Modeling the Impacts of Coastal Land Use Scenarios on Ecosystem Services Restoration in Southwest Ghana, West Africa

doi:10.21203/rs.3.rs-4432789/v1

Urbanization is a key driver of coastal habitats degradation in West Africa. Habitat restoration is strongly advocated to mitigate urbanization impacts in West African coastal areas. However, knowledge on the application of scenarios to envision land use patterns and ecosystem services (ES) supply in this region is still lacking in scientific literature. In this study, we applied land use scenario modelling to provide recommendations for habitat restoration planning and associated ES supply in coastal socio-ecological systems. Specifically, four land use scenarios (Urbanization Scenario (UBS), Urban Greening Scenario (UGS), Plantation Agriculture Scenario (PLAS) and Landscape Restoration (LRS)) were developed for the coastal zone of Southwest Ghana. Their impacts on land use patterns and ES (food, fuelwood, carbon sequestration and recreation benefit) were assessed and visualized by integrating benefits transfer and experts’ knowledge into a spatially explicit modelling platform. The simulated results showed that UBS would decrease the supply of food, fuelwood, carbon sequestration and recreation benefits in the region. LRS would create negative synergies between food and carbon sequestration but this relationship reversed to positive synergies with future intensification of restoration. Our findings also showed that LRS could lead to expansion of mixed swamp forests, no change in the spatial extent of palm swamp forests and decline of mangrove swamps. On this basis, we recommend planning regulations which target swamp forests in the region for enhanced protection and restoration in order to safeguard these critical coastal habitats and avert their future degradation due to urbanization.

Earth and environmental sciences/Ecology

Scientific community and society/Geography

Social science/Geography

Coastal ecosystems are integral components of landscapes and seascapes and provide the requisite biophysical and ecological conditions in which marine and terrestrial species thrive ^1,2. Across the land-sea continuum, multiple coastal ecosystems and their associated habitats interact to co-create ecological functions and enable interconnectivity between terrestrial, nearshore and marine ecosystems ^3,4,5. Ocean currents, nutrient transport and feeding patterns of marine organisms across different habitats drive these habitat interactions⁶. Moreover, environmental gradients such as salinity, temperature and elevation in coastal environments have effects on the spatial distribution of interconnected coastal and marine habitats. For instance, mangrove forests are limited in their range by coastal geomorphological features and estuarine tidal dynamics ^7,8.Similarly, variations in coastal microenvironments create the ecological conditions which enable some fish species to live their juvenile stages in nearshore areas and adult life in marine environments^2,9.These interconnections underpin ecological functions and the supply of ES in coastal regions¹⁰. Maintaining the spatial connectivity and heterogeneity of coastal habitats and their functions is fundamental to conserving biodiversity⁵. However, coastal habitats are predisposed to urbanization threats as human population and activities are concentrated on the narrow zone between land and ocean interfaces along coastal areas¹¹. Coastal urbanization is characterized by land use changes which convert natural areas into built infrastructure such as settlements, roads, wharfs and breakwaters¹². Such land use pressures distorts the spatial connectivity between coastal habitats and leads to the loss of ecosystem services (ES) ¹².

At the international, national and local levels, ecosystem restoration has been proposed as a policy and pragmatic response to the degradation of natural resources ^13,14. The UN Decade on Ecosystem Restoration acknowledges the centrality of actor-oriented land use planning in facilitating achievement of the restoration goals and targets linked to the 2030 Agenda for Sustainable Development¹⁵. Land use planning focused on restoration is gradually developing towards the application of frameworks and tools for the participation of a broad range of stakeholders in restoration decision making, including determination of socio-economic objectives and benefits of restoration ¹⁶. Goal-setting in landscape restoration becomes effective with the engagement of various land use actors as it allows to identify different interests, values and preferences for sustainable landscape planning ^17,18. Hence, a socio-ecological perspective integrating both biophysical and social components and their spatial and temporal interactions is a prerequisite for effective coastal landscape planning and restoration ¹⁹.Understanding the role of human agency, as a beneficiary of natural resources, in coastal habitat restoration planning paves the way for integrating ecosystem services (ES) supply into restoration goals ²⁰. Incorporating ES into restoration decisions increases the chances for successful outcomes by addressing human values and wellbeing dimensions ^21–23. Considerations of human values also reduces the risk of political and community backlash against restoration actions while promoting public support for restoration goals ^21,24.

Participatory scenario development and assessment have increasingly proven to be valuable heuristic tools for engaging ecosystem restoration actors in goal-setting processes and the collective development of plausible ES provision futures in West Africa ^25–27. Incorporating local knowledge in scenario composition revealed that future urbanization and deforestation land use patterns will reduce ES provision in West African agricultural landscapes^28,29. Furthermore, in Ghana’s forest-plantation mosaic landscapes, preferences for segregated rather than integrated land use scenarios potentially caused trade-offs between multiple ES supply ²⁶. Effective scenarios can be those appropriate for performing three interrelated functions of modelling, planning and making effective future-oriented decisions ³⁰. Different types of scenarios have been advanced in literature to create narratives and to characterize diverse coastal and marine socio-ecological contexts ^31–33. Exploratory scenarios are increasingly utilized by scientists to describe plausible future states of ES supply under different driving forces and environmental conditions ^34–36. Whereas exploratory scenarios investigate a range of plausible futures considering key drivers of change, normative scenarios on the other hand, define the vision of a preferred future and identify the goals and strategies required to achieve that vision ^33,35,37.

Coastal areas in West Africa are characterized by heavy reliance of the population on natural resources for their livelihoods, hence, the conditions of the region’s coastal habitats are interlinked with ES supply and human wellbeing ³⁸. Driven by the accelerating pace of urbanization in coastal West Africa, rapid land use changes poses threats to the sustainability of coastal ecosystems in the region ^39,40. Restoration of the region’s critical biodiversity habitats such as mangroves, peatlands, lagoons, estuaries, rocky reefs and benthic ecosystems can mitigate marine biodiversity loss and recover such ecosystems and the functions and services they provide. Particularly, in the coastal landscapes of Southwest Ghana, different restoration initiatives are underway and aim to advance the protection of pristine coastal habitats and restore degraded ecosystems to enhance ES supply ^27,41. However, research on land use scenario planning in West Africa has focused attention on ES provision potentials due to land use changes in agricultural and production landscapes ^26,28,42. There is less attempt of scenario applications in landscape restoration and for evaluating trade-offs in restoration outcomes⁴³. Furthermore, few studies have applied participatory scenarios in the coastal and marine contexts to understand spatial changes in habitats and ES supply dynamics ⁴⁴. This study addresses this knowledge gap by utilizing remote sensing, benefits transfer and participatory land use scenario modelling to deepen understanding of urbanization and coastal landscape restoration outcomes on ES recovery in Southwest Ghana.

The spatial dynamics of habits and land use patterns according to urbanization and landscape restoration outcomes can be analyzed using remote sensing data ^45,46. Benefits transfer is a proxy-based technique for ES assessment, utilized especially in data scarce contexts for extrapolating biophysical values from one site to another ^47,−,49. Application of the foregoing analytical techniques contributes to the understanding of coastal landscapes governance dynamics in the region⁵⁰. Furthermore, understanding and generation of such information and knowledge is necessary to inform multi-habitat restoration and strategic planning to mitigate risks posed by urbanization to the functionality of coastal habitats. Especially, application of the data and information generated in a spatially explicit scenario modelling approach is useful to identify risks and opportunities associated with the articulation of land use actors’ visions and values for landscape restoration. We anticipate that scenarios in favor of landscape restoration will lead to an increase in multiple ES supply and the spatial extent of habitats in coastal areas undergoing urbanization. This study specifically aims to address the following research questions in the context of Southwest Ghana; i) how will local experts’ perspectives on land use transitions influence future land use decisions in coastal landscapes undergoing urbanization? ii) what are the impacts of future land use scenarios on the spatial extent of critical coastal habitats and supply of ES in Southwest Ghana? iii) what potential trade-offs and synergies could result between multiple ES due to future land use scenarios in Southwest Ghana?

2.1 Description of study area

The study was conducted in the coastal zone of Southwest Ghana, specifically, the Greater Amanzule wetlands landscape. Covering approximately 60,000 hectares, the area lies between the Ankobra river estuary and the Tano river basin on Ghana’s Southwest border with Cote d’Ivoire⁵¹ (Fig. 1). The rainfall pattern is characterized by a peak season occurring between May to June and a minor season, spanning October to November of each year. Mean annual precipitation and relative humidity are 1600 mm and 87.5% respectively ⁵². The study area is known for its rich and diverse but critical marine and coastal habitats and biodiversity. These critical habitats comprise lagoon-wetland systems, mangroves, estuaries, sandy and rocky beaches, located in the only coastal area in Ghana which hosts intact peat swamp forests ^1,51. These habitats are vulnerable to the mounting threats of population growth, urbanization, industrialization and climate change¹.The study area spans three administrative boundaries in Southwest Ghana – Nzema East, Ellembelle and Jomoro districts.

Socio-economic activities in the study region revolve around fishing and farming. The fishing industry has witnessed a steep decline over the past decades, due primarily to overfishing⁵³. Recent land demand in the region is significantly shaped by urbanization, which is a consequence of land use/land cover (LULC) changes linked to the region’s growing oil and gas industry, large scale plantation agriculture development and population growth^54–56. Traditional arable land uses are being displaced in favor of mining and in some instances, illegal mining operations ^57,58. Food crop (including cassava, plantain and cocoyam) farming is undertaken at the subsistence level. Other land use options such as rubber and oil palm plantation development are commercial ventures dominated by land owners and agro-based private sectors ⁵⁵. There is high demand for mangrove wood as it is culturally preferred to other wood sources for fish smoking in traditional ovens.

2.2Methodological framework

The study methodology was implemented in three phases, resulting in key research outputs and outcomes (Fig. 2). The first phase involved land use/ land cover (LULC) classification and delineation of critical coastal habitats and interlinked terrestrial LULC types (section 2.3.1). In the second phase, workshops and questionnaire survey were implemented with selected land management experts (section 2.3.3). The workshop sessions were framed around land use situation revealed in the LULC maps, information contained in the relevant municipal and regional spatial and municipal plans and identification of relevant ES in the study region (section 2.3.4). Through the workshop and questionnaire survey, locally relevant ES were identified, land use transition rule-sets for simulating plausible future scenarios were co-created with land management experts. The second phase also involved the selection of indicators and extrapolation of biophysical values from existing studies for compilation into an ES assessment matrix (section 2.3.2). In the final phase, the land use scenarios were simulated and their impacts on habitats spatial extent and ES supply evaluated using the GISCAME simulation platform (section 2.3.5).

2.3 Data gathering and processing

2.3.1 Remote sensing

The land cover classification was conducted using the Google Earth Engine (GEE) platform. GEE provides access to free, georeferenced, and atmospherically corrected real-time remote sensing data, comprising a comprehensive catalog spanning over four decade⁵⁹. The classification utilized a blend of optical and radar satellite sensors to address the persistent challenge of cloud cover interference commonly encountered with optical sensors. Sentinel-1 Synthetic Aperture Radar (SAR), Ground Range Detected (GRD) product, with dual polarization and Landsat 8 OLI (Operational Land Imager) Collection-1 Tier-1 Top-of-Atmosphere Reflectance product, acquired in 2010, 2015 and 2020 served as the primary datasets. As a further preprocessing step of the SAR dual-polarized data, a speckle filter was applied to denoise the image. Speckle filtering enhances visual quality and improves qualitative parameters like Peak Signal-to-Noise Ratio (PSNR)⁶⁰. Subsequently, Gray Level Co-occurrence Matrix (GLCM) parameters were computed to capture textural information available in the data. Subsequently, various vegetation indices from Landsat 8 were derived to characterize vegetation properties and later integrated with the SAR bands. In addition, ancillary data, including slope, aspect, hillshade and elevation was derived from the Shuttle Radar Topography Mission (SRTM) data. The three categories of data were merged into a composite image. Training samples were collected over the composite image, representing each land cover class. These samples were grouped into one feature class, and a Random Forest classifier with 100 trees and 2 variables per split was trained using 70% of the samples as the training dataset. The classifier was then applied to the entire composite image, and the resulting classified image was evaluated using a 30% validation set to compute accuracy metrics. The area covered by each land cover class was calculated within the region of interest. As a result, twelve LULC types were included in the classification (Fig. 3 and Table. 1).

Table 1

Major LULC types and their areal coverage
LULC	Area (Ha)	%
Mangrove swamp	949.17	1.55
Mixed swamp forest	11,635.60	19.02
Palm swamp forest	3,282.67	5.37
Bog plain	1,594.12	2.61
Shrub land	17,956.20	29.35
Rubber	3,995.18	6.53
Coconut	3317.83	5.42
Oil palm	229.01	0.37
Bare surface	60.90	0.10
Built-up	2,300.82	3.76
Water	290.78	0.48
Food crop	15,563.60	25.44

2.3.2 Selection of ES and indicators

Coastal and interconnected marine ecosystems supply an array of ES for the wellbeing of the dependent coastal population in Ghana⁶¹. Based on the results of prior studies on ES supply in the study region, locally relevant ES were identified and selected for impact assessment of land use scenarios on ES restoration ^27,41. Cultural ES are generally created through human perception of ecosystems and the biophysical environment⁶². In this study, we utilized land management experts’ judgements to obtain estimates of recreation benefits based on a ranking criterion (see section 2.3.4). Recreation benefits were selected as the coastal landscape and adjoining seascape have unique tourist attractions including scenic beaches (bare surfaces), water bodies, pristine mangrove forests and green spaces (shrublands) that offer recreation opportunities²⁷. Among potentially relevant provisioning and regulating ES in the study context, a specific set of ES were determined according to data availability. As provisioning ES, we selected food and fuel wood supply. In addition to staple food supply by cropland, oil palm and coconut plantations are sources of vegetable oils in the region while mangroves, swamp forests and waterbodies are critical to meet the local population’s fish food needs. Mangroves, shrublands and rubber plantations are also important sources of fuelwood supply to the local population. For regulating ES, carbon sequestration was determined. This ES is supplied by all vegetation types in the study region, albeit at different magnitudes.

Considering data availability, indicators that quantitatively depict ES supply were derived in relation with the characteristics of LULC (Table 2 and Table S2). Specifically, food and fuelwood supply were respectively measured in Tonnes/ha and Kg/ha while carbon sequestration was quantified in MgC/ha ^41,63. Finally, all the ES indicator values derived through benefits transfer, remote sensing and experts’ consultation were standardized to a relative scale (0–100 value points) ⁶⁴. Standardization of the different indicator values was necessary to allow subsequent comparison of ES supply potentials across the different LULC classes with the same unit value ⁶⁴.

Table 2

Data generation methods for locally relevant ES and their proxy indicators
Relevant Ecosystem Services	Proxy indicators	Data Generation
Provisioning
Food	Yield in Tonnes/ha	Benefits transfer, remote sensing
Fuel wood	Biomass in Kg/ha	Benefits transfer, remote sensing
Regulating
Carbon sequestration	Above-ground carbon in MgC/ha	Benefits transfer, remote sensing
Cultural
Recreation benefits	Benefits derived from recreational activities in open and green spaces	Questionnaire survey, workshop

2.3.3 Selection of land management experts

Experts in this study were defined as knowledge holders about land use system dynamics of the study region by virtue of their training, research, skills and practical experience ⁶⁵. We selected land management experts across academic institutions, government agencies, private sector, non-governmental organizations and heads of land-owning clans in the study region. Through this representation, we elicited knowledge held at the local scale by land owners with non-professional interests as well as insights from professional knowledge holders at the regional scale. ⁶⁶. The experts were identified through snowball sampling, in which those selected experts identified other potential experts ⁶⁷. This sampling technique was suitable for this study as it allowed identification and recruitment of experts with more than a decade long experience in land system research and practice in the study region. In total, twenty-four land management experts comprising 4 coastal ecologists, 10 land use and development planners, 4 environmentalists, 2 rubber and oil palm plantation managers and 4 land owners were identified and recruited for the scenario co-creation workshops and questionnaire survey,

2.3.4 Future scenarios development workshop and questionnaire survey

A workshop with land management experts was utilized to facilitate development of land use scenarios. As preparation for the workshop, relevant regional land use planning and policy documents were screened. Literature on the drivers of land use changes and their influences on coastal ecosystems within the study region were also reviewed to enable contextual understanding of the LULC classification. This literature served as the basis for spatially explicit simulation of four land use scenarios, namely, Urbanization Scenario (UBS), Plantation Agriculture Scenario (PLAS), Landscape Restoration Scenario (LRS) and Urban Greening Scenario (UGS) using the GISCAME software. It also informed the development of scenario narratives (Table 3), which framed and contextualized the workshop discourses together with the baseline simulation results.

Table 3

Narratives of coastal land use scenarios for Southwest Ghana. UBS: Urbanization Scenario; PLAS: Plantation Agriculture Scenario; LRS: Landscape Restoration Scenario; UGS: Urban Greening Scenario
Scenario	Description
Urbanization Scenario (UBS)	The urbanization scenario is characterized by unfettered settlement expansion and encroachment of built-up infrastructure on natural areas⁶⁸. Drivers of these land use changes are population growth and implementation of transportation infrastructure development plans, notably for road network expansion and establishing railway line connections between Ghana and neighboring Cote D’Ivoire along the coastal corridor⁶⁹. Built-up infrastructure expansion is also related to coastal tourism development to enhance recreation experience in tourist destinations along the coast²⁷.
Plantation Agriculture Scenario (PLAS)	The plantation agriculture scenario represents future land use trends dominated by rubber and oil palm plantations ⁷⁰.Forests, shrublands and food crop areas are substituted for plantations as landowners and farmers realize higher returns from rubber and oil palm cultivation ^71,72. Plantation agriculture also expands in response to the increasing global demand for agricultural commodities such as natural rubber latex and palm oil^73,74.
Landscape Restoration Scenario (LRS)	The landscape restoration scenario prioritizes strict protection of swamp forests using planning and zoning regulations, community-based conservation norms and active restoration of degraded mangrove ecosystems and shrublands to create habitats for biodiversity⁷⁵. Abandoned illegal gold mining sites are reforested⁷⁶.
Urban Greening Scenario (UGS)	The UGS targets future coastal land uses characterized by developed urban core zones and sprawl along the urban peripheries ^56,68. Within this urban development context, green spaces are introduced into the built-up areas such as along road medians, around compounds of homes, schools and within recreational parks, in line with the government’s reforestation program ⁷⁷.

At the commencement of the workshop, the LULC information in the region was presented alongside spatially explicit representations of the land use scenarios. As ES is an unfamiliar concept to the experts, an overview of the potential ES supplied by the LULC types was also presented ²⁷. Subsequently, the experts were randomly assigned to one of three groups (comprising 8 experts per group) and asked to discuss the following questions: 1) which land use transitions, for instance, conversion from mangrove to built-up areas, neighborhood and environmental conditions align with urbanization, plantation agriculture, landscape restoration and urban greening scenarios in the region? 2) what are the timescales for the identified land use transitions to express in the region? Outputs of the workshop were plausible land use transition rule-sets which align with the land use scenarios (Table S1).

A post-workshop questionnaire survey was implemented with the same group of experts. The questionnaires were composed of two parts. The first part focused on the potential supply of cultural ES by the respective LULC types. Related to this aspect, the experts were firstly, invited to rank from a pre-identified list, the top three cultural ES supplied by the landscape. Secondly, an evaluation was performed on the prioritized cultural ES according to the following criteria: a) on a scale of 1(very low) to 5 (very high), and b) the relative potential of the twelve LULC types to supply cultural ES. The second part of the questionnaire elicited information on the transition probabilities of the LULC transitions identified at the workshop stage. Concerning this aspect, the experts were invited to rank the probability of occurrence of each LULC transition on a scale between 0 to 100%, given certain neighborhood effects and environmental conditions ^28,64.

2.3.5 Spatially explicit simulation and impact assessment

In order to simulate future land use patterns, transition rule-sets determined by the experts for each of the scenarios were iteratively applied on the current land use map using the cellular automata (CA) module in GISCAME. The CA comprises an array of cells which can change their state simultaneously at any given time as a function of their own state and the states of cells in their neighborhood ²⁸. An iteration in a CA model is dimensionless, however, a timescale can be assigned to an iteration in GISCAME based on the expected period of land use changes ²⁸. Following the questionnaire survey, realistic land use changes were determined by the experts to occur within 10-year cycles. CA simultaneously reassigns land use types to all raster cells in a land use map according to the defined transition rule-sets, thereby creating new land use patterns affected by future scenarios ^28,79,80. Using the 10-year period as the basis for regional land use transitions, one application of the transition probabilities simulated a change in 10 years, thus a five-time iterative application of the transition rule-sets represented impacts of land use changes on land use patterns for a period of 50 years. Environmental attributes of a landscape influence a region’s potential to supply ES, as well as perform as a control factor for individual conversions ⁸¹. In this study, environmental attributes and neighboring land use types were defined for each of the land use scenarios and reflected in the simulations of future land use patterns. For instance, elevation and tidal influences were considered to influence the ranges of mangrove swamps, mixed swamp forests and palm swamp forests. Relatively high elevations on coastal landscapes inhibit mangrove propagule dispersal and impede mangrove forest growth ⁸². Consequently, elevations above and below 14 meters were defined as environmental attributes and iteratively applied to the scenario simulation related to the conversion of mangrove swamp. Similarly, the neighboring cells were defined in order to reflect the proximity effect to the initial land use type as a driver of the land use changes. For instance, based on experts’ opinions, under a landscape restoration scenario, water body (brackish water) can be converted to mangrove swamp with 75% probability if the water body has mangroves as a neighboring land use type, whereas the probability of mangroves conversion to built-up areas is 80% under urbanization scenario if the mangroves are located adjacent to built-up areas.

The simulated land use patterns were coupled with the ES assessment matrix (Table 4) in GISCAME to display potential ES supply by the study region. The potential of the study region to provide ES was calculated as mean values for the ES supplied by each land use cell. Consequently, the final assessment score reflected the mean potential ES supply of the study region influenced by each of the scenarios. The assessed ES values for land use scenarios were presented in spider charts, provisioning maps and ES balance tables. The influence of land use scenarios on the spatial extent of critical coastal habitats were interpreted as the percentage change in the habitats’ extent based on original and simulated land use patterns. Similarly, the scenarios’ influence on ES recovery were derived as the difference between ES values based on original and simulated land use patterns. Increase of a certain ES and concurrent decrease of another represented trade-offs in ES while positive and negative synergies typified simultaneous increase or decline of multiple ES respectively ²⁸.

3.1 Impacts of coastal land use scenarios on future land use patterns and ES supply

Simulation results of the impacts of UBS, UGS, PLAS and LRS were presented as future land use patterns and ES supply in the region (Fig. 4, Fig. 5, Fig. 6 and Fig. 7). The impacts of UBS on future land use patterns and ES supply are presented by comparing with the status of the current land use pattern (Fig. 4). Under urbanization influence, built-up area showed prominent expansion towards the southern edges and mid-east portions of the landscape compared to the current land use pattern. The expansion of built-up area mostly displaced mangroves, palm swamp forests, bog plains and food crop areas which were neighboring land uses along the same area of the coastal landscape. However, food crop areas located towards the northern edges of the landscape were moderately influenced by intensifying urbanization within a 50-year timescale. Similarly, mixed swamp forests and water bodies located on the mid-west portions of the landscape remained relatively intact and less influenced by urbanization land use patterns. The integration of the simulated land use patterns with the ES assessment matrix (Table 4) displays the potential ES supply of the region, influenced by the scenarios. Considering urbanization scenario at decadal timescale, the results revealed decreased potential of the region to supply food, fuelwood, carbon sequestration and recreation benefits compared to the current status. Except for carbon sequestration which remained constant, the region’s potential to supply food, fuelwood and recreation benefits further declined due to intensifying urbanization (Fig. 4).

Table 4: Final assessment matrix showing normalized values for land use types and their potential to supply ecosystem services from 0 (lowest) to 100 (highest)

LULC	Ecosystem Services
	Food	Fuelwood	Carbon Sequestration	Recreation

Mangrove swamp	100	100	48	100
Mixed swamp forest	33	50	47	27
Palm swamp forest	33	5	50	21
Bog plain	10	1	50	41
Shrubland	5	22	7	95
Rubber	0	8	52	6
Coconut	5	29	28	27
Oil Palm	4	37	28	17
Bare surface	0	0	0	0
Built up	0	0	5	81
Water	9	0	5	87
Food crop	100	11	100	6

The simulation results of UGS compared to the current land use pattern are shown in Fig. 5. Both 10-year and 50-year timescales of urban greening showed expanded built-up areas on the southern edges of the landscape. Furthermore, under the influence of UGS, bog plains in the current land use pattern transitioned to shrublands within the vicinity of built-up spaces. In contrast to urbanization, UGS resulted in slight expansion of mixed swamp forests. Except recreation benefits which remained relatively stable between the current land use and 10-year urban greening timescale, the overall ES balance of the region was negatively influenced by UGS as shown in the spider chart (Fig. 5).

As shown in Fig. 6, the simulation results of PLAS in comparison to the current land use pattern revealed key trends in rubber plantation expansion over the landscape. Bog plains, oil palm and coconut located along the southern boundary of the landscape transitioned to rubber by plantation agriculture over 10-year timescale. With the intensification of plantation agriculture over a 50-year timescale, rubber became prominent and dominated land uses along the fringes of mixed swamp forests. The resultant impacts of PLAS on the region’s potential to supply ES showed mixed results as depicted in the spider chart. Whereas the region’s potential to supply food remained constant between the current land use and PLAS at 10-year timescale, food supply potential decreased with intensification of PLAS at 50-year timescale. Fuelwood supply potential initially increased under 10-year PLAS but later decreased with intensification of PLAS at 50-year timescale. On the other hand, carbon sequestration potential increased progressively from the current land use to PLAS at 10-year and 50-year timescales. On the contrary, the overall influence of PLAS on recreation benefits was negative as depicted by decreasing values in the balance tables.

In Fig. 7, the simulated results of the impacts of LRS on future land use patterns and ES supply are shown by comparing with the current land use. Mixed swamp forest expanded in the mid-west portions of the landscape under the influence of landscape restoration over a 10-year timescale. Intensification of restoration at 50-year timescale was accompanied by greater expansion of mixed swamp forest on the mid-west and east portions of the landscape. Similarly, rubber showed progressive expansion, particularly along the fringes of mixed swamp forest considering landscape restoration at 10-year and 50-year timescales. As shown by decreasing values in the ES balance tables, the overall influence of landscape restoration on the region’s potential to supply food and recreation benefits was negative. Conversely, the region’s potential to sequester carbon and supply fuelwood increased under the influence of landscape restoration compared to the current land use.

3.2 Impacts of coastal land use scenarios on spatial extent of critical habitats and arable land uses across temporal scales

Differences in the impacts of the scenarios on the spatial extent of critical coastal habitats (mangrove swamp, mixed swamp forest, palm swamp forest, bog plain and water) and interconnected terrestrial habitats (shrublands) were revealed by comparing scenarios across 10-year and 50-year timescales (Table 5 and Table 6). Regarding 10-year timescale, mangrove swamp decreased slightly in spatial extent and by the same magnitude under the influences of UBS and UGS (-0.9%). Similarly, mangroves decreased in spatial extent under LRS (-0.5%) but remained unchanged under PLAS influence. Furthermore, considering 50-year timescale, mangroves exhibited further decline in spatial extent under UBS (-1.5%) and LRS (-0.8%) respectively, as shown in Table 6. Conversely, mixed swamp forests increased in spatial extent under UGS (0.8%) and LRS (2.6%) and remained stable under UBS and PLAS influences, considering 10-year timescale. In addition, mixed swamp forests recorded 4.3% and 8.3% increase in spatial extent respectively, under the influences of UGS and LRS, considering 50-year timescale, which represent triple and five-fold expansion in the spatial extent of mixed swamp forests. Palm swamp forests remained stable under the influences of all the scenarios and their temporal scales. Bog plain declined under the influences of all the scenarios and their temporal scales, except UBS which exhibited zero influence on bog plain. The influence of all the scenarios on water in the region was not significant as its spatial extent remained stable across the respective temporal scales and scenarios.

As shown in Table 5 and Table 6, UBS and UGS significantly and inversely influenced changes in the spatial extent of shrubland and built-up areas. Considering 10-year timescale, shrubland declined in spatial extent under UBS (-24.9%) and UGS (-23.3%), whereas built-up areas simultaneously increased by 28.2% and 30.5% under the respective scenarios. Shrubland further declined in spatial extent under the influences of UBS (-29.3%) and UGS (-26.6%) with concomitant increase in built-up areas by 33.6% and 33.3% under the respective scenarios considering 50-year timescale.

Regarding arable land uses, rubber increased under the influence of PLAS (2.6%) and LRS (1.6%), considering 10-year timescale but recorded 11% and 2% respectively under the influence of same scenarios considering 50-year timescale. However, the influences of UBS and UGS on rubber plantation were neutral across all temporal scales. While the influence of PLAS on the spatial extent of food crop remained zero across temporal scales, food crop decreased between 2.3% and 2.6% in spatial extent across UBS, UGS and LRS considering all temporal scales.

Table 5. Areal change (%) of LULC types under UBS, UGS, PLAS and LRS over 10-year timescale. The temporal scale signifies the period required for land use transitions to manifest in reality. The values indicate the difference (%) in the areal coverage of LULC types compared to the areal coverage of same LULC types in the current land use map (Fig.3).

LULC Types	Scenarios
	Change (%) in spatial extent of LULC types
	UBS	UGS	PLAS	LRS
Mangrove swamp	-0.93	-0.93	0	-0.54
Mixed swamp forest	0	0.82	0	2.6
Palm swamp forest	0	0	0	0
Bog plain	0	-2.44	-1.67	-0.15
Shrub land	-24.9	-23.28	-5.97	0.53
Rubber	0	0	2.58	1.55
Coconut	0	-2.22	-0.68	-1.49
Oil palm	0	0	5.73	-0.05
Bare surface	-0.09	-0.1	0	0
Built-up	28.24	30.46	0	0
Water	0	0	0	-0.06
Food crop	-2.33	-2.33	0	-2.36

Table 6 Areal change (%) of LULC types under UBS, UGS, PLAS and LRS over 50-year timescale. The temporal scale signifies the period required for land use transitions to manifest in reality. The values indicate the difference (%) in the areal coverage of LULC types compared to the areal coverage of same LULC types in the current land use map (Fig.3).

LULC Types	Scenarios
	Change (%) in spatial extent of LULC types
	UBS	UGS	PLAS	LRS
Mangrove swamp	-1.54	-0.93	0	-0.76
Mixed swamp forest	0	4.31	0	8.32
Palm swamp forest	0	0	0	0
Bog plain	0	-2.6	-2.58	-0.3
Shrub land	-29.29	-26.61	-6.08	-4.59
Rubber	0	0	10.52	2.01
Coconut	0	-5.03	-1.85	-1.92
Oil palm	0	0	0	-0.08
Bare surface	-0.1	-0.1	0	0
Built-up	33.55	33.27	0	0
Water	0	0	0	-0.06
Food crop	-2.62	-2.33	0	-2.61

3.3 Trade-offs and synergies between potential ES supply

Trade-offs and synergies were evident in the potential supply of ES by the region considering the scenarios and their respective timescales (Table 7 and Table 8). UBS resulted in negative synergies between the potential supply of food, fuelwood, carbon sequestration and recreation benefits. Similarly, UGS created negative synergies between the potential supply of food, fuelwood, carbon sequestration (Table 7) and recreation benefits (Table 8). Synergies and trade-offs associated with PLAS were slightly more nuanced at different temporal scales. PLAS within a 10-year timescale created positive synergies between fuelwood supply and carbon sequestration. Trade-offs resulted between the potential supply of fuelwood, carbon sequestration and recreation benefits for the same scenario (Table 7). However, intensifying PLAS within a 50-year timescale resulted in negative synergies between potential supply of food, fuelwood and recreation benefits and trade-offs between such ES and carbon sequestration (Table 8). LRS created negative synergies between potential food supply and carbon sequestration considering 10-year timescale (Table 7). Furthermore, intensification of LRS at 50-year timescale resulted in negative synergies between food supply and recreation benefits and positive synergies between fuelwood supply and carbon sequestration. Furthermore, LRS was associated with trade-offs between the supply of fuel wood and carbon sequestration and potential food supply and recreation benefits.

Table 7

Trade-offs and synergies between potential ES supply values in the study region considering 10-year timescale. Changes in ES values were identified by the differences in ES values of the current land use and ES values of simulations of UBS, UGS, PLAS and LRS (reference values in the balance tables in Fig. 4, Fig. 5 Fig. 6 and Fig. 7). Across each scenario, positive values depict positive synergies and negative values illustrate negative synergies. Both positive and negative ES values across each scenario illustrate trade-offs in the supply of the respective ES.
Coastal land use scenario	Ecosystem Services
	Food	Fuelwood	Carbon sequestration	Recreation
UBS	-4	-7	-3	-2
UGS	-4	-7	-4	0
PLAS	0	+ 1	+ 2	-6
LRS	-2	0	-1	0

Table 8

Trade-offs and synergies between potential ES supply values by the study region considering 50-year timescale. Changes in ES values were identified by the differences in ES values of the current land use and ES values of simulations of UBS, UGS, PLAS and LRS (reference values in the balance tables in Fig. 4, Fig. 5 Fig. 6 and Fig. 7). Across each scenario, positive values depict positive synergies and negative values illustrate negative synergies. Both positive and negative ES values across each scenario illustrate trade-offs in the supply of the respective ES.
Coastal land use scenario	Ecosystem Services
	Food	Fuelwood	Carbon sequestration	Recreation
UBS	-6	-8	-3	-3
UGS	-4	-7	-3	-1
PLAS	-1	-1	+ 4	-7
LRS	-1	+ 2	+ 1	-4

4.1 Influence of experts’ knowledge and perspectives on scenario outcomes

Studies have utilized expert knowledge to develop participatory land use scenarios and to shape the scenario outcomes based on shared preferences and visions for the future^83,84. This is because experts are key knowledge holders in land use decision making processes, hence their knowledge are valuable in land use planning research to address uncertainties and fill data gaps ^85,86. In this study, we elicited experts’ knowledge and perspectives on the drivers of land use changes in coastal environments to complement existing data and develop exploratory scenarios on future sustainable land use pathways. Such knowledge and perspectives underpinned LULC conversions and their transition probabilities in the study region and also influenced the scenario outcomes. For instance, the transition rule-sets suggest that mangroves rather than mixed swamp forests and water bodies can be converted to built-up land uses under the urbanization scenario (Table S1). Spatially explicit representation of urbanization scenario showed that mixed swamp forests and water bodies were persistent in the face of urbanization threats, and consequently, maintained their spatial extents into the future (Fig. 5). This can be explained by the local knowledge of permanent inundated site conditions of mixed swamp forests and water bodies which hamper their conversions to built-up land uses in the study region. Similar studies using participatory approaches also found that understanding of local environmental and biophysical constraints informed the outcomes of future land use scenarios^87,88. Furthermore, the results showed that future land uses driven by urbanization, and which manifest in reality, as transitions to built-up areas, became pronounced and intensified over time, within the immediate vicinity of coastal ecosystems along the southern borders of the region (Fig. 6). Such locations experiencing urban expansion were coterminous with the range of mangrove ecosystems, thereby reinforcing perceptions of mangrove forests vulnerability to urbanization pressures.

The results also showed that experts perspectives on coastal landscape restoration favored transitions from other land uses to mixed swamp forests, mangrove swamps, shrublands, food crop and rubber (Table S1). Such perspectives suggest that experts perceive coastal restoration more broadly, and inclusive of land use options that maintains the heterogenous characteristics of the landscape. However, spatially explicit representation of restoration scenario revealed expansion of rubber plantation within the vicinity of mixed swamp forests (Fig. 7). By utilizing spatially explicit scenarios to showcase experts’ perspectives on future land uses, our approach reveals place-based and temporal risks to, and opportunities for increasing, future ES supply in the study region in the contexts of urbanization and restoration (see section 4.2).

4.2 Future supply of ecosystem services

The analysis of ES supply provides guidance for the planning and implementation of ecological restoration^89,90. By applying spatially explicit scenarios to inform decisions on the locations for enhancing ES supply in the region through restoration, and where risks to ES supply due to urbanization can be avoided, our approach bridges the gap between the theory and practice of integrating the concept of ES in land use planning ⁹¹. The land use scenarios in this study represented major future land use changes. Potential impacts of the scenarios were illustrated by rearranged land use patterns and their synergistic or trade-off effects between multiple ES. The impact assessments of UBS and UGS (Fig. 4 and Fig. 5) showed temporal decline in the region’s potential to supply food, fuelwood, carbon sequestration and recreation benefits. The negative synergies between such ES can be explained by the fact that, according to the UBS, built up areas replaced mangrove swamp (Table S1) which are considered to have high potential to provide multiple ES such as food, fuelwood and recreation benefits (Table 4). Mangroves provide spawning, feeding and resting habitats for a diversity of fish species, hence, are rich sources of fish food ⁹². In the artisanal fisheries sector, there is high preference for mangrove wood for smoking fish in traditional ovens⁵². Besides, mangrove ecosystems are important destinations for nature-based recreation and low impact tourism activities. Similarly, food crop land, including agroforestry systems which have high potential for food supply and carbon sequestration (Table 4) were converted to built-up areas (Table S1) which have neither the potential to supply food nor sequester carbon. Although built up areas showed moderate levels of potential to provide recreation benefits (Table 4), this potential was concentrated along the southern portions of the landscape according to the UGS (Fig. 5). With the concentration of recreation benefits to few locations on the landscape, the results showed decline in recreation benefits over time. This implies the creation of green spaces in developed urban areas will not necessarily improve recreation benefits, especially where urban green spaces are not interconnected with other green land uses on the entire landscape.

Impact assessment of PLAS revealed steep declines in recreation benefits alongside positive synergies between fuelwood and carbon sequestration over a decade. However, such positive synergies reversed to trade-offs between fuelwood and carbon sequestration as plantation agriculture intensified into the future (Fig. 7). Rubber, oil palm and coconut showed low potential to supply recreation benefits. Land use conversions which favored such plantations explains the decline in recreation benefits according to the PLAS (Fig. 6). It is noteworthy that rubber had a significant influence on the land use configuration and dominated the LULC according to the PLAS (Fig. 6 and Table 6). Although rubber fuelwood is utilized to meet energy demands at the household level, the region showed a low potential to supply rubber fuelwood (Table 4). This can be attributed to the prohibitive costs associated with harvesting rubber trees for fuelwood.

Impacts on potential ES supply of future land use patterns linked to restoration over a decade showed decline in food and carbon sequestration which resulted in negative synergies between such ES. However, food and carbon sequestration potential increased and resulted in positive synergies with the intensification of restoration. This reinforces understanding of the relationships between long-term restoration actions and ecosystem services supply ⁸⁹. Furthermore, analysis of restoration land use patterns indicates dominance of mixed swamp forest and rubber in future land use patterns. Mangroves showed reduction in extent under the LRS. Yet mangrove swamps in the region have relatively high potential to supply food, fuelwood and recreation benefits. This can explain the low potential supply of food which resulted from the LRS. Given their relatively high potential for multiple ES supply, mangroves require inclusion in future restoration land use patterns through effective land use and habitat restoration planning (see section 4.3).

4.3 Implications for land use and habitat restoration planning

Coastal habitats in developing regions such as the study area, are vulnerable to deforestation and degradation from pressures of urbanization and over-exploitation ¹¹. Coastal habitat restoration is a complex undertaking that involves deployment of a range of passive natural recovery strategies or active human-mediated conservation actions to achieve pre-defined outcomes ⁹³. In the study area, conservation planning decisions are focused on restoring degraded mangrove swamps without regard for ES considerations or protection of habitat diversity. In this vein, our results are instructive for optimizing the benefits of conservation through multiple habitat restoration planning. Collectively, mangrove swamp, mixed swamp forest and palm swamp forests comprise the peat swamp forests of Southwest Ghana⁵¹. Peatlands are recognized for their multiple conservation benefits⁹⁴. However, our results show the loss of mangrove swamps considering urbanization and landscape restoration futures (Table 5 and Table 6). Urbanization threats which pose risks to mangrove restoration success are prominent along the southern edges of the coastal landscape (Fig. 4). This implies, in order to pave way for successful mangrove restoration and realize the ES supply potential of mangrove swamps in the region (Table 4), urbanization threats have to be addressed. This can be achieved by applications of planning regulations that restrict conversion of mangroves into built up areas along the intertidal zones of the region. Relatedly, loss of mangroves associated with landscape restoration highlights the need for human-mediated restoration actions to consider habitat complexity and other environmental factors such as elevation and changes in tidal flows which can impede mangrove restoration success ^82,95.

The results also highlight the need to re-evaluate conservation planning opportunities in the region as mixed swamp forests showed potential for persistence in the face of urbanization and expansion in spatial extent considering landscape restoration (Table 5 and Table 6). This finding suggests that, landscape restoration planning in the region should also prioritize mixed swamp forests in order to generate greater ecological benefits from multiple coastal habitats. Activation of planning instruments to guarantee future protection for mixed swamp forests will be necessary to avert their potential conversion to paddy rice fields ⁹⁶. Similar landscape protection regulations will be essential for maintaining the stability in the spatial extent of palm swamp forests (Table 5 and Table 6) into the future. Results of the study also highlight other risks to restoration in the region such as rubber expansion along the fringes of mixed swamp forests (Fig. 7). Additional data and validation processes will be required to evaluate how beneficial or otherwise, to biodiversity, of the potential connectivity between swamp forests and rubber plantations. A similar study reported that, significant decrease in ES resulted from conversion of peat swamps to monoculture crops⁹⁴.

Despite the strength of our approach, there are inherent shortcomings which limit its applicability. This relates to data gaps in ecosystem services supply potential of all the mapped land cover classes. This will impact the significance of the results. Besides, the literature values on ES supply potential of land cover classes were based on benefit transfer which could introduce errors in the estimation of potential ES supply of land cover classes due to differences between transfer sites and the study site ⁹⁷. We mitigated error propagation between transfer sites and study sits by selecting studies conducted within Ghana’s coastal zone. Finally, our contribution was to showcase utilization of the concept of ES and land use scenarios for planning and prioritization of multi-habitat restoration efforts in coastal areas. Consequently, utilization of the scenarios in planning should be approached with caution by embracing underlining uncertainty and ambiguity. In that sense, the scenarios present a range of futures, rather than predictions, of urbanization and landscape restoration outcomes, hence provide a starting point to shape pragmatic discourses around coastal land uses within the context of regional and municipal spatial planning.

We presented an approach to develop and assess land use scenarios under the contexts of future urbanization and landscape restoration in coastal areas. Designing four land use scenarios which aligned with urbanization, urban greening, plantation agriculture and landscape restoration futures, provides a starting point for re-examination of conservation planning decisions to facilitate multiple habitats and ecosystem services restoration in Southwest Ghana. Our scenarios building approach elicited regional land management experts’ knowledge of the future drivers of land use transitions in the study region. A key strength of our approach was the opportunity to integrate expert knowledge and benefits transfer data. Integrative approaches were building blocks for ecosystem-based management of coastal areas. Regarding potential ES supply, we found evidence that urbanization decreased the supply of food, fuelwood, carbon sequestration and recreation benefits while landscape restoration scenario revealed both synergies and trade-offs in the supply of relevant ES. Concerning landscape restoration, negative synergies resulted between food and carbon sequestration but this relationship reversed to positive synergies with future intensification of landscape restoration. We also found evidence that mangrove restoration risks failure along the southern boundaries of the landscape due to prevailing urbanization threats which manifest in reality, as land use conversions to built-up areas. Relatedly, our findings showed that mangrove swamps can be reduced in spatial extent under future landscape restoration scenarios, requiring restoration interventions to consider site-specific factors such as elevation and tidal influences to enhance mangrove restoration success. On the basis of our evidence which showed future expansion of mixed swamp forests, no change in the spatial extent of palm swamp forests and decline of mangroves under landscape restoration, we recommend planning regulations to target peat swamp forests in the region to reinforce protection of these ecosystems and avert their future degradation.

Author Contribution

S.K conceptualized the research and wrote the main manuscript text. HK supported conceptualization of the research and provided review comments on the draft manuscript. JNI provided review comments. CF supervised the research and commented on the final draft manuscript.

Acknowledgement

We acknowledge the contribution of the United States Forest Service (USFS) for providing funding support to Hen Mpoano (www. henmpoano. org) under the project, “Enhancing Greater Amanzule Wetland Conservation through Mangrove Ecosystem Monitoring and Management Planning”, within which this research data was gathered. We thank all the workshop participants for sharing their knowledge and insights.

Data Availability

Data is provided within the manuscript or supplementary information files

Resources Center, C. & of the Nation, F. Assessment of Critical Coastal Habitats of the Western Region, Ghana. Integr. Coast. Fish. Gov. Initiat. West. Reg. Ghana. 132 (2011).
Dunne, A. et al. Importance of coastal vegetated habitats for tropical marine fishes in the Red Sea. Mar. Biol. 170, 1–17 (2023).
Nagelkerken, I. et al. The habitat function of mangroves for terrestrial and marine fauna: A review. Aquat. Bot. 89, 155–185 (2008).
McAfee, D. et al. Multi-habitat seascape restoration: optimising marine restoration for coastal repair and social benefit. Front. Mar. Sci. 9, 1–10 (2022).
Carlson, R. R. et al. Synergistic benefits of conserving land-sea ecosystems. Glob. Ecol. Conserv. 28, e01684 (2021).
Gao, S. et al. Research progress on habitat connectivity in coastal waters: A review. Ecohydrology 16, (2023).
Raw, J. L. et al. Dispersal and coastal geomorphology limit potential for mangrove range expansion under climate change. J. Ecol. 111, 139–155 (2023).
Alongi, D. M. Carbon Cycling and Storage in Mangrove Forests. Ann. Rev. Mar. Sci. 6, 195–219 (2014).
Hylkema, A., Vogelaar, W., Meesters, H. W. G., Nagelkerken, I. & Debrot, A. O. Fish Species Utilization of Contrasting sub-Habitats Distributed Along an Ocean-to-Land Environmental Gradient in a Tropical Mangrove and Seagrass Lagoon. Estuaries and Coasts 38, 1448–1465 (2015).
Rodrigues-Filho, J. L. et al. From ecological functions to ecosystem services: linking coastal lagoons biodiversity with human well-being. Hydrobiologia vol. 850 (Springer International Publishing, 2023).
Ramesh, R. et al. Land-Ocean Interactions in the Coastal Zone: Past, present & future. Anthropocene 12, 85–98 (2015).
Aguilera, M. A., Tapia, J., Gallardo, C., Núñez, P. & Varas-Belemmi, K. Loss of coastal ecosystem spatial connectivity and services by urbanization: Natural-to-urban integration for bay management. J. Environ. Manage. 276, (2020).
Stanturf, J. A. et al. Implementing forest landscape restoration under the Bonn Challenge: a systematic approach. Ann. For. Sci. 76, (2019).
Strassburg, B. B. N. et al. Strategic approaches to restoring ecosystems can triple conservation gains and halve costs. Nat. Ecol. Evol. 3, 62–70 (2019).
UN. The United Nations Decade on Ecosystem Restoration. Restor. Ecol. 27, 1–48 (2019).
Duarte, G. T. et al. A user-inspired framework and tool for restoring multifunctional landscapes: putting into practice stakeholder and scientific knowledge of landscape services. Landsc. Ecol. 35, 2535–2548 (2020).
Brown, G. & Donovan, S. Measuring Change in Place Values for Environmental and Natural Resource Planning Using Public Participation GIS (PPGIS): Results and Challenges for Longitudinal Research. Soc. Nat. Resour. 27, 36–54 (2014).
Brown, G. & Fagerholm, N. Empirical PPGIS/PGIS mapping of ecosystem services: A review and evaluation. Ecosyst. Serv. 13, 119–133 (2015).
Refulio-Coronado, S. et al. Coastal and Marine Socio-Ecological Systems: A Systematic Review of the Literature. Front. Mar. Sci. 8, 1–17 (2021).
Liu, N. & Ma, Z. Ecological restoration of coastal wetlands in China: Current status and suggestions. Biol. Conserv. 291, 110513 (2024).
Milner-Gulland, E. J. et al. Accounting for the impact of conservation on human well-being. Conserv. Biol. 28, 1160–1166 (2014).
Stephanson, S. L. & Mascia, M. B. Putting People on the Map through an Approach That Integrates Social Data in Conservation Planning. Conserv. Biol. 28, 1236–1248 (2014).
Annis, G. M. et al. Designing coastal conservation to deliver ecosystem and human well-being benefits. PLoS One 12, 1–21 (2017).
McAfee, D., Reinhold, S. L., Alleway, H. K. & Connell, S. D. Environmental solutions fast-tracked: Reversing public scepticism to public engagement. Biol. Conserv. 253, 108899 (2021).
Koo, H., Kleemann, J. & Fürst, C. Land use scenario modeling based on local knowledge for the provision of ecosystem services in northern Ghana. Land 7, (2018).
Asubonteng, K. O., Ros-Tonen, M. A. F., Baud, I. & Pfeffer, K. Envisioning the Future of Mosaic Landscapes: Actor Perceptions in a Mixed Cocoa/Oil-Palm Area in Ghana. Environ. Manage. 68, 701–719 (2021).
Kankam, S., Inkoom, J. N., Koo, H. & Fürst, C. Envisioning alternative futures of cultural ecosystem services supply in the coastal landscapes of Southwestern Ghana, West Africa. Socio-Ecological Pract. Res. 3, 309–328 (2021).
Koo, H., Kleemann, J. & Fürst, C. Impact assessment of land use changes using local knowledge for the provision of ecosystem services in northern Ghana, West Africa. Ecol. Indic. 103, 156–172 (2019).
Koo, H., Kleemann, J. & Fürst, C. Land use scenario modeling based on local knowledge for the provision of ecosystem services in northern Ghana. Land 7, (2018).
Xiangô, W. The use of scenarios in land-use planning The use of scenarios in land-use planning. (2003) doi:10.1068/b2945.
Allan, A. et al. Developing socio-ecological scenarios: A participatory process for engaging stakeholders. Sci. Total Environ. 807, (2022).
DasGupta, R., Hashimoto, S., Okuro, T. & Basu, M. Scenario-based land change modelling in the Indian Sundarban delta: an exploratory analysis of plausible alternative regional futures. Sustain. Sci. 14, 221–240 (2019).
Saito, O. et al. Co-design of national-scale future scenarios in Japan to predict and assess natural capital and ecosystem services. Sustain. Sci. 14, 5–21 (2019).
Alcamo, J., van Vuuren, D. & Ringler, C. Methodology for Developing the MA Scenarios. Ecosyst. Hum. Well-being. Vol. 2 Scenar. 145–172 (2005).
IPBES. Scenarios and models of biodiversity and ecosystem services - summary for policymakers. (2016).
Kok, K., van Vliet Mathijs, M., Bärlund Ilona, I., Dubel, A. & Sendzimir, J. Combining participative backcasting and exploratory scenario development: Experiences from the SCENES project. Technol. Forecast. Soc. Change 78, 835–851 (2011).
Kok, K., Biggs, R. & Zurek, M. Methods for developing multiscale participatory scenarios: Insights from Southern Africa and Europe. Ecol. Soc. 12, (2007).
IPBES. The regional assessment report on biodiversity and ecosystem services for Africa of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services. ISBN No: 978-3-947851-05-8. Ipbes (2018).
Dada, O., Almar, R., Morand, P. & Menard, F. “Towards West African coastal social-ecosystems sustainability: Interdisciplinary approaches”. Ocean Coast. Manag. 211, 105746 (2021).
Almar, R. et al. Coastal Zone Changes in West Africa: Challenges and Opportunities for Satellite Earth Observations. Surv. Geophys. 44, 249–275 (2023).
Kankam, S., Osman, A., Inkoom, J. N. & Fürst, C. Implications of Spatio-Temporal Land Use / Cover Changes for Ecosystem Services Supply in the Coastal Landscapes of Southwestern Ghana, West Africa. 1–26 (2022).
Koo, H., Kleemann, J. & Fürst, C. Integrating ecosystem services into land-use modeling to assess the effects of future land-use strategies in northern ghana. Land 9, 1–24 (2020).
Moore, E., Howson, P., Grainger, M., Teh, Y. A. & Pfeifer, M. The role of participatory scenarios in ecological restoration: a systematic map protocol. Environ. Evid. 11, 1–9 (2022).
Acosta, A. L., d’Albertas, F., de Souza Leite, M., Saraiva, A. M. & Walter Metzger, J. P. Gaps and limitations in the use of restoration scenarios: a review. Restor. Ecol. 26, 1108–1119 (2018).
Zhou, W. et al. Zoning for Spatial Conservation and Restoration Based on Ecosystem Services in Highly Urbanized Region: A Case Study in Beijing-Tianjin-Hebei, China. Land 12, (2023).
Smiraglia, D., Cavalli, A., Giuliani, C. & Assennato, F. The Increasing Coastal Urbanization in the Mediterranean Environment: The State of the Art in Italy. Land 12, (2023).
Eigenbrod, F. et al. The impact of proxy-based methods on mapping the distribution of ecosystem services. J. Appl. Ecol. 47, 377–385 (2010).
Rosenberger, R. S. & Loomis, J. B. Benefit Transfer of Outdoor Recreation Use Values. A Tech. Doc. Support. For. Serv. Strateg. Plan 59 (2001) doi:https://www.fs.fed.us/rm/pubs/rmrs_gtr072.pdf.
Plummer, M. L. Assessing benefit transfer for the valuation of ecosystem services. Front. Ecol. Environ. 7, 38–45 (2009).
Laze, K. Remote sensing analysis of the changes in vegetated coastland: The case of southeastern Europe. ISPRS Ann. Photogramm. Remote Sens. Spat. Inf. Sci. 5, 55–59 (2021).
Amoakoh, A. O. et al. Testing the contribution of multi-source remote sensing fea- tures for random forest classification of the Greater Amanzule tropical peatland. (2021).
USAID Integr. Coast. Fish. Gov. Progr. West. Reg. Ghana … 1–30 (2011).
Lazar, N., et al. Status of Ghana ’ s small pelagic stocks and recommendations to achieve sustainable fishing 2017. Sci. Tech. Work. Group. USAID/Ghana Sustain. Fish. Manag. Proj. (SFMP). Coast. Resour. Center, Grad. Sch. Oceanogr. Univ. Rhode Isl. (2018).
Adjei Mensah, C., Kweku Eshun, J., Asamoah, Y. & Ofori, E. Changing land use/cover of Ghana’s oil city (Sekondi-Takoradi Metropolis): implications for sustainable urban development. Int. J. Urban Sustain. Dev. 11, 223–233 (2019).
Asante-yeboah, E. et al. Analyzing Variations in Size and Intensities in Land Use Dynamics for Sustainable Land Use Management: A Case of the Coastal Landscapes of South-Western Ghana. 1–28 (2022).
Kleemann, J. et al. Peri-urban land use pattern and its relation to land use planning in Ghana, West Africa. Landsc. Urban Plan. 165, 280–294 (2017).
Owusu-Nimo, F., Mantey, J., Nyarko, K. B., Appiah-Effah, E. & Aubynn, A. Spatial distribution patterns of illegal artisanal small scale gold mining (Galamsey) operations in Ghana: A focus on the Western Region. Heliyon 4, (2018).
Obodai, J., Adjei, K. A., Odai, S. N. & Lumor, M. Land use/land cover dynamics using landsat data in a gold mining basin-the Ankobra, Ghana. Remote Sens. Appl. Soc. Environ. 13, 247–256 (2019).
Gorelick, N. et al. Google Earth Engine: Planetary-scale geospatial analysis for everyone. Remote Sens. Environ. 202, 18–27 (2017).
Tang, W. et al. Big geospatial data analytics for global mangrove biomass and carbon estimation. Sustain. 10, 1–17 (2018).
Duku, E., Agbeko, P., Mattah, D., Angnuureng, D. B. & Adotey, J. Understanding the Complexities of Human Well-Being in the Context of Ecosystem Services within Coastal Ghana. (2022).
Dickinson, D. C. & Hobbs, R. J. Cultural ecosystem services: Characteristics, challenges and lessons for urban green space research. Ecosyst. Serv. 25, 179–194 (2017).
de Groot, R. S., Alkemade, R., Braat, L., Hein, L. & Willemen, L. Challenges in integrating the concept of ecosystem services and values in landscape planning, management and decision making. Ecol. Complex. 7, 260–272 (2010).
Koschke, L., Fürst, C., Frank, S. & Makeschin, F. A multi-criteria approach for an integrated land-cover-based assessment of ecosystem services provision to support landscape planning. Ecol. Indic. 21, 54–66 (2012).
Martin, T. G. et al. Eliciting Expert Knowledge in Conservation Science. Conserv. Biol. 26, 29–38 (2012).
Price, J. et al. Eliciting expert knowledge to inform landscape modeling of conservation scenarios. Ecol. Modell. 229, 76–87 (2012).
Naderifar, M., Goli, H. & Ghaljaie, F. Snowball Sampling: A Purposeful Method of Sampling in Qualitative Research. Strides Dev. Med. Educ. 14, (2017).
Mensah, C. A., Gough, K. V. & Simon, D. Urban green spaces in growing oil cities: The case of Sekondi-Takoradi Metropolis, Ghana. Int. Dev. Plan. Rev. 40, 371–395 (2018).
TCPD. Western Region Spatial Development Framework. 1–209 (2012).
Kankam, S. Land Use Land Cover assessment of Cape Three Points Forest and linked ecosystems: Assessment Report. 56.
Sarku, R. & Appiah, D. O. Analysis of palm fruit value addition for socio-economic development in the Kwaebibirem District, Ghana. 6, 149–167 (2017).
Asante-Yeboah, E., Koo, H. M., Ros-Tonen, M. A. F., Sieber, S. & Fürst, C. Participatory and Spatially Explicit Assessment to Envision the Future of Land-Use/Land-Cover Change Scenarios on Selected Ecosystem Services in Southwestern Ghana. Environ. Manage. (2024) doi:10.1007/s00267-024-01943-z.
Voora, V., Bermúdez, S., Farrell, J. J., Larrea, C. & Luna, E. Palm oil prices and sustainability. (2023).
Umar, H. Y., Giroh, D. Y., Agbonkpolor, N. B. & Mesike, C. S. An Overview of World Natural Rubber Production and Consumption: An Implication for Economic Empowerment and Poverty Alleviation in Nigeria. J. Hum. Ecol. 33, 53–59 (2011).
Coastal Resources Center. Building Capacity for Adapting to a Rapidly Changing Coastal Zone; Graduate School of Oceanography, University of Rhode Island: Narragansett, RI, USA, 2010.
Nero, B. F. Structure, composition and diversity of restored forest ecosystems on mine-spoils in South-Western Ghana. PLoS One 16, 1–22 (2021).
Foli, E. G. & CIFOR. Reshaping the terrain Forest landscape restoration efforts in Ghana. 1–4 (2018).
Fürst, C., Frank, S., Witt, A., Koschke, L. & Makeschin, F. Assessment of the effects of forest land use strategies on the provision of ecosystem services at regional scale. J. Environ. Manage. 127, S96–S116 (2013).
Fürst, C., König, H., Pietzsch, K., Ende, H. P. & Makeschin, F. Pimp your landscape - a generic approach for integrating regional stakeholder needs into land use planning. Ecol. Soc. 15, (2010).
Fürst, C. et al. How to better consider sectoral planning information in regional planning: Example afforestation and forest conversion. J. Environ. Plan. Manag. 55, 855–883 (2012).
Frank, S., Fürst, C., Koschke, L. & Makeschin, F. A contribution towards a transfer of the ecosystem service concept to landscape planning using landscape metrics. Ecol. Indic. 21, 30–38 (2012).
Triest, L. & Van der Stocken, T. Coastal Landform Constrains Dispersal in Mangroves. Front. Mar. Sci. 8, 1–11 (2021).
Lacher, I. et al. Modeling alternative future scenarios for direct application in land use and conservation planning. Conserv. Sci. Pract. 5, 1–17 (2023).
Kariuki, R. W. et al. Integrating stakeholders’ perspectives and spatial modelling to develop scenarios of future land use and land cover change in northern Tanzania. PLoS ONE vol. 16 (2021).
Price, J. et al. Eliciting expert knowledge to inform landscape modeling of conservation scenarios. Ecol. Modell. 229, 76–87 (2012).
Martin, T. G. et al. Eliciting Expert Knowledge in Conservation Science. Conserv. Biol. 26, 29–38 (2012).
Penny, J., Djordjević, S. & Chen, A. S. Using public participation within land use change scenarios for analysing environmental and socioeconomic drivers. Environ. Res. Lett. 17, (2022).
Sandhu, H. et al. Scenario planning including ecosystem services for a coastal region in South Australia. Ecosyst. Serv. 31, 194–207 (2018).
Alexander, S., Aronson, J., Whaley, O. & Lamb, D. The relationship between ecological restoration and the ecosystem services concept. Ecol. Soc. 21, (2016).
Shimamoto, C. Y., Padial, A. A., Da Rosa, C. M. & Marques, M. C. M. Restoration of ecosystem services in tropical forests: A global meta-analysis. PLoS One 13, 1–16 (2018).
de Groot, R. S., Alkemade, R., Braat, L., Hein, L. & Willemen, L. Challenges in integrating the concept of ecosystem services and values in landscape planning, management and decision making. Ecol. Complex. 7, 260–272 (2010).
Corcoran, E.; Ravilious, C.; Skuja, M. Mangroves of Western and Central Africa, UNEP World Conservation Monitoring Center: Cambridge, United Kingdo, 2007. ISBN 9789280727920.l.
Elliott, M., Burdon, D., Hemingway, K. L. & Apitz, S. E. Estuarine, coastal and marine ecosystem restoration: Confusing management and science - A revision of concepts. Estuar. Coast. Shelf Sci. 74, 349–366 (2007).
Tarigan, S. et al. Peatlands Are More Beneficial if Conserved and Restored than Drained for Monoculture Crops. Front. Environ. Sci. 9, 1–12 (2021).
Rastogi, R. P., Phulwaria, M. & Gupta, D. K. Mangroves: Ecology, Biodiversity and.
Owino, C. N., Kitaka, N., Kipkemboi, J. & Ondiek, R. A. Assessment of greenhouse gases emission in smallholder rice paddies converted from anyiko wetland, kenya. Front. Environ. Sci. 8, 1–13 (2020).
Eigenbrod, F. et al. Error propagation associated with benefits transfer-based mapping of ecosystem services. Biol. Conserv. 143, 2487–2493 (2010).

No competing interests reported.

Modeling the Impacts of Coastal Land Use Scenarios on Ecosystem Services Restoration in Southwest Ghana, West Africa

Status:

Version 1

Abstract

Figures

1.0 Introduction

2.0 Methods