outlined in the national biodiversity strategy to 2010 and the vision to 2020 [57]. Additionally, diseases and water pollution negatively impacted shrimp aquaculture during this period [39].

Most critically, the enormous expansion in shrimp aquaculture beginning in 2014 coincides with the intensive destruction of Ca Mau’s mangrove forests (Figure 3). It highlights a direct correlation between these two land uses. Due to the tremendous economic prospects of the shrimp industry, a portion of Ca Mau’s mangrove area was converted into shrimp ponds. Despite conservation efforts, the mangrove forests in Ca Mau have not recovered from the significant losses experienced after reaching their peak in 2014. Previous studies indicated that the major loss of the Ca Mau mangrove forest is mostly due to shrimp cultivation expansion [12], [58]. Similarly, on the southern coast of Tamil Nadu, India, meticulous planning and strategic actions are essential to prevent the degradation of mangrove habitats due to the growth of aquaculture operations [59].

From 2022 to 2050, the BAU Scenario’s projection data indicated a gradual increase in the mangrove area, starting at 73,015 ha and culminating in 76,484 ha by the mid-century. It represented a moderate growth of approximately 4.7%. Throughout these years, the increase was not uniform, displaying slight fluctuations, which suggest a degree of variability in mangrove expansion due to unregulated shrimp farming development. A slight reduction was observed in 2049, hinting at a potential levelling off in the growth of mangrove areas as the scenario approached mid-century.

In contrast, the Policy Scenario showed a more robust growth pattern, starting from 73,518 ha in 2022 and exhibited a more consistent and significant increase. By 2050, the mangrove area under this scenario will reach 88,901 ha, marking an increase of roughly 20.9% from the initial area. This trend indicates a proactive policy effect that promotes but likely incentivises the expansion of mangroves, resulting in a more pronounced increase.

The difference in projected shrimp pond and mangrove area in 2050 between the two scenarios is 37,037 ha and 12,418 ha, respectively. It implies that for a 37,037 ha decrease in shrimp ponds, there will be a corresponding increase of 12,418 ha in the mangrove area by 2050. Therefore, stabilising the shrimp farming area has indirect positive consequences for mangrove protection. The Policy Scenario’s trajectory suggests that the policy measures implemented effectively promote mangrove conservation and expansion, as compared to the BAU Scenario, where the growth in mangrove areas is more conservative and subject to fluctuations that could indicate less stringent conservation efforts. The Policy Scenario’s proactive stance on environmental conservation seems to create a more favourable condition for mangrove expansion, which could contribute to enhanced ecosystem services and coastal protection in the long term. In Thailand, one of the world’s leading shrimp producers, “feebate” policy action also balances mangrove regions [46].

However, given the current pressures on mangroves resulting from the rapid expansion of shrimp aquaculture in Ca Mau, it is crucial to implement better-adjusted land use management plans. It involves balancing the needs of shrimp farming with the preservation and restoration of mangrove forests to ensure their long-term sustainability. There have been efforts made to reduce deforestation, such as the introduction of Decision No. 19/2010/QĐ-UBND in 2010 [60], authorising recipients of mangrove forest land to use up to 40% of it for combined agriculture and aquaculture, preserving the ecosystem. Following this, the International Union for Conservation of Nature and the Netherlands Development Organization launched the Mangroves and Markets (MAM) project in 2012, which instructed 5,500 Ca Mau residents in sustainable shrimp farming, safeguarding 12,600 ha of mangroves [61]. Continuing these efforts, since 2021, the “Mangrove and climate protection with income generation for vulnerable communities (VM069)” project has been operational, spearheaded by the SRD in Mui Ca Mau National Park and Tam Giang Protection Management Board, aiming to enhance biodiversity and develop mangroves to bolster carbon sequestration [62]. Unfortunately, the conversion of mangroves into shrimp ponds still occurred. Therefore, there is a demand for more effective measures to be implemented to prevent further deforestation.

The potential carbon stock per hectare was represented by a constant auxiliary [42], and the stock variable Carbon sink’s shape followed the Mangrove area’s shape in both the BAU and Policy scenarios (Figure 4).

Figure 4.

Carbon sink from Ca Mau mangrove forest 2007-2050

Under the BAU Scenario, the carbon sink values varied from 4.59×10⁶ MgC in 2022 to 4.83×10⁶ MgC in 2050. This variation occurred with an initial increase until 2023, when the carbon sink reached 4.608×10⁶ MgC, followed by a fluctuating decline over the subsequent years. The decline steadied towards 2030 and then showed upsurges till 2050.

Contrastingly, in the Policy Scenario, the carbon sink capacity showed an upward trend, starting from 4.625×10⁶ MgC in 2022 and increasing each year to reach 5.686×10⁶ MgC by 2050. This consistent rise reflects the potential effectiveness of policy interventions to enhance carbon sequestration. The increase was steady and did not exhibit the volatility in the BAU Scenario, indicating a robust and sustained enhancement of the carbon sink capacity over time.

These projections suggest that under the Policy Scenario, the carbon sink in the area are likely to improve significantly, assuming that the policies continue to support activities that enhance carbon sequestration. The increasing trend in the Policy Scenario underscores the potential benefits of targeted environmental policies. In contrast, the BAU Scenario indicates that without targeted interventions, carbon sink capacities may not maintain their initial levels and could potentially decrease, highlighting the importance of policy measures in sustaining environmental health and combating climate change.

Our analysis examined the potential contribution of carbon sink within the context of Vietnam’s National Climate Change Strategy and the commitment to a net zero emission target by 2050. For the year 2050, the BAU Scenario showed a carbon sink capacity of 4.83×10⁶ MgC. When converted to carbon dioxide equivalent (CO2e), this capacity contributes 17.7261×10⁶ MgCO2e. This figure equals 9.6% of Vietnam’s total emission cap of 185 MtCO2e as stipulated by the government [63]. On the other hand, the Policy Scenario indicated a more substantial carbon sink capacity of 5.686×10⁶ MgC in 2050. This capacity contributes to 20.867×10⁶ MgCO2e, mitigating 11.2% of the national emission cap.

These projections suggest that the Policy Scenario’s enhanced carbon sequestration efforts could yield a more significant impact in meeting the country’s emission reduction targets. The increase from 9.6% to 11.2% of the total allowed emissions underlines the critical role of policy-driven environmental management. It implies that adopting and implementing the proposed policies can amplify the effectiveness of natural carbon sinks, a crucial factor in the broader strategy to mitigate climate change. The discussion underscores the importance of strategic policy interventions in achieving Vietnam’s climate goals and contributing to global emission reduction efforts.

It is worth noting that the estimation of carbon stocks serves as a rough indicator for the potential of climate change mitigation via carbon storage of Ca Mau mangrove forests because determining carbon sequestration or carbon storage is extremely complicated and beyond the scope of this study. Mangrove forests are the most important carbon sink in the tropics [64]. The quantification of carbon storage in mangrove forests is essential for developing climate change mitigation strategies and implementing REDD+ schemes in Vietnam [65]. It also contributes to the larger puzzle of addressing climate change and planning for a sustainable future. By assessing the carbon storage capacity of mangrove ecosystems, we can better incorporate their conservation and restoration into broader climate action efforts [66]. In terms of monetary value, the carbon sequestration value of the Ca Mau mangrove is approximately 136.64 USD per ha per year [67]. In addition to carbon sequestration, effective management of coastal ecosystems can generate beneficial ecosystem services, including protection against tropical cyclones [68].

During the analysis period, Ca Mau agriculture concentrated on implementing the Agricultural Restructuring Project, particularly from 2016 to 2020. From 2007 to 2021, the total output value of the agricultural, forestry, and fisheries sectors rose from 493,710,000 USD to 940,128,000 USD, contributing nearly 34% to the province’s total output value. Labourers in the agricultural, forestry, and fisheries sectors earned 1680 USD per person-year, contributing to economic prosperity, political stability, and social stability [11], [69], [70]. Ca Mau is the largest producer of black tiger shrimp and operates farms with the highest cost efficiency but low environmental efficiency [71].

The shrimp farming industry is identified as a key economic sector and plays an important role in the economic development of Ca Mau province. The shrimp industry in Ca Mau accounts for 80% of the total production value in the aquaculture sector and 49% of the total production value in the agricultural sector. The output value of the shrimp farming area is approximately 45 million VND/ha (1890 USD/ha). The province’s seafood export turnover, mainly driven by shrimp, reaches 1.2 billion USD annually, accounting for more than 40% of the country’s shrimp export turnover, with the processing volume exceeding 138,000 tons annually [11], [69], [70].

Under the BAU Scenario, the shrimp output value experienced a consistent upward trend, beginning at 12,733 billion VND in 2022 and projected to reach 25,153 billion VND by 2050 (Figure 5). The growth was continuous, reflecting an overall positive trajectory for the shrimp industry in the absence of additional policy interventions. A steady income increase for farmers was also projected in Malaysia due to the expansion of the aquaculture area [32].

Figure 5.

Shrimp production output value in Ca Mau 2007-2050

Comparatively, the Policy Scenario began with a shrimp output value slightly lower than the BAU at 12,727 billion VND in 2022. However, this scenario projected an ascent to 22,214 billion VND by 2050, indicating a substantial increase, although at a lower ending than the BAU Scenario. It suggests that while policy interventions may promote more sustainable or regulated growth, they do not necessarily inhibit the economic output of the shrimp industry; instead, they seem to guide it towards a sustainable path.

From the prediction, both scenarios indicated growth in the shrimp sector’s output value. The BAU Scenario assumes an unrestrained industry expansion, whereas the Policy Scenario incorporates sustainable growth practices that may slightly temper economic output but potentially offer long-term benefits. These findings underscore the need to balance economic development with sustainable practices in the shrimp industry, aligning with Vietnam’s broader economic and environmental goals. Moreover, in developing countries, it is important to consider the scale of farming operations and the livelihoods of individuals. Shrimp farming, being capital-intensive, allows larger entrepreneurs to easily engage in high-profit activities. At the same time, small-scale producers would require institutional, technical, and financial assistance to comply with enhanced regulations to promote mangrove conservation [56].

Various strategies should be implemented to further bolster the shrimp industry’s economic growth in the Policy Scenario. These strategies include improving productivity, enhancing production quality, adopting advanced high-tech farming models, and strengthening production linkages. Ca Mau may secure its shrimp industry’s long-term success and development by carefully considering these factors and implementing appropriate measures. One emerging solution in Vietnam and Indonesia is adopting mangrove silvo-aquaculture farming systems (particularly integrated mangrove-shrimp farming). The adoption of these systems holds promise in enhancing resilience in coastal aquaculture landscapes by offering opportunities for livelihoods, restoring mangroves, sequestering blue carbon, reducing disease risks, improving water quality, generating additional income through diversified aquatic products and timber production, and providing coastal protection [72].

Furthermore, while shrimp aquaculture significantly bolsters Ca Mau’s economy, diversifying its economic activities remains essential. Encouraging the growth of other sectors, such as agriculture, tourism, manufacturing, and services, can reduce reliance on shrimp farming and mitigate risks associated with capped aquaculture growth. Relying solely on shrimp farming is unsustainable due to its high investment costs and risks, particularly for small rural households. Additionally, promoting diverse economic sectors will aid in redistributing the labour force, attracting more employees to alternative employment opportunities. By fostering various economic sectors, Ca Mau can create new income sources, jobs, and resilience, promoting economic development and stability. It aligns with the plan for restructuring agriculture and rural development in Ca Mau towards 2050 [19].

Under the BAU Scenario, shrimp industry employment is projected to increase steadily from 147,605 people in 2022 to 171,666 in 2050 (Figure 6). Meanwhile, the share of shrimp employees within the total labour force shows a gradual rise from 22.06% to 25.14% over the same period. This trend suggests that, without policy intervention, the shrimp industry’s share of the total labour force will continue to grow, potentially exceeding the government’s target of reducing the combined share of agricultural, fishery, and forestry employees to 40% by 2025 and 20% by 2030.

In contrast, the Policy Scenario reveals a different trajectory. The number of shrimp industry employees starts at a comparable level to the BAU Scenario in 2022. However, it reaches an early plateau at 151,611 people from 2028 onwards due to the stabilised shrimp farming area, maintaining this level through to 2050 (Figure 6). The share of employment generation in the shrimp industry begins at 19.42% in 2022 and shows a subtle increase to 23.58% by 2050. This trend indicates that the Policy Scenario is more closely aligned with the Ca Mau government’s labour force targets. These projections illustrate that the Policy Scenario is in closer adherence to the government’s aim of restructuring the labour force.

Figure 6.

Labour force (a), shrimp employees (b) and shrimp industry’s contribution to employment generation in Ca Mau 2007-2050 (c)

While the BAU Scenario implies growth beyond the set targets, potentially challenging the government’s strategic labour force composition, the Policy Scenario demonstrates a controlled approach, stabilising the number of employees in the shrimp industry. This stabilisation, along with the incremental increase in the industry’s labour share, supports the targeted reduction of the agriculture, fishery, and forestry workforce, thus contributing to the overall objective of diversifying Vietnam’s labour force and economic activities.

Shrimp farming, as the predominant industry in the rural regions of Ca Mau, significantly influences local livelihoods, accounting for 60%-70% of annual rural employment [73]. Within the BAU scenario, the projected expansion of shrimp ponds up to 2050 is expected to create additional employment opportunities, reinforcing this sector’s role in job creation. However, it is important to recognise that intensive shrimp culture can lead to socioeconomic challenges, as the industry is susceptible to high risks, including potential economic losses due to diseases and environmental degradation, such as water pollution [74].

Contrastingly, the Policy Scenario contributes significantly to rural employment, although the number of shrimp farming jobs is anticipated to stabilise. Despite this, its employment impact is projected to be less than that of the BAU scenario. A cap on shrimp pond expansion could shift the dynamics of the rural labour market, possibly resulting in increased migration of workers from rural areas to urban centres in search of alternative employment. Such a transition underscores the need for policymakers and stakeholders to develop strategies that balance the shrimp industry’s role in rural job provision with the pursuit of sustainable employment options.

Moreover, conservation efforts to protect regions rich in biodiversity must consider the economic needs of local communities [75]. The predicted difference in the number of shrimp employees between the BAU and Policy Scenario in 2050 is 20 thousand people, indicating that there would be a potential loss of employment opportunities in rural areas resulting from capped shrimp aquaculture. Implementing policies in Ca Mau should be phased and include promoting alternative livelihood options to mitigate any adverse effects on local employment [56]. This approach helps ensure that environmental conservation efforts do not undermine the economic stability of the local communities. Promoting alternative livelihoods includes supporting the development of other agricultural sectors, promoting rural entrepreneurship, and facilitating skill development programs to ensure a smooth transition for affected workers. The government can play a pivotal role by building on existing initiatives by allowing rural households with access to mangrove forests to engage in various agricultural and fishery activities, while also being eligible for capital support from programs like the Payments for Forest Ecosystem Services (PFES) Program [76]. Additionally, exploring alternative economic activities such as beekeeping and ecotourism can reduce households’ reliance on shrimp farming for income. Furthermore, implementing a payment for carbon services mechanism in Ca Mau’s mangrove forests could promote forest conservation and enhance local communities’ livelihoods [77].