Coastal Migration Index for Coastal Flooding Events Increased by Sea Level Rise due to Climate Change: Mexico and Cuba Case Studies

Abstract

This paper presents a coastal migration index (CMI) useful for decision-making in the current scenario of sea-level rise (SLR) due to climate change. The CMI includes coastal human population density, degree of urbanization, and coastal-flooding penetration. Quantitative and qualitative statistical techniques and the geographic information system ArcGIS View 9.0 were used. Further, a panel of fifteen international experts in coastal management issues was consulted to establish and validate the CMI. Results led to three index components based on 22 indicators. CMI was applied in the state of Tamaulipas, Mexico and in Santiago de Cuba province, Cuba. According to CMI estimates, the risk levels associated with SLR for human settlements analyzed in Mexico and Cuba were 5.3% and 11.0%, respectively. The most severely affected communities will require resettlement. Meanwhile, the CMI determined that 15.8% of the Mexican territory studied will be able to withstand the effects of SLR through the management of engineering works that will protect human settlements. The CMI determined that 79.0%, in the case of Tamaulipas, as well as 89.0% of the Cuban territory, will not require new policies or guidelines to promote conservation and protection of coastal natural resources. Lastly, the method used allowed for creation of a CMI stoplight map useful to coastal decision-makers to adopt sound management actions.

1. Introduction

Forecasts and evidence of sea level rise (SLR) in the current era of the Anthropocene engender increased vulnerability of coastal areas around the world [1,2,3,4,5,6,7], and this paper deals with two study areas located in Mexico and Cuba.

Mexico is located at the northern tip of Central America. The country has coasts towards the Gulf of Mexico (GoM) and the Caribbean Sea. It has important geographical features, among them, the Mexican Plateau, the Sierra Madre del Sur, Sierra Madre Occidental, and Sierra Madre Oriental Mountain systems. The latter is inclined towards the Gulf of Mexico between the Isthmus of Tehuantepec and the Rio Grande, including both the Frontal Sierras and the Sierras de San Carlos and Tamaulipas, with basaltic plateaus to the south. Such features influence the behavior of the meteorological variables in the Gulf of Mexico, giving rise to abundant rains and a notable occurrence of cyclones, whose trajectories are determined by pressure differences because of the heating of the northern lands during the summer and the lower temperatures of the Antillean seas and the Caribbean [8]. Along the GoM, SLR has increased, on average, 1.7 mm/year since 1993, with a maximum of 3 mm/year [9]. In coastal wetlands of the GoM, such as those associated with the Mississippi Delta, SLR increases of up to 12 mm/year are reported [10].

The Republic of Cuba is an insular country consisting of an archipelago of various-sized islands that contribute to an extensive coastal zone [11,12,13,14,15]. Coastal flooding took place in 117 different locations in Cuba, 35 of them in the southeastern region [16,17,18], during the 1791–2018 period due to sea level rise, the impact of energetic swell waves, cold fronts, and other extreme meteorological events, e.g., hurricanes. Therefore, SLR constitutes an immediate and growing future danger for the Cuban coast, where an increase of 6.77 cm of sea level was reported from 1966 to 2017 [19,20]. Future projections of SLR increase indicate scenarios of sea level rise of 29.5 cm for 2050 and 95.0 cm for 2100 [19,20]. It is projected that by 2050, Cuba will lose 2691 km² of coastal land to permanent submergence (2.4% of its territory) that, in the 2100 scenario, would increase to 6370 km² (5.8% of its territory). This prediction may underestimate the real losses due to insufficient data from the cays [21].

In Cuba, there are extensive coastal plains and marine terraces at different altitudes [22,23,24]. Submerged coastal plains predominate in the southeastern region [25]. In Santiago de Cuba, 28.6% of rural settlements occur in the coastal area and on riverine flood plains. These settlements face severe risk of flooding [19,20,26,27,28,29,30,31].

Both countries are at high risk because climate change is driving sea level rise and an increase in the intensity and magnitude of extreme meteorological phenomena (e.g., hurricanes). Both threats pose a challenge to the coastal management of the two countries. This paper addresses decision-making for the protection of coastal communities facing the challenge of SLR and a changing climate.

Tools to assess the impact of SLR increases in coastal areas include vulnerability and risk indices [32,33,34,35,36,37,38,39,40,41,42,43]. Some of these are based on territorial, urban development, and socioeconomic indicators. However, none provides land-use planning recommendations for the public to use as a tool to meet the challenge of SLR.

In this paper, we propose a methodology for determining a coastal migration index (CMI) based upon projected SLR values, the presence and typology of urban infrastructure, and population density. For this purpose, analyses and measurements were conducted to assess the following: (1) use of a statistical measure for calculating the value of structures endangered by SLR in the medium- and long-terms and (2) determination of geographic polygons for the application of CMI. Lastly, the method was validated for the coasts of Tamaulipas state in Mexico and the municipality of Guama, Cuba.

This article is written from the perspective of coastal risk management [1,13,15,18]. The goal is to prevent or limit damages due to coastal flooding events linked to extreme hydrometeorological phenomena such as storms and hurricanes the effects of which are enhanced by sea level rise. The proposed CMI is applicable for coastal urban, semi-developed, rural, and natural areas. The CMI is useful for regions with different physical and geographical characteristics where information and scientific data incorporating the effects of coastal forcing factors area available to determine the landward limit of flooded areas in case of energetic events. The extent (distance from shore) of coastal areas potentially affected by flooding processes in each of the two study cases was obtained from existing available information (reports, papers, etc.). Thus, the CMI often will not require additional data collection beyond what is normally available to most communities.

As with other approaches, the CMI provides a data-based provisional assessment of the extent of coastal areas requiring management in the face of SLR. The results presented here address both regional (Mexico) and provincial (Cuba) scales, demonstrating the applicability of the CMI in different situations. While the CMI will aid in management decision-making, it also alerts authorities to the need for further, more detailed studies in particular situations facing extreme energetic events.

2. Materials and Methods

The development of the CMI was based on the use of statistical techniques, a geographic information system (ARCGIS View. 9.0), and consultations with expert panels. SLR was estimated with existing data. Available data allowed us to adjust the CMI to the specific conditions of each country, including hydrodynamic and high-resolution terrain-elevation models [44,45,46]. This research was carried out in three phases: (1) design of the CMI; (2) definition of polygons in the study area for application of the CMI; and (3) integration of results for establishing a CMI stoplight map in the test areas (see Figure 1).

2.1. Steps Followed in the Design of the Coastal Migration Index

2.1.1. Phase 1. Design of the CMI

This article presents the novel concept of developing a coastal migration index (CMI) that is a statistical measure constructed from a standardized weighting of three territorial indicators: (1) coastal flooding-penetration length (distance from shore), (2) population density, and (3) degree of urbanization of the coastal area. This allows for the determination of which of the elements built in the coastal zone are likely to be affected in the short term by the penetration of the sea on land due to the intensity and frequency of extreme hydrometeorological phenomena (hurricanes). It also addresses medium- and long-term, situations taking into account the average elevation of the SLR due to climate change. This index will allow decision-makers at different levels to establish responses for mitigation (coastal engineering), adaptation (conservation of ecosystems), and human migration (actions to move the population and infrastructures to less vulnerable areas) for sustainable coastal management purposes. The CMI provides a traffic light as a graphic expression of the situations resulting from the calculation of the index in coastal areas at different scales. This aids in the communication of risk to decision-makers and the general population.

Step 1: Selecting experts.

The design of the coastal migration index resulted from discussion and consensus among specialists in coastal issues. Criteria for selecting the experts were: (a) professionals and technicians with more than ten years of experience in risk management of coastal areas, (b) extensive knowledge and information on the impact of sea level rise on both countries, (c) experts endorsed by one institution that attests to their expertise or the person being a scientific leader in the discipline. The leaders of the studies were, for the Cuban case, the Center for Multidisciplinary Studies of Coastal Zones (CEMZOC) of the Universidad de Oriente (Santiago de Cuba), and for Mexico, the Center for Research and Development in Port, Maritime, and Coastal Engineering (CIDIPORT) of the University of Tamaulipas (Tamaulipas). Fifteen experts actively participated in the workshops supporting the design and validation of the CMI.

Step 2: Workshops.

We conducted three workshops. The first addressed the question of the need for a coastal migration index. Participants considered relevant literature related to SLR and coastal-vulnerability topics [33,47,48,49,50,51]. A consensus of experts designed an instrument that defines the levels of intervention and adaptation necessary to face the impact of energetic events under a scenario of sea level rise, therefore useful for coastal decision-making, i.e., the protection of coastal human settlements in both countries.

The second Workshop defined the components of the coastal migration index. The experts applied the brainstorming technique and weighed three variables as the most significant to evaluate the impact of sea level rise on coastal settlements. They were: C₁—Coastal flooding-penetration length (km). C₂—Population density at an average width of 1 km. C₃—Degree of urbanization of the coastal area. The data for these components were taken from the official information existing for the two countries (Cuba and Mexico) [1,9,11,18,20,30,45].

Experts in the third workshop determined the weighting factors of the indicators that can interact with each component of the index. An electronic worksheet allowed experts to score the components and their indicators. Each indicator is accompanied by a weight factor, n₁, n₂, and n₃, respectively, which is estimated by the expert valuation of each of the 22 components listed in Table 1. The selected indicators were activities, uses, or policies that can interact with the components and generate changes in these, according to the criteria of the experts. Components were evaluated according to the criteria developed in the workshops by the experts (Table 2).

By consensus, the experts determined the weights of each indicator in the three compartments. They considered the direct impact of the proposed indicators on the behavior of each component using the rating scale previously provided. The values assigned to the indicators for each component were then added, achieving the individual partial weight. The total partial weight was obtained by the sum of the three components.

Step 3: Designing the coastal migration index (CMI).

The quotient between the individual partial weights and the total partial weights was determined, obtaining the relative weight reached by each component of the index in its interaction with the 22 indicators. In this case, decimal numbers up to the value of hundredths, applying rounding up (≥5) or down (<5), were considered. The sum of the relative weights assigned to the components is equal to 1. In correspondence with the relative weight, the weight factors (n₁, n₂, n₃) were determined, taking for each one the respective relative weight values, up to the tenths place, applying the rounding rules.

Once the components and indicators had been evaluated, the coastal migration index was constructed, which resulted in the following mathematical expression:

$$CMI={\left(\frac{\mathrm{Coastal}\mathrm{flooding}\mathrm{penetration}\mathrm{length}}{Max.Pen.Len.}\right)}_{}^{n_1}*{\left(\frac{PopulationDensity}{Max.Pop.Dens.}\right)}_{}^{n_2}*{\left(\frac{Urbanarea}{T.Area}\right)}_{}^{n_3}$$

(1)

Legend: Maximum penetration length (Max.Pen.Len)

      Maximum population density (Max.Pop.Dens).

Next, formulas to calculate three components of the coastal migration index are shown:

$$C_1={\left(\frac{\mathrm{Coastal}\mathrm{flooding}\mathrm{penetration}\mathrm{length}}{Max.Pen.Len.}\right)}_{}^{n_1}$$

(2)

The coastal flooding-penetration length (Equation (2)) is the landward distance (from the shoreline) reached by flooding processes and obtained by a registered historical event or from coastal flooding modeling projected using historical data. The maximum penetration length corresponds to the physical distance between the shoreline and the territory’s physical, geographic limits established for the study area based upon the relief characteristics and the coastal flooding forecasts due to extreme meteorological events. In each case, existing bibliographic data were considered.

Population density is the number of people per km² that live in the risk area of coastal flooding as previously calculated, while the maximum population density is considered as the total amount of people per km² in a certain area (Equation (3)).

$$C_2={\left(\frac{Population Density}{Max.Pop.Dens.}\right)}_{}^{n_2}$$

(3)

The urban area corresponds to the geographical urban space, given in km², that can be affected by coastal flooding, while the total area is the combined urban and non-urban space (Equation (4)).

$$C_3={\left(\frac{Urbanarea}{T.Area}\right)}_{}^{n_3}$$

(4)

The n₁, n₂, and n₃ parameters are the results of the weighting calculus done by the experts. The index values range between 0 and 1 when it is defined that:

$${\displaystyle \sum }_{i=1}^3{n}_i=1$$

(5)

$$CMI \left[0-1\right].$$

$$C_i\left[0-1\right];$$

Through videoconferences, the experts agreed on the results obtained and set about the task of constructing the polygons to which the index would be applied. Methodological graphics of the index design are shown in Figure 2a,b.

2.1.2. Phase 2: Definition of Polygons in the Study Area for the Application of the CMI

Steps 4 and 5: To build polygons using ARCGIS View 9.0 and identification of zones with threat due to SLR using satellite images.

With the use of satellite images and databases of the National Institute of Statistics and Geography (INEGI) of Mexico processed with software of ARCGIS View. 9.0, each polygon was characterized by determining its associated coverage or urban spot, as well as the population density of each. In the case of Mexico, the state of Tamaulipas was used to test the CMI proposed in this paper.

Each polygon has a constant coastal width of 10 km and a varied length of landward penetration delimited by land elevation according to the maximum topographic height that flooding processes could reach in the study area. To do this, the sum of the sea level rise due to effect of wind, waves, atmospheric pressure, and the historical semi-amplitude of the tide in each place is considered. This is the highest value assumed by the tide, registered in a period after the occurrence of hurricanes and other natural events, such as gales and storms. This upper limit is derived from the expectations for a category V hurricane on the Saffir-Simpson scale [19,20]. The approximate maximum altitude above mean sea level is 10 and 5 m for Mexico and Cuba, respectively. We also considered the geomorphological characteristics present in each coastal territory.

In the case of Cuba, the coastal area of the municipality of Guama in the province of Santiago de Cuba was considered using the current administrative political divisions. For Cuba, the information came from the studies of hazards, vulnerabilities, and risks by coastal floods [20,23,52,53], as well as data published by the National Statistics and Information Office in recent years [16,17].

2.1.3. Phase 3: Integration of Results for Establishing a CMI Stoplight Map in the Test Areas

The Coastal Migration Index (CMI) is calculated for each polygon by applying the aforementioned integrated formula. The interpretation ranges for each action to be implemented are determined according to the scale of low, moderate, or high-risk scenarios. Generally, in these ranges, the numerical position that the index values occupy in different polygons is considered with respect to the lower or intermediate limits, as appropriate. In the entire class of data corresponding to the CMI by polygons, the minimum non-zero value was selected, which is assumed as the lower limit (Xmin). It was also necessary to obtain the intermediate limit (Xmed). For this, the maximum limit (Xmax), was determined as the highest value in the class of data under analysis, and the arithmetic mean of the minimum and maximum limits was calculated [Xmed = (Xmin + Xmax)/2].

Ranges for the actions to be implemented are determined according to the low, moderate, or high-risk scenarios, linking them to: REMAIN ON THE SITE (green traffic light) when the risk is low; MITIGATION (yellow traffic light) when moderation of risk requires mitigation measures (such as coastal walls, hydraulic infrastructure against floods, early warning systems, among other actions); and MIGRATE (red traffic light) in the face of high risk that would mainly force the relocation of human settlements or not allow them to expand in such conditions. The result of applying the CMI is expressed in a semaphore. As mentioned, the CMI establishes three scenarios: permanence (low risk), mitigation (moderate risk, allowing staying in a place with appropriate actions), and migration (high risk requiring relocation) (Figure 3).

2.2. Study Areas

Choosing Mexico and Cuba provided the opportunity to compare the CMI in continental and insular coastal settings. We selected the Mexican state of Tamaulipas and the Guama coastal municipality in Santiago de Cuba province, Cuba (Figure 4).

Mexico and Cuba differ in socioeconomical, physical, geographical, and population characteristics (Table 3), but both countries will experience significant impacts due to climate change induced SLR.

Mexico is a coastal country with more than 11 thousand kilometers of coastline, which accounts for 67% of its land, classified as marine territory. About 40% of the population and most of the bigger cities are on the coast. The coast of Mexico is between 32°34′59″ and 14°26′35″ N and between the 117°09′07″ and 87°25′39″ W. Tamaulipas has a population of 3,527,735 people, of which 49.2% are men and 50.8% are women. The GoM, to the south, the central coast on the Pacific Ocean, and the land along the Caribbean Sea are heavily vulnerable to the impact of hurricanes [54]. SLR presents critical scenarios for the Mexican coasts [9,55]. Coastlines of Tamaulipas (one of the 17 Mexican coastal states) are on extensive coastal plains located less than 6 m above mean sea level, rendering many residents and properties vulnerable to the effects of storm-surge impacts [56].

Table 1. Component ratio of the weights, n_i, of each weight, C_i, that make up the CMI.

Categories	Indicators	Criteria for Weighing the Impact Factor of Each Indicator and Their Interaction with the Components of CMI
Policies	Environmental	To consider the impact that the existence (or absence) of environmental policies can have on the implementation of the CMI for decision-makers in environmental-management programs, projects, and initiatives, as well as on its implementation in urban areas with different levels of population density and coastal flooding.
	Social	To analyze the presence of institutions, interest groups, and communities in urban areas that are densely populated and vulnerable to coastal flooding [13,24,30].
	Economic	To analyze the role of economic policies and the allocation of budget items to climate change adaptation and mitigation measures, in particular, the rise in mean sea level and the impact of extreme hydrometeorological phenomena [9,13,24,30,47,55].
	Sectorial	To consider all sectoral conflicts of interest existing in coastal areas.
	Political	To consider the role played by the existence of political will to confront climate change in the execution and implementation of climate-change adaptation and mitigation measures [9,13,18].
Uses	Land uses	To weigh land uses (residential, agricultural, recreational, transport, commercial, others) on an urban scale in the study areas.
	Infrastructure	To weigh the level of urbanization, as well as the presence of heritage, built on the urban scale [13,14,30,47].
Activities	Tourism	To consider the presence of tourism as an important socioeconomic and cultural activity in the studied polygons, as well as the location of facilities and levels of vulnerability.
	Fishing	To weigh fishing as a socioeconomic and cultural activity and considering if it is sport fishing or the subsistence of fishing communities. Considering the existence of fishing areas, diving centers, and training areas, areas for unloading or maintenance of boats or fishing gear due to the impact on decision-making in the application of the CMI.
	Aquaculture	To consider the role of aquaculture as a socioeconomic and cultural activity within the coastal migration index.
	Agriculture	To weigh the role of agriculture as a socioeconomic and cultural activity of coastal communities in relation to the components of the index, as well as the impact of penetration of the sea on crops and the quality of life of inhabitants in the coastal communities.
	Livestock	To analyze the existence of cattle (cows, sheep) in the study areas, as well as pastoral practices in the community.
	Energy	To consider the role and impact of electrical or gas networks in the studied polygons and their interrelation with the study components.
	Ports	To analyze the existence of ports in the study area and their importance for the territory [14,24].
	Commerce	To consider the existence of free zones and maritime or land-trade flows.
Coastal Management	Conservation	To consider stocks of protected areas, national parks, or other protected natural heritage features in the coastal zone [14].
	Legal	To analyze the existence of legal frameworks that support policies for confronting climate change and coastal risks.
	Governance	To evaluate the existence of integration measures for participation of social actors in the implementation, mitigation, and adaptation of actions to face climate change and coastal risks.
	Authority	To consider the existence and implementation of coastal management policies by institutions with competence in these issues.
	Resilience	To weigh the importance of the response and adaptation capacity for communities and natural systems to face the threats associated with climate change.
	Salt intrusion	To analyze saline intrusion as a consequence of sea level rise on currently productive areas on which coastal communities depend [13,14].
	Vulnerability	To weigh different levels of vulnerabilities (structural, ecological, social, economic, functional, non-structural) in the coastal area being analyzed [18,19].

Table 1. Component ratio of the weights, n_i, of each weight, C_i, that make up the CMI.

Categories	Indicators	Criteria for Weighing the Impact Factor of Each Indicator and Their Interaction with the Components of CMI
Policies	Environmental	To consider the impact that the existence (or absence) of environmental policies can have on the implementation of the CMI for decision-makers in environmental-management programs, projects, and initiatives, as well as on its implementation in urban areas with different levels of population density and coastal flooding.
	Social	To analyze the presence of institutions, interest groups, and communities in urban areas that are densely populated and vulnerable to coastal flooding [13,24,30].
	Economic	To analyze the role of economic policies and the allocation of budget items to climate change adaptation and mitigation measures, in particular, the rise in mean sea level and the impact of extreme hydrometeorological phenomena [9,13,24,30,47,55].
	Sectorial	To consider all sectoral conflicts of interest existing in coastal areas.
	Political	To consider the role played by the existence of political will to confront climate change in the execution and implementation of climate-change adaptation and mitigation measures [9,13,18].
Uses	Land uses	To weigh land uses (residential, agricultural, recreational, transport, commercial, others) on an urban scale in the study areas.
	Infrastructure	To weigh the level of urbanization, as well as the presence of heritage, built on the urban scale [13,14,30,47].
Activities	Tourism	To consider the presence of tourism as an important socioeconomic and cultural activity in the studied polygons, as well as the location of facilities and levels of vulnerability.
	Fishing	To weigh fishing as a socioeconomic and cultural activity and considering if it is sport fishing or the subsistence of fishing communities. Considering the existence of fishing areas, diving centers, and training areas, areas for unloading or maintenance of boats or fishing gear due to the impact on decision-making in the application of the CMI.
	Aquaculture	To consider the role of aquaculture as a socioeconomic and cultural activity within the coastal migration index.
	Agriculture	To weigh the role of agriculture as a socioeconomic and cultural activity of coastal communities in relation to the components of the index, as well as the impact of penetration of the sea on crops and the quality of life of inhabitants in the coastal communities.
	Livestock	To analyze the existence of cattle (cows, sheep) in the study areas, as well as pastoral practices in the community.
	Energy	To consider the role and impact of electrical or gas networks in the studied polygons and their interrelation with the study components.
	Ports	To analyze the existence of ports in the study area and their importance for the territory [14,24].
	Commerce	To consider the existence of free zones and maritime or land-trade flows.
Coastal Management	Conservation	To consider stocks of protected areas, national parks, or other protected natural heritage features in the coastal zone [14].
	Legal	To analyze the existence of legal frameworks that support policies for confronting climate change and coastal risks.
	Governance	To evaluate the existence of integration measures for participation of social actors in the implementation, mitigation, and adaptation of actions to face climate change and coastal risks.
	Authority	To consider the existence and implementation of coastal management policies by institutions with competence in these issues.
	Resilience	To weigh the importance of the response and adaptation capacity for communities and natural systems to face the threats associated with climate change.
	Salt intrusion	To analyze saline intrusion as a consequence of sea level rise on currently productive areas on which coastal communities depend [13,14].
	Vulnerability	To weigh different levels of vulnerabilities (structural, ecological, social, economic, functional, non-structural) in the coastal area being analyzed [18,19].

Table 2. Criteria for assessing the contribution of each component to each of the C_i indicators that comprise the CMI.

No.	Scale for Assessing Criteria
1	Null
3	Unimportant
5	Unreferenced
7	Important
9	Very Important

Table 2. Criteria for assessing the contribution of each component to each of the C_i indicators that comprise the CMI.

No.	Scale for Assessing Criteria
1	Null
3	Unimportant
5	Unreferenced
7	Important
9	Very Important

Table 3. Relevant comparative aspects between Mexico and Cuba.

Aspects to Compare	Mexico	Cuba
Land Surface (km2) a	1,964,375	110,860
Coastline (km) b	11,000	5980
Coastline length c	23,761 km	14,519 km
Gross Domestic Product (see FMI, 2016) d	1,063,606	62,705
1990 population (millions of people)	83.94	10.6
2010 population (millions of people)	114.1	11.2
2030 estimated population (millions of people)	138	11
Ecological footprint (hectare per capita) 2007 e	2.60 h	1.95 h
Global Peace Index (2017) f 163 countries ranking.	140	81
Human Development Index (IDH, Spanish acronym) g	0.767 h	0.778 h

Index: ^a,b National Institute of Statistics, Geography, and Informatics of Mexico. Atlas of the physical environment, 1981. https://www.inegi.org.mx/app/biblioteca/ficha.html?upc=702825220525, accessed on 23 May 2021. ^c Coastline length, World Resources Institute. https://es.wikipedia.org/wiki/Anexo:Pa%C3%ADses_por_longitud_de_costa, accessed on 23 May 2021. ^d GDP as FMI 2016. https://economipedia.com/ranking/paises-por-pib.html, accessed on 23 May 2021. ^e https://www.footprintnetwork.org/content/images/uploads/Ecological_Footprint_Atlas_2010.pdf, 2006. Living planet report 2006. https://es.wikipedia.org/wiki/Huella_ecol%C3%B3gica, accessed on 23 May 2021. ^f Global Peace Index 2018, Institute for Economics and Peace (IEP). https://CMIo.org.mx/indice-paz-global-2018-via-iep/, accessed on 23 May 2021. ^g Human Development Index Report, 2018. https://es.wikipedia.org/wiki/Anexo:Pa%C3%ADses_por_%C3%ADndice_de_desarrollo_humano, accessed on 23 May 2021.

Table 3. Relevant comparative aspects between Mexico and Cuba.

Aspects to Compare	Mexico	Cuba
Land Surface (km2) a	1,964,375	110,860
Coastline (km) b	11,000	5980
Coastline length c	23,761 km	14,519 km
Gross Domestic Product (see FMI, 2016) d	1,063,606	62,705
1990 population (millions of people)	83.94	10.6
2010 population (millions of people)	114.1	11.2
2030 estimated population (millions of people)	138	11
Ecological footprint (hectare per capita) 2007 e	2.60 h	1.95 h
Global Peace Index (2017) f 163 countries ranking.	140	81
Human Development Index (IDH, Spanish acronym) g	0.767 h	0.778 h

Index: ^a,b National Institute of Statistics, Geography, and Informatics of Mexico. Atlas of the physical environment, 1981. https://www.inegi.org.mx/app/biblioteca/ficha.html?upc=702825220525, accessed on 23 May 2021. ^c Coastline length, World Resources Institute. https://es.wikipedia.org/wiki/Anexo:Pa%C3%ADses_por_longitud_de_costa, accessed on 23 May 2021. ^d GDP as FMI 2016. https://economipedia.com/ranking/paises-por-pib.html, accessed on 23 May 2021. ^e https://www.footprintnetwork.org/content/images/uploads/Ecological_Footprint_Atlas_2010.pdf, 2006. Living planet report 2006. https://es.wikipedia.org/wiki/Huella_ecol%C3%B3gica, accessed on 23 May 2021. ^f Global Peace Index 2018, Institute for Economics and Peace (IEP). https://CMIo.org.mx/indice-paz-global-2018-via-iep/, accessed on 23 May 2021. ^g Human Development Index Report, 2018. https://es.wikipedia.org/wiki/Anexo:Pa%C3%ADses_por_%C3%ADndice_de_desarrollo_humano, accessed on 23 May 2021.

Cuba’s archipelago includes the island of Cuba, the Isla de la Juventud, and more than 4194 smaller islands and cays with a total area of 109,884 km² [16,17]. The archipelago of Cuba is between 19°49′36″ and 23°17′09″ N and between the 74°07′52″ and 84°57′10″ W meridians, which includes the Cuban Island and islands adjacent to its coasts. Cuba extends 1256 km from Cape San Antonio to Maisi Tip [57]. In 2019, 11,238,000 people lived in Cuba [16,17]. Guama coastal municipality is in the province most vulnerable to hydrometeorological events, such as hurricanes and SLR [52,53,58,59,60,61]. The Cuban study region extends to the north with the state of Gramma, and to the south with the Caribbean Sea [62]. It is in a large and narrow belt that extends from east to west across 104 km, with an area of 94,969 km². Guama has a population of 34,394 people. The population is established in rural and urban settlements in a coastal mountainous relief.

3. Results

3.1. Determination of Polygons for the Application of CMI in the Study Areas

Valuation of the three components, C₁, C₂, and C₃, of the CMI for the Tamaulipas (Mexico) study site was determined by constructing 57 polygons (Figure 5a). In the case of Guama municipality [63] in Cuba, nine polygons were obtained (Figure 5b).

3.2. Assessing the Contribution of Each Component to Each of the C_i Indicators That Comprise the CMI Used

The experts determined the value of the interactions between the components of the index and the 22 indicators selected by them. The values of the coefficients (n₁, n₂, and n₃) of the components (C₁, C₂, and C₃) of the CMI stem from the expert workshops. These are organized in Appendix A and Appendix B.

3.3. Calculation, Interpretation, and Representation for the Index That Determines the Need for Populations to Migrate to Safer Regions

The assessed components (Appendix A and Appendix B) in (Equation (1)) were used for each one of the 57 Mexican and 9 Cuban polygons (see Figure 5a,b). Height above mean sea level established the landward boundary of each polygon, 10 m altitude in the case of Mexico and 5 m for Cuba [64]. Note that the two different boundary altitudes reflect the contrasting landforms in the study sites. We also checked all points affected by coastal floods observed in 2012 due to the effects of Hurricane Sandy (a category III storm on the Saffir-Simpson scale during impact) [18,19,20]. We also considered the forecasts of 28.4 cm and 93.1 cm of flooding in Chivirico, the urban municipal head of Guama (projections for the years 2050 and 2100, respectively) [19,20]. In Mexico, taking into account the estimated increase of 3 mm/year [9], by 2100, approximately 30 cm of sea level rise would be reached.

We included population density and degree of coastal urbanization in defining the nature of each polygon. Results obtained for the components, C₁, C₂, and C₃, are shown in Appendix C and Appendix D. Calculation of the population density in Guama is based on the reference values from Citma [19,20]. Guama includes a diverse topography, resulting in many isolated settlements.

Value of Coastal Migration Index for Each Polygon of Action

We used the results presented in Appendix C and Appendix D to assess the equation with the exponents determined in Appendix A and Appendix B (respectively for Mexico and Cuba). This produced the CMI values for each of the 57 polygons shown in Figure 6a. Similarly, the nine polygons processed in the case of Cuba are showed in Figure 6b.

The range of values shown in Figure 6a was obtained using the ArcGis Jenks function mapping tool. Its use made it possible to determine the natural groupings inherent in the values obtained from the CMI (Appendix C) and to organize them into three classes linked to their recommendations for action. This classification was as follows: REMAIN for CMI values less than 0.035; MITIGATION for CMI values greater than or equal to 0.035 and less than 0.305; and MIGRATION for CMI values greater than or equal to 0.305 (Figure 7a).

In the case of Mexico, the numerical ranges of the stoplight associated with the CMI values presented in Appendix C (ranging from 0 to 0.636) and the Jenks function of ArcGis were used, which helped to determine the natural grouping inherent in the values obtained from the CMI, allowing them to be classified into three classes linked to their action recommendations, which were: REMAIN for CMI values less than 0.035; MITIGATION for CMI values greater than or equal to 0.035 and less than 0.305; and MIGRATION for MIC values greater than or equal to 0.305.

For the case of Cuba, the numerical ranges of the traffic light associated with the CMI values are presented in Appendix D (which vary between 0 and 0.49). The recommendations for action were: REMAIN for CMI values less than 0.0369; MITIGATION for CMI values greater than or equal to 0.0369 and less than 0.4004; and MIGRATION for CMI values greater than or equal to 0.4004 (Figure 7b).

Using the values shown in Appendix C and Figure 6a, we organized the results by range. The application of the CMI for the site in Mexico scored 79.0% of polygons as permanence, 15.8% needing mitigation, and 5.3% requiring migration. For Cuba, 89.0% of the polygons scored in the permanence range, with 11.1% requiring migration (Appendix D and Figure 6b, respectively). Unlike the situation in Mexico, Cuba presents no polygons requiring mitigation strategies.

4. Discussion

Coastal vulnerability indices developed by different authors have been based on specific territorial characteristics and are dynamic due to changing rates of sea level rise (SLR) [34,39,43,46,50,65,66,67,68,69]. However, many of them depend on local information availability and require annual updates, rendering them difficult to apply in coastal environments [70]. Considering that the current scenario of SLR is increasing faster than previous predictions, it has become necessary to build versatile indices applicable on a more general basis [71].

The CMI proposed in this paper incorporates coastal physical characteristics, the degree of urbanization, and population dynamics for appropriate physical divisions of the landscape [72]. The CMI also accounts for marine dynamics, such as coastal flooding-penetration length. The component is directly associated with the beach slope that is easily applied to distinct places in time and space. Even though authors such as Dou et al. [72] include the beach slope to evaluate their coastal-risk index, the consideration of multiple parameters in CMI evaluation makes it very specific for spatial applications in all types of coastal ecosystems.

Some authors state that not all indices are applicable in all contexts [69]. Some indices used to assess coastal vulnerability in an SLR context are presented at a national scale of application [43,73]. However, most authors validate vulnerability indices at the regional scale [74,75,76,77,78,79,80,81]. The coastal migration index (CMI) was developed to be applied on a regional and local scale. Our proposal lines up with the index designed by McLaughlin and Cooper [39] and the CMI uses commonly available data for the assessment of these three components (1) coastal flooding-penetration length (km); (2) population density at an average width of 1 km; and (3) degree of urbanization of the coastal area. Our proposal is in correspondence with the methodological proposal of Özyurt and Ergin [77]. The authors focus on identifying those regions where the various effects of sea-level rise may be the greatest.

Other indices also include population density, urbanization, and measures of vulnerability [82,83]. However, they use the rise in mean sea level with little account for extreme events. The CMI offers most coastal countries and insular nations an effective tool to address SLR [84]. Further, our CMI proposal was validated in two different countries with good results. In both cases, the CMI provided decision-makers with a novel and comprehensive tool for making choices about adaptation actions in the face of rising sea levels.

Various coastal-vulnerability indices provide information on possible changes in the coastline [85,86]. Some metrics include a hazard subindex [87], a sensitivity indicator [88], and a susceptibility index [82]. The research of Rizzo et al. [89] integrated into one index the analysis of seven subindices that include beach morphology, dune morphology, shoreline evolution, coastal runup, the fluvial system, storm surge, and topography [89]. Bautista and Bautista-Hernández [90] designed a set of routines for calculating derived variables most used in land evaluation, including agroclimatic indices, probability of rainfall by month, and others [90].

The CMI does not analyze changes in coastal dynamics and forcing agents. Once the impacts of flooding on the built and natural heritage of coastal areas have been determined, the index offers a tool to decide what action should be taken by coastal managers. The CMI complements other coastal management tools. To be clear, the CMI does not address the dynamic drivers of coastal environments, nor does it predict the future. Instead, it informs the decision-making process by synthesizing what is already documented about the coastal system at hand.

The CMI expands this approach with the inclusion of the range of values required to inform appropriate decisions by policy makers. This novel aspect is missing from other studies.

The CMI has three basic components: characterization of a coastal area in geomorphological, urban, and demographic terms. This simplicity allows for rapid quantification of the CMI. The basic information for evaluating the CMI is generally available in most of the world’s coastal areas [91]. The versatility of the use and application of the CMI is demonstrated for the evaluation of coastal risks with basic and easy-to-determine information.

The CMI offers the opportunity to manage risks in the face of hydrometeorological threats (Figure 3). This not only applies to human settlements and the built environment but also allows for consideration of adaptation and/or mitigation alternatives to preserve the integrity of ecosystems [92].

The CMI allows the inclusion of important ecological information. For example, the dialectical relationship between biotic potential and environmental resistance to population growth for different populations of organisms can be included. Such aspects were incorporated during the analysis of the interactions between the 22 indicators and the components of the index (Appendix A and Appendix B). The CMI addresses ecological security and ecosystem services by including consideration of biogeochemical cycles, climate regulation and soil formation. Ecosystem services, such as protection against natural phenomena, prevention of coastal flooding, and agricultural and recreational uses of the land depend on these functions.

The analysis of the aforementioned indicators (Table 1) also corresponds to the ecosystem functions of production (environmental measures, energy, and agriculture) of information (social, economic, sectoral, political, risk, tourism, ports, trade, legal, governance, governability), habitats (infrastructure, fishing, aquaculture, livestock) and protection (saline intrusion, resilience).

One of the challenges for the CMI was its validation, which was done on two levels. The first of these was the consultation with local experts to determine the weight factor of each of the three components of the CMI, where the experience of the participants was fundamental in the evaluation of the 22 factors involved. Although the evaluation process has a quantitative representation, it is not without subjectivity. But it is a requirement that the participants in the evaluation have proven experience in coastal management in some of its variants: urban, environmental, and/or risk [93].

The 22 indicators resulting from this research have different scopes. These correspond to the implementation of policies and land uses on the urban scale (states, municipalities, cities). Others correspond to uses, activities, and characteristics of coastal management considering socioeconomic and cultural aspects of the communities that need protection or permanent migration actions. All the indicators determine the behavior of the CMI. The CMI is very flexible to adapt or adjust the indicators according to different realities and contexts.

The second level of validation of the CMI was its application in two coastal regions with very different geomorphological, urban, and sociodemographic characteristics, such as the coasts of the state of Tamaulipas in Mexico (sandy coast with a gentle slope, population centers, and large urban areas) and the coasts of the municipality of Guama in Cuba (rocky coast with steep slopes and lower population density and less urban coverage). An example of this type of risk-analysis approach is also presented by Pricope et al. [94] and Jaranovic et al. [95]. These works mandate information that is more difficult to achieve, requiring extensive information on erosion and the return periods of hydrometeorological phenomena, such as hurricanes and gales.

With these comparative results, it was possible to demonstrate the adaptability of the CMI to assess coastal risks with basic information, and quickly [13,29,96,97], contrary to other indices that require very specific information from each site, which is expensive to obtain [49,98,99,100,101,102,103,104,105,106]. We agree with authors [107,108] who have explained the importance of choosing and implementing specific actions in each polygon according to the communities’ needs and vulnerabilities.

In the case of Mexico, the CMI assessment produced a low percentage of coastal areas in red zones that require relocation (5.26%). The highly concentrated populations, gradual slopes, and fragile infrastructure of the red zones are consistent with their CMI scores. The CMI values provide a technical way to quantify risk and to establish and support policies that promote the relocation of high-risk human settlements or to no longer allow their expansion under such conditions.

Bartlett & Smith [108] and Seenath et al. [109] mentioned the importance of geographic information systems (GIS) for coastal zone management and coastal flooding-vulnerability assessment. In the second article, the authors presented two types of tools that are commonly employed for SLR analysis. The first one is based on hydrodynamic models, and the second one on GIS tools [109]. The CMI takes into account a GIS coastal digital elevation model, as well as storm surge and SLR due to the effect of wind, waves, atmospheric pressure, and the historical semi-amplitude of the tide for the Cuban [11,12,13,14,18,19,20,23] and the Mexican cases [9,55]. In both situations, we used data on bathymetry and elevation for scenarios of SLR, as done by others for Trinidad and Tobago [109].

We demonstrated the applicability of the CMI at regional and provincial scales. This facilitates evaluation across the urbanization spectrum from low to high population densities. Additionally, the use of the traffic-light method contributes to raising the knowledge of the population when it is used within environmental education programs.

The strength of the CMI lies in the consultation of experts for the weighting of its components. The aforementioned validates the locally achieved results as a support in the establishment of public policies and the promotion of sustainable coastal development. Additionally, the CMI allows strengthening actions to mitigate risk in areas of less vulnerability and increase their resilience in the face of increasing intensities and duration of storm surges due to global climate change.

5. Conclusions

The CMI offers a statistical measure built on a few elements to evaluate the risks of damage in coastal zones due to extreme events such as hurricanes in a time of SLR and climate change. This index will enable decision-makers at different levels to establish mitigation (coastal engineering infrastructure), as well as adaptive (ecosystem conservation) and migration responses (displacement of populations and infrastructure to less vulnerable areas) for greater protection.

The CMI provides an affordable and easily adaptable approach for making informed decisions regarding the growing threat of SLR. It combines objective criteria (population density, degree of urbanization, elevation, historic flooding data) with the wisdom of experts familiar with local conditions.

The expert advice ranks the various parameters or indicators in the context of the local conditions. The application of the CMI to portions of coastal Mexico and Cuba demonstrates its validity for suggesting appropriate actions to protect communities in the face of SLR. Policymakers and managers can use the CMI to determine if the community is safe to remain in place (permanence), requires actions to prepare the community to stay put (mitigation), or must move to a new and safer area (migration).