Risk assessment as tool for coastal erosion management

Ocean and Coastal Management 186 (2020) 105099

Contents lists available at ScienceDirect

Ocean and Coastal Management journal homepage: http://www.elsevier.com/locate/ocecoaman

Risk assessment as tool for coastal erosion management Nelson Rangel-Buitrago a, *, William J. Neal b, Victor N. de Jonge c a

Departamentos de Física y Biología, Facultad de Ciencias B� asicas, Universidad del Atl� antico, Km 7 Antigua vía Puerto Colombia, Barranquilla, Atl� antico, Colombia Department of Geology, Grand Valley State University, The Seymour K. & Esther R. Padnos Hall of Science 213A, Allendale, MI, USA c The University of Hull, Department of Biological and Marine Sciences, Hull, HU6 7RX, United Kingdom b

A R T I C L E I N F O

A B S T R A C T

Keywords: Hazard Vulnerability Risk Management Coastal erosion Cartagena

The coastal zone of Cartagena city, Colombia provides illustration for applying risk assessments as a basis for guiding coastal management. Much of this coastal reach is heavily urbanized with a few areas of suburban to rural zones. In this study 32.6 km of shoreline under the city’s jurisdiction were evaluated for coastal erosion management, calculating the Hazard and Vulnerability Indexes, which together constitute the Coastal Erosion Risk Index which provides a single numerical evaluation of risk. A range of variables were utilized in the cal­ culations including forcing processes (wave height, storm surge, littoral exposure), coastal susceptibility (sandy and rocky shores: dune height, percent washovers, beach width, beach slope, cliff/platform, rock types, struc­ ture, weathering), as well as socio-economic, ecological and cultural factors (e.g., land use, percent urbanized, population density, infrastructure, cultural heritage, agency conservation/protection, ethnic communities, etc.). Coastal forcing and susceptibility are the basis for determining the Hazard, resulting mostly from induced sus­ ceptibility as a result of 60 years of over-reliance on shore-hardening structures, interrupting/reducing sediment supply, and over-development in terms of urbanization (loss of protective landforms, irreversible coastline modification, beach narrowing, and the growing coastal squeeze). From a management viewpoint, much of the Cartagena urban area has limited options, but the buffer zone between the shoreline and development must be increased (e.g., no new development; planned retreat), building codes updated (e.g., building floors at/below ground level should be flood-proofed, and have uses compatible with/recoverable from flooding/wave impact), and minimize additional shore-hardening (e.g., utilize beach nourishment). Rural and suburban areas in the region still have time to adopt more stringent management including development bans, disallowing replace­ ment of buildings lost to erosion/flooding, large set-back requirements for new development, utilizing ecosystem management for natural protection, and removal of some groin systems. Virtually all management solutions will be costly, so plans are needed at the government level to develop financing systems.

1. Introduction Coastal erosion can be defined as the removal of material that typi­ cally causes a landward retreat of the coastline (British Geological Survey, 2012). It is a normal process that has existed since land emerged from the sea, and as such, it will exist in the future. Coastal erosion encompasses many different specific processes, but in general terms, it is the result of a suite of natural process and anthropogenic influences that can act in distinct or combined ways (de Jonge, 2009). As a normal process, coastal erosion only becomes a problem where there is no room to accommodate the occurring changes (Neal et al., 2007; Pilkey and Cooper, 2014; Pilkey and Pilkey, 2019; Rangel-Bui­ trago, 2019). In this sense, coastal erosion is a major issue along

low-elevated shore areas as well as coastlines of uplands underlain by unstable rocks, or that are tectonically active, whether rural or urban­ ized (Bell, 1999; Neal et al., 2018; Martinez et al., 2018). Currently, coastal erosion reaches unmanageable magnitudes and impacts due to increasing urbanization along coastal zones (Pilkey and Cooper, 2014; Barragan and Andreis, 2015; Pilkey and Pilkey, 2019; Cooper and Jackson, 2019). Regardless of rates, causes or sites, the coastal erosion processes are becoming a severe hazard to ecosystems and human activities, and therefore must be managed in the best way available. Coastal Erosion Management (CEM) is defined as the interactive, dynamic and multidisciplinary approach to cope the coastal erosion processes (Bush et al., 1996; Rangel-Buitrago and Neal, 2019; Williams

* Corresponding author. E-mail address: [email protected] (N. Rangel-Buitrago). https://doi.org/10.1016/j.ocecoaman.2020.105099 Received 1 October 2019; Received in revised form 8 January 2020; Accepted 8 January 2020 0964-5691/© 2020 Elsevier Ltd. All rights reserved.

N. Rangel-Buitrago et al.

Ocean and Coastal Management 186 (2020) 105099

et al., 2018). Specifically, the CEM process seeks to prevent, mitigate, or in the best scenario, avoid any ecological, social and economic losses derived from the coastal erosion process (Viles and Spence, 1995; Cooper and Pilkey, 2012; Gracia et al., 2018; Rangel-Buitrago et al., 2018a). The CEM is becoming much more necessary and urgent as coastlines constitute the most densely populated and developed land �n et al., 2015; De Andr�es zones in the world (Pilkey et al., 2011; Barraga et al., 2018). Such concentration of population isn’t likely to decline because the coastal zone is the perfect area for a variety of activities (IPCC, 1990; de Ruig, 1998; Gracia et al., 2018; Rangel-Buitrago et al., 2018b). Currently, five strategies are considered as primary CEM practices: Protection: preserving population centers, economic activities and natural resources (vulnerable areas) using hard structures and/or soft protection measures. Accommodation: occupying sensitive regions, but acceptance of a higher degree of flooding, erosion, or other hazards by changing land use, construction methods and improving preparedness. Planned Retreat: removing threatened structures in developed areas, resettling inhabitants, and requiring new development to be set back from the shore, as appropriate. Use of Ecosystems: Influencing processes related to coastal erosion by means of the creation and restoration of coastal ecosystems, such as wetlands (e.g., low marshes and mangroves), biogenic reef structures, seagrass beds, and dune vegetation. Sacrifice (do nothing): allowing property loss when the suggested protection is not viable, or the accommodation and retreat option does not exist. Of these five, the accommodation approach is one of the less used because it is based on the occupation of sensitive areas, accepting the existence of a serious degree of hazard. A sensu stricto CEM based on accommodation lies in the use of approaches that allow the revision and reorganization of human activities in that coastal zone area, or even at the river basin level (Rangel-Buitrago and Neal, 2019). The above means use of land at risk without attempting to prevent the area from being damaged by coastal erosion, and allowing habitat conservation in the best feasible way. Despite the fact that the accommodation approach provides oppor­ tunities for damaged lands to be used for new purposes, the negative economic, social and cultural consequences of its implementation can be significant (Williams et al., 2018). For example, elevated costs may be involved during the planning and restructuring of existing land uses that can generate significant economic stress on local, regional or national budgets. The accommodation approach can be divided into two specific fronts of work: Technologies that allow accommodation/adaptation while coastal erosion is occurring (e.g., flood proofing). Information systems that allow the understanding of coastalerosion risk to develop optimal responses to minimize the related im­ pacts (e.g. with help of geo-indicators, mapping/zoning). Mapping is the standard management approach as preparation for appropriate land use planning, especially in coastal-erosion susceptible areas. Ideally, coastal hazard mapping should be the first step in any CEM scheme, because this is the broadest approach to define those coastal areas which are at risk of erosion. Such mapping consists of a range of map typologies from traditional (e.g., topography) to specific processes (e.g., erosion rate maps). The evaluation and mapping of coastal risk, specifically the coastalerosion risk, are critical issues in the coastal management field, and an extensive literature on assessment methodologies exists (Del Rio and Gracia, 2009; Bonetti and Woodroffe, 2017). Methodologies used to assess coastal-erosion risk can be classified according to different char­ acteristics, but the establishment of a succinct classification often results in a system where limits between classes are not exact (Di Paola et al., 2011; Rizzo et al., 2018). The intrinsic coastal-erosion hazard is

determined using different information types such as physical and ecological coastal features, human occupation, present and future shoreline trends (Gornitz, 1991; McLaughlin and Cooper, 2010; Ng et al., 2014). Initial studies employed single approach methods (i.e., UNEP Methodology, Gornitz, 1991; Gornitz et al., 1994; Carter et al., 1994). These methods progressively evolved and were superseded by more specific techniques. Improved consideration of physical and non-physical factors, as well as the associated uncertainties, has given rise to more consistent methods for assessing and defining coastal haz­ ard, vulnerability and risk assessments (i.e. USGS-CVI, Gornitz et al., 1994; SURVAS, Benassai et al., 2009). The associated maps have been obtained for several coastal sectors around the world through Geographical Information Systems (GIS), computer-assisted multivar­ iate analysis and numerical models (Cooper and McLaughlin, 1998; Gerkema and Duran-Matute, 2017; Silvera and Bonetti, 2019; Skilodi­ mou et al., 2019). Specifically, recent works have been focused on the determination of coastal hazards and risks related to the specific impacts of extreme wave events, and the sea-level rise under a climate change context (Li and Li, 2011; Ng et al., 2014). Risk assessments based on maps provide information on the forces to which the coastal zone is exposed and its adaptive capacity (Bush et al., 1996; Small and Nicholls, 2003). In these kinds of assessments, it is essential to examine not only the interacting physical attributes but also these physical attributes in combination with socioeconomic, conser­ vational, and archaeological–cultural characteristics. Within this context, the considerable amount of information, that must be inte­ grated and processed, requires an organized working methodology to show spatial and temporal relationships between the hazard phenome­ non and the elements at risk. The applied goal of risk maps is risk reduction based on accurate information. Risk maps also support the CEM in working to reduce any impact of coastal erosion. Mapping shows the expected extension of the erosion process and related impacts in a given location based on mul­ tiple variables and diverse scenarios. Stakeholders need a robust and transparent management framework to incorporate risk assessment and mapping as a tool for CEM. This framework must be focused on four specific points: The assessment and mapping of the reality. This only can be achieved with the comprehension of the underlying processes involved in the coastal erosion process. The creation of optimal and comprehensible maps. This will facilitate community involvement on coastal erosion and their man­ agement issues. The use of maps as a starting point to evaluate and determine the optimal strategy for coastal erosion control and management. This is achieved by taking into account the viewpoints of all stake­ holders as well as all the technical concepts. The planning takes into account a long-term perspective. Always giving specific solutions and flexible measures. Risk mapping is usually the starting point for zoning which is the division of an area into zones assigned for different uses or restrictions, based on the detailed knowledge of the erosion hazard to reduce coastal vulnerability. Adequate zoning can provide efficient mechanisms for allocating coastal space for appropriate uses based on the suitability with environmental, social and economic conditions, as well as compatibility with sustainable development objectives and principles, and with policies and legal requirements. In essence, coastal zoning schemes can constitute the perfect regulatory and planning framework for CEM and other management issues (Veersalu, 2011). This paper deals with a methodological approach to the coastalerosion risk determination through the use of matrices concerning physical parameters, socioeconomic activities, ecological units and cultural assets. The goal is to assess coastal-erosion risk as a manage­ ment tool, opening and improving opportunities for optimal coastal zoning as exemplified for the city of Cartagena located on Colombia’s 2

N. Rangel-Buitrago et al.

Ocean and Coastal Management 186 (2020) 105099

Caribbean coast. The approach presented here is based on the selection and evaluation of three types of variables: i) the forcing variables contributing to coastal erosion, ii) dynamic variables that determine the resilience to erosion (susceptibility) and iii) the vulnerable targets grouped in three different contexts (socioeconomic, ecological and cultural). These three groups are mathematically combined into two separate indices, the Hazard Index (combining forcing and suscepti­ bility) and the Vulnerability Index, which together constitute the Coast Erosion Risk Index which is a single numerical measure of the risk.

due to changes in the influence of the trade winds of the Northern Hemisphere. These seasonal changes also influence wind and wave conditions. Significant mean wave height and peak period along the study area are 1.5 m and 7.5 s, respectively (Rangel-Buitrago and Anfuso, 2015; Orejarena-Rondon et al., 2019). Wave systems are dominated by NE swells from November to July when shores in the area periodically experience large waves (up to 3.5 m, with a regional average significant height of 1.5 m). From August to December, waves are variable from the NW, WSW and even SW. These seasonal variations in wave direction occur with a decrease in significant wave height. As a result, the highest energy conditions occur from November to July. Along-shore sand drift is dominantly southwestward along this entire reach, except for minor reversals due to head-land induced wave-refraction patterns, and also during rainy periods when southerly winds become dominant and set up short waves (Rangel-Buitrago et al., 2015). The dominant direction of sand transport is obvious in the large spit structures, and patterns of down-drift sediment starvation due to groins and jetties. Tides are mixed, semi-diurnal, microtidal (ranging from 0.3 to 0.5 m during spring and neap tides). Coastal flooding occurs in low-lying areas during perigean spring tides. Tectonic, climate and oceanographic processes have resulted in coastal settings characterized by i) dissipative beaches and sand spits composed of sand sediments of terrigenous and carbonate origin; ii) marine terraces and cliff sectors, formed by Tertiary sandstones; iii) coastal plains associated with fluvial-marine sedimentary processes, iv) log-spiral shores and bays and v) coastal lagoons with mangrove swamps (Stronkhorst et al., 2018). This coastal morphology is characterized by a sediment deficit resulting in erosion processes (Rangel-Buitrago et al., 2018c; Orejarena-Rondon et al., 2019). Changes in erosion processes were induced during 1935 by the construction of the jetty of Bocas de Cenizas which is the mouth of the Magdalena River (The major sediment supplier on the Caribbean coast of Colombia) in the neighboring city of �pez, 2008). By a few decades ago, the Barranquilla (Restrepo and Lo evident coastal erosion and interests to preserve the historical and cul­ tural heritage in the study area had motivated the implementation of hard protection measures (Rangel-Buitrago et al., 2018b). Cartagena has a population of ~1 million inhabitants, 96% in its urban area and the remaining 4% in rural sectors. The Cartagena Mu­ nicipality is the fifth-largest urban area in Colombia. Currently, its coastal occupation shows important tourist activities linked to both cultural resources and beach attractiveness.

2. Study area Cartagena city is located at the central Caribbean coast of Colombia (Fig. 1). Founded in 1533, this city was the most important Spanish settlement in the Americas during colonial times. Currently, Cartagena is the most famous tourist city in the country, and also the main port on the Colombian Caribbean coast. Specifically, this work is focused on a 32.6 km length coastal strip located between Punta Canoas (N) and El Laguito (S), including a sector of Tierrabomba island located in front of Cartagena (Fig. 1). The climatic and oceanic setting of the study area largely determine coastal erosional processes. Rainfall patterns, in this semi-arid tropical environment, are bimodal with wet periods in April/May and October/ November, and dry periods in between. Although the average monthly ~ o Southern Oscillation strongly affects rainfall is 20.2 mm, the El Nin precipitation patterns (Blanco et al., 2006; Molares-Babra and Mestres-Ridge, 2012). So, the annual range is from 300 to 1400 mm/yr,

3. Materials and methods 3.1. Coastal erosion assessment Shoreline change and evolution trends are essential issues because they can provide an understanding of complex and dynamic large scale systems (Addo and Lamptey, 2013; Stanchev et al., 2018). The 32.6 km of shoreline that composes the study area were assessed for the 1984–2019 period by the use of satellite images downloaded from the earth explorer server (https://earthexplorer.usgs.gov/). An important issue is the selection of an adequate feature as a shoreline indicator which correctly reflects real shoreline position and thus its evolution over time (van Heuvel and Hillen, 1994; Crowell et al., 1999; Boak and Turner, 2005). For differentiation and later digitizing, the fourth band on the Landsat images was selected. In a micro-tidal environment, such as the study area, this shoreline indicator can be defined as the waterline when the image was taken (de Ruig, 1998; Boak and Turner, 2005; Alberico et al., 2012). After shoreline position identification, images were digitized in a GIS environment (ARCGIS 10) for subsequent shoreline change analysis. The DSAS 3.2 extension for ArcGIS, developed by the USGS (Thieler et al., 2005), was used to estimate shoreline changes. DSAS uses as an input a series of shoreline positions, which need to be referenced to an arbitrary

Fig. 1. General location of Cartagena city, Colombia, and shoreline study area Districts. The shaded area in inset B represents the Bolivar Department. 3

N. Rangel-Buitrago et al.

Ocean and Coastal Management 186 (2020) 105099

baseline. Transects perpendicular to the shoreline were generated at 500 m intervals. The DSAS allowed calculation of erosion/accretion rates between points, based on the distance between them and the elapsed time, assuming changes to be linear processes (Thieler et al., 2005). Retreat/accretion rates were calculated and grouped into four cate­ gories “high erosion” (�-1.5 m/yr), “erosion” (between 0.2 and 1.5 m/yr), “stability” (between 0.2 and þ 0.2 m/yr) and “accretion” (�þ0.2 m/yr) according the methodology presented by Rangel-Buitrago et al. (2015).

Cf and CS are Coastal Forcing and Coastal Susceptibility values assigned (ranging from 1 to 5) and n is the number of indicators used for the index calculation. Hazard ¼

P Vulnerability ¼

CSn nCS * 100 nCS*n

(4)

Vn is vulnerability values assigned (ranging from 1 to 5) and n is the number of indicators used for the index calculation.

The study area was divided into 61 segments of 500 m length. For each segment the Coastal Erosion Risk Index was calculated. This index is based on the mathematical combination of two sub-indices within a GIS environment according the methodology proposed by Rangel-Bui­ trago and Anfuso (2015). These two sub-indices are: Hazard sub-index which is composed by forcing variables contrib­ uting to coastal erosion (Table 1) and the Susceptibility which describes the coastal resilience and susceptibility to erosion according to its spe­ cific morphological characteristics (Table 2). Four factors were chosen as variables for coastal forcing index estimation (Table 1). Similarly, the susceptibility sub-index was estimated as a function of the intrinsic coastline characteristics and twelve variables were chosen (5 for sandy coast and 7 for rocky coast, Table 2). Vulnerability sub-index which is concerned with vulnerable tar­ gets, including socio-economic, ecological and cultural aspects (Tables 3–5). Analysis of potentially at-risk targets also includes a series of vulnerability related variables in the socio-economic, ecological and cultural contexts. A total of twenty one variables (8 socio-economic, 7 ecological and 6 cultural) were selected (based on its availability) to obtain the vulnerability sub-index (Tables 3–5). The designated variables (for hazard and vulnerability sub-indexes) were classified on a 1–5 scale; 1 indicating a low contribution to the specific key variable for the studied sector, while 5 indicates a high contribution. Classes were set on a numerical base, and an ordinal scale approach was adopted in the case of any semi-quantitative variable difficult to quantify (Cooper and McLaughlin, 1998). All variables were combined under a GIS environment into the Forcing, Susceptibility, Hazard and Vulnerability sub-indexes (socioeconomic, ecological and cultural contexts). The scores of each variable have been summed with the scope of obtaining an absolute value for each sub-index according to the follow equations: P Cfn nCf Coastal Forcing Index ¼ * 100 (1) nCf *n P

Vn nV * 100 nV*n

Where:

3.2. Coastal erosion risk assessment

Coastal Susceptibility Index ¼

ðCoastalForcingIndex*nCf Þ þ ðCoastal SusceptibilityIndex*nCSÞ nCf þ nCS (3)

In order to obtain an index in each context, the Coastal Erosion Risk Index has been calculated and expressed as a value based on the theo­ retical hazard and the coastal vulnerability. That value was normalized according to the index technique suggested by McLaughlin et al. (2002): Risk ¼

½Hazard*ðnCf þ nCSÞ�*½Vulnerability*ð nV*nÞ� ðnCf þ nCSÞ þ ðnV*nÞ

(5)

The total Coastal Erosion Risk Index is a single numerical value ob­ tained by means of a weighted average of all the calculated risks (Social, Ecological and Cultural), according to the number of variables included in each in order not to overestimate their individual weight. Once final indices were calculated, they were categorized by means of the natural breaks function analysis (Jenks and Caspal, 1971) into five classes of risk ranging from very low (1) to very high (5). 4. Results 4.1. The coastal erosion problem in Cartagena (magnitudes and causes) The analysis of coastal evolution trends for the study area revealed that 71.7% of Cartagena coastline is undergoing severe erosion problems (Table 6 and Fig. 2). Spatial and temporal variability of these results are related to a coastal heterogeneity that is affected by a diversity of factors that contribute to erosion behavior with different influences. The recent coastal evolution calculated for the Cartagena city is presented in Table 6 and Fig. 2. The erosion rates observed suggest the importance of human-induced processes. Disruptions of sediment supply are a primary factor contributing to observed high erosion rates, which are mainly related to ecosystem destruction such as mangroves and dunes (FAO, 2007; Aguirre-Rubi et al., 2018; Orejarena-Rondon et al., 2019) and emplacement of coastal engineering structures such as groins, breakwaters and seawalls (Stancheva et al., 2011; Rangel-Buitrago et al., 2011, 2015 and 2018a). The high erosion rates observed on this coast highlight that sediment supply to the coast is much less than the littoral transport capacity �ntara, 2005; Rangel-Buitrago et al., 2015; Orejar­ (Correa and Alca ena-Rondon et al., 2019). Sediment supplies derived from eroding cliffs, beaches, and dunes are insufficient to balance the sediment budget deficit. The morphology of the Magdalena river delta (the leading

(2)

Where:

Table 1 Forcing variables contributing to coastal erosion. COSTAL FORCING Variable

Null/Very Low (1)

Low (2)

Medium (3)

High (4)

Very High (5)

Significant wave height at a specific coastal sector (% of initial Hs) Rangel-Buitrago and Anfuso (2015) Storm Surge at a specific coastal sector (Rangel-Buitrago and Anfuso, 2015) Degree of littoral exposure to wave fronts (García Mora et al., 2001)

Less than 20%

20–40%

40–60%

60–80%

80–100%

Less than 20% 10–45� Oblique Macrotidal

20–40% x

40–60% 0–10� Sub-parallel Mesotidal

60–80% x

80–100% 0� Parallel Microtidal

Tidal Range (McLaughlin and Cooper, 2010)

4

x

x

N. Rangel-Buitrago et al.

Ocean and Coastal Management 186 (2020) 105099

Table 2 Coast Susceptibility Index for A) Sandy and B) Rocky shores. A.) COAST SUSCEPTIBILITY – SANDY Variable

Null/Very Low (1)

Low (2)

Medium (3)

High (4)

Very High (5)

Dune height (Gracia et al., 2009) Percentage of washovers (García Mora et al., 2001) Dry beach width as a multiple of the ICZ ( Anfuso et al., 2013) Beach slope/morphodynamic state, Foreshore slope (Anfuso, 2002) K Index (Aybulatov and Artyukhin, 1993)

�6 0%

�3 �5%

�2 �25%

�1 �50%

<1 �50%

5 times ICZ

4 times ICZ

3 times ICZ

2 times ICZ

Equal to ICZ

Dissipative (tan β � 0.02) Extreme (K > 1)

x

Intermediate (0.02 < tan β < 0.08) Average (K ¼ 0.11 � 0.5)

x

Reflective (tan β � 0.08)

Minimum (K ¼ 0.0001 � 0.1)

No structures (K ¼ 0)

Maximum (K ¼ 0.51 � 1)

B.) COAST SUSCEPTIBILITY – ROCKY Variable

Null/Very Low (1)

Low (2)

Medium (3)

High (4)

Very High (5)

Type (Sunamura, 1992)

Cliff with horizontal shore platform Granitic rocks, resistant metamorphic Virtual absence of discontinuities, cracks, joints, faults <30� 5 times ICZ

x

Cliff with sloping shore platform

x

Plunging cliff

limestone

Flysch, shale, Tertiary sedimentary rocks

Quaternary deposits

Volcanic ejecta

x

Some evidence of discontinuities, cracks, faults

x

High density of discontinuities, cracks, faults

31� �40� 4 times ICZ

41� �50� 3 times ICZ

51� �60� 2 times ICZ

>60� equal to ICZ

Unweathered Extreme (K > 1)

Slightly Unweathered Maximum (K ¼ 0.51 � 1)

Moderately weathered Average (K ¼ 0.11 � 0.5)

Highly weathered Minimum (K ¼ 0.0001 � 0.1)

Decomposed No structures (K ¼ 0)

Lithology (Sunamura, 1992) Structures (Bieniawski, 1989)

Slope (Anfuso et al., 2013) Cliff edge width as a multiple of the ICZ (Anfuso et al., 2013) Weathering (Bieniawski, 1989) K Index (Aybulatov and Artyukhin, 1993)

Table 3 Socioeconomic variables associated with the Vulnerability sub-index. SOCIO-ECONOMIC VULNERABILITY INDEX Variable

Null/Very Low (1)

Low (2)

Medium (3)

High (4)

Very High (5)

Land uses (CORINE Project)

Bushes and scrubs

Swamp area, Salt marsh, Coastal lagoon, Wet area, Gallery forest

Agricultural pond, Cropland Complex, cultivation area

Recreational structures, Airports, Industrial-Commercial area, Urban area, Mining area

Percentage of urbanized area (Li and Li, 2011) Population density (Li and Li, 2011)

Lower than 20%

Pastures (dense grass cover), Pastures (grass þ crop), Pastures (grass þ threes) 20 � 40%

40 � 60%

60 � 80%

Larger than 80%

Lower than 10 inhabitants per square kilometer Absent Absent

11 � 75

76 � 300

301 � 999

Greater than 1000 inhabitants per square kilometer

x Local

x National

x x

Present International

Less than 10

11–15

16–20

21–25

More than 25

Occasional

x

Seasonal

x

Full Time

Low Income

x

Medium income

x

High income

Roads (Drejza et al., 2019) Conservation designation ( Contreras and Kienberger, 2011) Number of infrastructure services (Cardona, 2007) Tourism (Rangel-Buitrago, 2019) Economic activities ( Rangel-Buitrago, 2019)

sediment supplier of the area) also has been subject to substantial change due to multiple interventions inside its basin such as damming, deforestation and canalization (Restrepo and Escobar, 2018; Restrepo et al., 2018). Another critical factor is the emplacement of coastal engineering structures. At the beginning of 2019, more than 300 hard structures (both cross and longshore - groins, concrete walls, breakwaters, among others) had been built along the Cartagena area. These data show an increase over that presented by Rangel-Buitrago et al. (2011) which observed 110 hard structures for the same area as of 2011. A high per­ centage (close to 90%) of these hard structures have had minimal

success, but instead have had adverse environmental impacts including interruption of coastal sediment-transport patterns, loss of beaches and other natural habitat, and a decrease in scenic ambiance. The global sea-level rise is another forcing factor in terms of coastal erosion, and general impact for managment (Cazenave, 2014; Geiser and Currents, 2017; Reimann et al., 2018; Zemp et al., 2019). In terms of the probable impact of sea-level rise on Cartagena, there is only a record database for the last 43 years as collected by the Global Sea Level Observing System (http://www.psmsl.org/data/obtaining/stations /572.php). According to Parker and Ollier (2017), this time interval is half the period necessary for a reliable estimation. However, as the only 5

N. Rangel-Buitrago et al.

Ocean and Coastal Management 186 (2020) 105099

Table 4 Ecological variables associated with the Vulnerability sub-index. ECOLOGICAL VULNERABILITY INDEX Variable

Null/Very Low (1)

Low (2)

Medium (3)

High (4)

Very High (5)

Protected area (IUCN, 2008) Ecosystem and habitat cover (Li and Li, 2011) Level of human intervention (Li and Li, 2011) Protected species (Gracia et al., 2018) Ecosystem services (Gracia et al., 2018) Litter presence ( Rangel-Buitrago, 2019) Non-built environment ( Rangel-Buitrago, 2019)

Strict Nature Reserve Unvegetated area

x x

x x

Very High (more than 80% of the area) 0

High (80 � 60%) 1–2

Natural Monument Bushes, stubble, grassland, bare rocks Medium (60 � 40%)

Habitat/species management area Strategic ecosystems: salt marsh, mangroves, marine seaweed, coral reef, lagoons Very low (Lower than 20%)

0–1 Continuous accumulations Field mixed cultivation

X

Low (40 � 20%) 3–5

x

2

x

more than 2

Full strand line x

Local or discontinuous accumulation Hedgerow/terracing/ monoculture

Few scattered items x

Virtually absent

More than 5

None

Table 5 Cultural variables associated with the Vulnerability sub-index. CULTURAL VULNERABILITY INDEX Variable

Null/Very Low (1)

Low (2)

Medium (3)

High (4)

Very High (5)

Cultural heritage (McLaughlin and Cooper, 2010) Ethnographic interest (Rangel-Buitrago, 2019) State of conservation (McLaughlin et al., 2002) National protection (Rangel-Buitrago, 2019) Ethnic communities (McLaughlin et al., 2002) Cultural built environment ( Rangel-Buitrago, 2019)

Absent

Local interest

Regional interest

0

1–2

x

National interest 3–5

International interest, UNESCO World Heritage Site More than 5

Poor

x

Moderate

x

Good

No

x

x

x

Yes

Absent

x

x

x

Present

Heavy industry

Heavy tourism/ urban

Light tourism/sensitive industry

Sensitive tourism

Historic

per year (Ortiz et al., 2013; Perez et al., 2018). These high-energy events generate severe damage, and an extended recovery period is necessary before beaches can return to equilibrium (Rangel-Buitrago and Anfuso, 2011) as the system requires higher sediment volumes and more extended periods for recovery (Lentz and Hapke, 2011). The understanding of climate change-driven impacts on the sea-level rise and extreme wave events are paramount to an effective coastal erosion management planning (Morim et al., 2019). Establishing robust projections of global sea-level rise and wave characteristics (by identi­ fying and assessing regions with lack of climate signal), and quantifying the uncertainties introduced by the complex modeling processes used for that purpose, is paramount to preventing potentially costly maladapta­ tion and wrong coastal erosion management.

Table 6 Coastal evolution trend categories along the Cartagena coastline. Type

Length (km)

Percentage %

High Erosion Erosion Stability Accumulation

20.0 3.4 2.3 6.9

61.3 10.4 7.1 21.3

Total

32.6

100

data set existing for the study area, it can serve to give an overall panorama of the situation. The analysis of monthly mean sea level data shows the relative mean sea level (MSL) trend at Cartagena is þ5.27 mm/year with a 95% confidence interval of �0.37 mm/year (a value of þ5.3 � 0.3 mm/year was calculated by Torres and Tsimplis, 2013). This apparent increment fits the global trend estimated by the IPCC (2014 and 2018), inferring an approximate elevation of 0.50 m in the next 100 years. Also, there is known subsidence along this coast (Andrade et al., 2017; Mora et al., 2018), but more data are needed to establish its real influence on the coastal erosion process. Despite such data limitations, it can be inferred that there has not been enough time for constructive forces to reach equilibrium along these beaches. Extreme wave events (e.g., hurricanes and cold fronts) also play a significant role in coastal erosion magnitudes observed in the study area (Rangel-Buitrago and Anfuso, 2015; Orejarena-Rondon et al., 2019). Hurricanes often originate in the Caribbean from June to November and affect coasts with high winds, heavy rains, and storm waves. Cold fronts have their origins during January–March period and cause strong swell waves whose impact may be increased by trade winds blowing from ENE, often striking the coast with an average occurrence of six events

4.2. Spatial distribution of hazard, vulnerability and risk 4.2.1. Hazard UNISDR (2009) defines Hazard as a dangerous phenomenon, sub­ stance, human activity or condition that may cause loss of life, injury or other health impacts, property damage, loss of livelihoods and services, social and economic disruption, or environmental damage. In this work, the Hazard Index is defined as a numerical value that indicates the po­ tential of an area to experience damages when it is subjected to coastal erosion, and it depends on two sub-indexes: Coastal Forcing and Susceptibility. Coastal Forcing is a numeric value that measures the energy level of all physical processes involved in coastal erosion. This sub-index mea­ sures the level of physical stress that any coastal segment could expe­ rience during erosion. Along the study area, 13.6 km of coastline (41.6%) have a low coastal forcing index while 11.9 km (36.5%) and 7.1 km (21.9%) have 6

N. Rangel-Buitrago et al.

Ocean and Coastal Management 186 (2020) 105099

sandy vs rocky shorelines). The coastal susceptibility index includes all factors that control littoral susceptibility to erosion in the function of the coast typology. The susceptibility value depends on the geological and geomorphological characteristics of the coast, and for assessment, it is necessary to differentiate the coast type (between sandy and rocky) because each typology has its own characteristics (Table 2), and these cannot be combined. 19.7 km (60.3%) of the total coastal length were classified inside the medium susceptibility class, while 11.3 km (34.7%) and 1.6 km (5%) belong to very high to high and low to very low classes, respectively (Fig. 3 and Table 7). These coastal susceptibility values are related to a coastline mainly composed of dissipative urban beaches, and diapiric Tertiary cliffs subjected to coastal erosion rates that sometimes exceed 1.5 m/yr (Rangel-Buitrago and Posada, 2013; Rangel-Buitrago et al., 2015 and 2018b). Cartagena is the perfect example of “induced susceptibility” in an area because of the developed numerous hard structures (i.e., groins, breakwaters, and seawalls) as the primary coastal erosion management strategy. These structures (Rangel-Buitrago et al., 2018c) are mainly responsible for the medium to high susceptibility values resulting from the loss of the original natural landforms. The current high level of coastal armoring (or hardening) not only is altering the natural susceptibility of the Cartagena coastline, but is also responsible for the formation of a narrow, swash aligned coastline with a typical “zig-zag” shape that is bringing adverse effects to coastal morphology that include: � Damage of the natural landforms � Irreversible coastline modifications � Interruption or reduction of sediment inputs (e.g., from eroding cliffs to adjacent beaches) � Increases downdrift erosion � Loss of sand material from beaches and shallow-water areas � Coastal squeeze. Fig. 2. Coastal evolution trend for the 1984–2019 period and erosion examples along Cartagena. Note that the designated High Erosion and Erosion areas are typically characterized by narrow to absent beaches, and typically shorehardening responses to the erosion.

Hazards arise when human life/health and/or property are placed or come to be in the path of an otherwise natural process. Coastal erosion processes usually are defined as hazards when the probability that the generated shoreline retreat will generate damage (specific or potential) in a given area. However, it is important to highlight that coastal erosion is a normal geological process, and only becomes a problem when there is no room to accommodate that change because urban development is too close to the sea, such as is happening along Cartagena. That then leads to “coastal squeeze.” Medium values of hazard accounting for 18.7 km (57.3%) of the coastline, and primarily distributed along the tourist sector (Fig. 3 and Table 7). High values accounted for 11.8 km (36.1%) of the total coastline, and occur in the Bocagrande, El Laguito and Tierrabomba sectors (Fig. 3 and Table 7). Low hazard values covered just 2.1 km (6.6%) of the coastline in areas of cliffs located north of Cartagena (Fig. 3 and Table 7). The 30.5 km of coastline with high and very high hazard index values has had significant erosion rates in past decades. An elevated percentage of areas with high and very high hazard index values, have erosion rates that exceeded 1.5 m/yr as were observed by Correa et al. (2005), Ran­ gel-Buitrago and Posada (2013), Rangel-Buitrago and Anfuso (2015), Rangel-Buitrago et al. (2018a) and Stronkhorst et al. (2018). This agreement confirms the validity of the methodology presented in this work, e.g. the higher the Hazard index value the greater the recorded erosion of a site, as observed in similar circumstances for temperate cliffs of Spain by Del Rio and Gracia (2009), coastal dunes of Brazil and Italy by Bertoni et al. (2019) and along archaeological heritage sites of Naples Gulf by Mattei et al. (2019). Coastal flooding is another process strongly linked with very high and high hazard index values. In Cartagena, medium and high hazard­ ous sectors are affected by high values of wave run-up recorded during

medium and high coastal forcing index values, respectively (Fig. 3 and Table 7). These values are related to specific factors that control how wave energy is distributed along Cartagena coastline. These factors mainly include local bathymetric conditions, the degree of littoral exposure to wave fronts and the presence of hard structures (mainly revetments and seawalls) that reflect the wave’s energy. Medium to high coastal forcing values measured along the study area are related to the degree of exposure of the coast, which is in some parts parallel the incoming wave fronts that mainly approach from the NE. The general NE-SW coastal orientation of Cartagena determines the level of exposure, and hence the prevalence of a longshore transport (towards SW) that is important in the coastal erosion process. Also, the rectilinear orientation in some specific segments of the coastline, plus homoge­ neous characteristics of the nearshore allow the arrival of high-energy waves, generating elevated storm surges. In contrast, low coastal forc­ ing index values along the area are related to sheltered zones. Tidal range and the sea-level rise always have been linked to coastal flooding and erosion hazards (Gornitz et al., 1994; Costas et al., 2015). In this work, the tidal range and sea-level rise were considered as vari­ ables with a constant value due to the short length of the study area (32.6 km). Cartagena which is a microtidal environment with a relative sea-level rise of 5.27 mm/year (as measured over the last 43 years) can be considered as having a medium to high level of physical stress when compared with meso or macrotidal environments. Coastal Susceptibility is a numeric value that reflects the level of exposure, and definitive intrinsic characteristics of the coastline (e.g., 7

N. Rangel-Buitrago et al.

Ocean and Coastal Management 186 (2020) 105099

Fig. 3. Forcing, Susceptibility and Hazard Indexes calculated for Cartagena city. Table 7 Distribution of forcing, susceptibility and hazard calculated for Cartagena. Type Forcing Susceptibility Hazard

Length (km) Percentage (%)

Very low

Low

Medium

High

Very high

Total

0.0 0.0 1.1 3.2 0.0 0.0

13.6 41.6 0.6 1.8 2.1 6.6

11.9 36.5 19.7 60.3 18.7 57.3

7.1 21.9 9.8 29.9 11.8 36.1

0.0 0.0 1.6 4.8 0.0 0.0

32.6 100 32.6 100 32.6 100

extreme wave conditions. Recent examples of this process were observed during September 2018 when the run-up related to extreme wave conditions generated during Hurricane Matthew reached values up to 1.5 m at the Cartagena beaches.

vulnerability indices calculated for Cartagena city are presented in Fig. 4 and Table 8. The socio-economic context of the vulnerability index has been constituted by a series of variables representing social, economic and human activities that, because of their intrinsic characteristics, may be negatively impacted by coastal erosion (Table 3). The socio-economic context reaches significant importance in any vulnerability assessment because the original concept of vulnerability always was linked to humans and society (Li and Li, 2011; De Serio et al., 2018; Kiat et al., 2019). Cartagena is the fifth-largest urban area in Colombia with a popu­ lation density of 1811.91 inhabitants per km2. In that sense, high up to very high values of the Socio-economic Vulnerability were observed along 22.2 km, or 68.1% of the total coastline (Fig. 4 and Table 8). Very low values of socio-economic vulnerability were found along low human intervention areas such as the northern part of Cartagena for 8.8 km of coastline (26.9%). Medium values of this index (1.6 km, 4.9%) were observed just in specific sites located in the north of the study area (Fig. 4 and Table 8). The distribution of the socio-economic vulnerability along Cartagena city suggests that variables such as population density, percentage of the urbanized area, and economic activities determine the degree of impacts associated with current coastal erosion. Such observations agree with McLaughlin et al. (2002) and Del Rio and Gracia (2009) which suggested

4.2.2. Vulnerability Vulnerability can be defined as the characteristics and circumstances of a community, system, or asset that make it susceptible to the damaging effects of a hazard (UNISDR, 2009). The vulnerability index calculated here corresponds with a value that denotes the ability of the coastal system to cope with and recover from an erosion event. This index allows the evaluation of the potential impacts of coastal erosion in a socioeconomic, ecological, and cultural framework. There are many aspects of vulnerability arising from various socioeconomic, ecological, cultural, and even physical factors. Examples may include: i) poor building design and/or construction, ii) inadequate protection of assets, iii) absence of public information and awareness, iv) limited recognition of hazards and preparedness measures, amongst others. Vulnerability can vary significantly within the same community and over time. The vulnerability definition used here identifies these aspects as a characteristic of the element of interest (the coast), which is inde­ pendent of its exposure to eroding forces. However, it is essential to highlight that in common use, the vulnerability concept is often used more broadly to describe the degree of exposure. The different 8

N. Rangel-Buitrago et al.

Ocean and Coastal Management 186 (2020) 105099

Fig. 4. Vulnerability Indexes calculated for Cartagena city. Table 8 Distribution of Vulnerability calculated for Cartagena. Type Socio-economic Ecological Cultural Total

Length (km) Percentage (%)

Very low

Low

Medium

High

Very high

Total

8.8 26.9 0.0 0.0 9.9 30.5 0.0 0.0

0.0 0.0 19.9 61.1 2.8 8.5 2.9 8.9

1.6 4.9 3.9 11.9 1.6 4.9 9.8 30.1

2.3 7.1 1.1 3.3 16.7 51.0 19.9 61.1

19.9 61.1 7.7 23.7 1.7 5.2 0.0 0.0

32.6 100 32.6 100 32.6 100 32.6 100

that the variability of land use type, percentage of urbanized area and economic activities are the most effective variables in the discrimination of impact levels, in this case from a socio-economic point of view. These same authors stated that the number of people living in a specific area is a significant issue in the analysis of any coastal risk. In addition, pop­ ulation density constitutes a critical factor in the calculation of the so­ cioeconomic vulnerability because of its relative nature that makes it more widely applicable than absolute population features. The ecological context is based on the conservation premise of natural habitats since these provide natural protection that reduces vulnerability to coastal erosion. Most of the existing vulnerability assessment methodologies only take into account the socio-economic aspect (human component). However, when coastal erosion strikes, all ecosystems located over the coastal environment can be affected, losing their quality, health, status, and conservation degree (Gracia et al., 2018). The integration of ecological variables inside of any vulnerability assessment represents a significant challenge (McLaughlin et al., 2002). The main question lies in deciding how to rank specific sites that sometimes lose their “natural status” because of human alteration. However, ‘protection’ of a conservation site can hardly include protec­ tion from the action of natural processes that formed a particular habitat (e.g., waves and corals). The final calculation of the ecological vulnerability showed values

inversely proportional to the socio-economic vulnerability. The results showed very high and high values of ecological vulnerability in 7.7 km (23.7%) and 1 km (3.3%) respectively. On the other hand, low values of ecological vulnerability were observed along 19.9 km (61.1%) while medium classes reached values of 3.8 km (11.9%) of the total area (Fig. 4 and Table 8). This dominance of low values of ecological vulnerability was because Cartagena does not have a high number of areas with any particular ecological significance or protection status because of the high degree of human intervention along the entire coastline. The cultural context is that aspect of vulnerability that places emphasis on protecting/conserving important cultural components such as archaeological, historical, heritage, scientific, and scenic sites. Un­ fortunately, coastal erosion is responsible for the destruction and loss of world archaeological and heritage sites (Hoogland and Hofman, 2015; Stancioff et al., 2018). Many examples exist where coastal erosion has affected many natural and cultural World Heritage properties (e.g., the Moais on Easter Island, Chile; Slave huts and obelisks in Bonaire; Cos­ tiera Amalfitana in Italy). In this paper the cultural context was taken into account using five variables that follow the UNESCO (1972) cultural heritage classifica­ tion, and that ranked each location according to its interest, from local to international. The international and national interest, linked to cultural and heritage importance, led to very high and high values of cultural 9

N. Rangel-Buitrago et al.

Ocean and Coastal Management 186 (2020) 105099

vulnerability for Cartagena city (18.3 km or 56.2%) as is shown in Fig. 4 and Table 8. Areas of interest were represented by the World Heritage Site called “Port, Fortresses, and Group of Monuments of Cartagena,” which are the most extensive fortifications in all South America. This famous example of the military architecture of the 16th, 17th and 18th centuries also was one of the most important ports of the Caribbean. The port of Cartagena – together with Havana, Cuba and San Juan, Puerto Rico – was an essential link in the route of the West Indies, and thus an essential chapter in the history of world exploration and the great commercial maritime routes. The combination of all the Vulnerability indexes calculated (socioeconomic, ecological, and cultural) gave a general scene of the vulner­ ability of Cartagena (Fig. 4 and Table 8). The total vulnerability index ranged from low to high values. High values were dominant accounting for 19.9 km (61%) of the coastline. Medium values accounted for 9.8 km (30.1%), while low values were observed along 2.9 km (8.9%).

and Table 9). Previously mentioned data showed a strong relation be­ tween wave energy, the degree of littoral exposure, coastal erosion, associated vulnerability, and the characteristics of human interventions. The ecological risk represented the probability of loss in areas with a certain level of ecological importance. Along the Cartagena area, me­ dium values are dominant and reach a percentage of 70.7%, equivalent to 23 km of coastline (Fig. 5 and Table 9). The Heritage risk assessment showed how the areas of high risk coincide with the world heritage area that encompasses 8.8 km (27.1%) of the total study area (Fig. 5 and Table 9). The combination of each one of the previous risk indexes is the basis for calculating the total risk for Cartagena city (Fig. 5 and Table 9). The results showed high values of risk in 8.8 km (27.1%) of the coastline. On the other hand, low values of risk were observed along 1 km (3.2%), while medium risk was the dominant trend along the area reaching values of 22.7 km (69.7%). 5. Discussion

4.2.3. Risk Coastal Risk can be defined as the combination of the probability of an event (e.g., erosion) and its negative consequences over the coastal environment. Any coastal risk assessment must include the two separate components that constitute the risk: hazard and the associated impact expressed through the vulnerability (Birkmann, 2007; Menoni and Margottini, 2011; Rangel-Buitrago and Anfuso, 2015). In this paper, hazard and vulnerability were combined into the Coastal Erosion Risk Index which is the combination of the probability of an event (coastal erosion) and its negative consequences in a socioeconomic, ecological and cultural contexts (Fig. 5 and Table 9). This index is a numerical value obtained using a weighted average of both indexes (hazard and vulnerability) according to the number of variables used to calculate them (Del Rio and Gracia, 2009; Rangel-Buitrago and Anfuso, 2015). From a socio-economic point of view, the coastal risk assessment showed that the most sensitive zones were located over urbanized areas of Cartagena that correspond with 61.5% (20 km) of the total area (Fig. 5

5.1. Results validation All methodological approaches to asset hazard, vulnerability, and risk should be tested and validated before being considered adequate to be used as management tool (Mcfadden, 2010; Micallef et al., 2018). Results obtained in this work have shown that 31.5 km (96.8%) of the Cartagena coastline is subjected to medium to high erosion risk. The methodologies proposed by McLaughlin (2002) and Rangel-Buitrago and Anfuso (2015) were used to test the obtained results by comparing coastal risk with the coastline evolution recorded during the 1984–2019 period. Also, a linear multiple regression method was used in each segment of coastline to evaluate the existing correlation between the calculated risk and coastline evolution, through the following expression: RI ¼ f(CE) Where RI is Risk Index and CE is Coastal Evolution.

Fig. 5. Risk Indexes calculated for Cartagena city. 10

(6)

N. Rangel-Buitrago et al.

Ocean and Coastal Management 186 (2020) 105099

Table 9 Distribution of Risk calculated for Cartagena. Type Socioeconomic

Length (km) Percentage (%)

Ecological Cultural Total

Very low

Low

Medium

High

Very high

Total

0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

3.2 9.8 1.1 3.4 3.7 11.4 1.1 3.2

9.4 28.8 23.1 70.7 20.1 61.5 22.7 69.7

20.1 61.5 8.5 26.0 8.8 27.1 8.8 27.1

0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

32.6 100.0 32.6 100.0 32.6 100.0 32.6 100.0

To corroborate the validation processes, coastal evolution (CE) was expressed using different tree methods: the Net Shoreline Movement (NSM), the End Point Rate (EPR) and the Linear Regression (LRR), ac­ cording to the method applied by Himmelstoss (2009). Results of this validation are presented in Table 10. Data shows optimal goodness for fit of the multiple regression model according to the coefficient of multiple determinations R, with around 86% of the variation in the RI being explained by the model. Here it is essential to highlight that correlation values will be high if the RI is dominated by the hazard (forcing and susceptibility). A low correlation value does not necessarily indicate a failure in the validation process. It can be interpreted as a more significant contribution of vulnerability to RI. According to Cooper and McLaughlin (1998) if the value of the Risk Index is not in sufficient agreement with the coastal erosion data, other important factors that are not included in the index maybe influence the coastline erosion. The above is not the case in this study confirming the validity of the used method and the absence of external, local factors that can affect the current coastal erosion process observed in the city of Cartagena. This kind of information can be easily used in its entirety or partially, i.e., at different levels and scales, by local planning staff or another variety of end-users according to specific goals. Working under a GIS environment allows the collection of a significant amount of different data and, one of its main advantages is the possibility of regularly update the initial information with new sets of data and then re-calculate the indexes. In this sense, all variables presented in this work, as well as new ones that want to be incorporated, can be easily integrated, spatially analyzed, and updated into a GIS environment to serve as a coastal erosion management tool. The methodological approach presented in this work gives an “instantaneous picture” analysis of coastal risk to coastal erosion related hazards for a given area. Significant advantages of the proposed method are the fact that it is based on recent coastal evolution and erosion processes effects as well as its flexibility and adaptability to the specific features of each area to be investigated. In this sense, it represents a primary and appropriate evaluation tool previous to the elaboration of any coastal erosion management plan.

(Brundtland, 1987). Such management requires the implementation of practical solutions based on real knowledge of magnitudes, distribution, trends, and causes of the coastal erosion process. Traditionally, the coastal erosion management approach has been concentrated on protection, using hard defenses (de Jonge, 2009; Pranzini and Williams, 2013; Pranzini et al., 2015; Williams et al., 2018). However, during the last decades, coastal erosion management has moved away from the “hard structure” mentality and now seeks to formulate adaptation strategies that begin with coastal risk mapping (ESCAP/UNISDR, 2012; Salik et al., 2015; Gracia et al., 2018). Coastal risk mapping is an essential tool in coastal management and planning. A well-developed map can contain essential information to assess the best coastal erosion management strategies to be used, and also can be used as technical and legal support to make decisions as well enact devel­ opment policies. Coastal mapping as an adaptation strategy can be encouraged by the use of existing Coastal Zone Management Laws. The incorporation of mapping into coastal laws will allow stakeholders to have a technical and legal basis for the best management available. Thus, regulations to limit or even ban further exploitation of erosion-prone coastal areas can be enacted, implemented, and supported based upon real information, and not in a subjective way. Currently, policies and actions are increasingly taking an integrated and holistic approach to address and counteract the increased coastal erosion impacts (UNEP, 2009; Jones and Phillips, 2011). In that sense, prioritizing coastal risk mapping in all coastal countries is urgent. As humans, we can get a sustainable global civilization if we start evalu­ ating the past from a broad perspective and placing greater emphasis on planning for the future by using mapping. World coastal countries, especially the most vulnerable nations, can manage coastal erosion with new and innovative strategies that must begin with the coastal risk mapping. Also, these countries must plan to replace old and less efficient management processes, particularly those that are known to generate more problems. Lastly, it is essential to highlight that coastal risk assessment and mapping also can be used as: � A tool to be incorporated into the spatial and temporal knowledge of an area. � Material to create indicators and indices. � A decision support tool for social and economic development. � An integrative system for all aspects involved (i.e., socio-economic, ecological, and cultural).

5.2. Some aspects to take into account Optimal coastal erosion management demands that uses of coastal resources “meet population needs without compromising the ability of eco­ systems and future generations to respond to their natural needs”

Considering the high-risk scores obtained in this work, the following aspects must be incorporated in the management and coastal planning of Cartagena city: Defining and delimiting of a buffer zone: A buffer zone in Carta­ gena city must be defined taking into consideration the dominant coastal erosion hazard. Areas of current high risk must receive special attention. Substantial restrictions for the development of new settlements must be applied along these areas. Creation of a coastal risk database: A database should be

Table 10 Results of the linear multiple regression analysis performed in order to validate the Risk Index. Net Shoreline Movement (NSM), End Point Rate (EPR) and Linear Regression (LRR). RI ¼ f (NSM) Multiple R Multiple R2 Adjusted R2

RI ¼ f (EPR) 0.75 0.73 0.70

Multiple R Multiple R2 Adjusted R2

RI ¼ f (LRR) 0.86 0.79 0.77

Multiple R Multiple R2 Adjusted R2

0.80 0.77 0.69

11

N. Rangel-Buitrago et al.

Ocean and Coastal Management 186 (2020) 105099

established and maintained. Also, all risk management must be devel­ oped in view of the economic, ecological, and cultural components which will help in the achievement of an optimal coastal protection policy. Building codes update: Due to high coastal erosion risk, buildings should be adapted or modified. Buildings with low foundations in the tourist zone experience frequent inundation on the lower floors, including the basement and first floor. Hence another option is to restrict how such lower floors are used in order to decrease the loss of assets. Incorporate the use of ecosystems for coastal erosion manage­ ment: This is an integrated process to conserve and improve ecosystem health that sustains some services for human well-being. This approach combines two particular actions: i) Develop or maintain ecosystems quality and ii) Ensure the best way for delivery of different ecosystem services to human well-being.

Forecasting and Early Warning Systems: Long-term weather forecasting is necessary along the study area. This will provide some guidance in focusing on where coastal erosion events may occur in the future. Also, it allows both managers and the public to be warned so that optimal decisions can be taken to reduce the adverse effects of erosion. As such, the primary goal of forecasting and early warning systems is to reduce vulnerability understanding the coastal forcing. Mapping also can serve as a baseline for the implementation of a managed/planned retreat approach. This coastal erosion management strategy has as a primary objective of reducing population, property, and infrastructure at risk through the planned withdrawal from coastal hazard zones, including erosion-prone areas that are defined after a risk assessment (Neal et al., 2018; Rangel-Buitrago and Neal, 2019). Specific management considerations are presented in Fig. 6. Con­ cerning coastal erosion hazard, managed retreat must implement

Fig. 6. Specific considerations to take into account along Cartagena city area. 12

N. Rangel-Buitrago et al.

Ocean and Coastal Management 186 (2020) 105099

mitigation tools designed to move existing and proposed development out of the path of both short- and long-term coastal erosion. Factors yet to be determined, such as the rate of sea-level rise for the Cartagena coast, must be included in management plans. The managed retreat approach may include managed realignment where shore-protection structures are removed selectively to allow natural coastal environ­ ments to be re-established. All the above are determined by means risk assessment and mapping. Here the approach is based on a simple phi­ losophy: know the problem, determine location, get out, avoid damage, and recognize that coastal zone dynamics should dictate the type of management to be employed.

increasing of the complexity of the index, so in any case, a balance should be found among applicability, scientific validity, and facility of use. Acknowledgements This work is a contribution to research groups: “Geology, Geophysics �ntico (Colombia), and Marine - Coastal Process”, Universidad del Atla “Department of Geology”, Grand Valley State University (USA) and The University of Hull (United Kingdom). References

6. Conclusions

Addo, K., Lamptey, E., 2013. Innovative technique of predicting shoreline change in developing countries: case of Accra erosion and causal factors. In: Finkl, C. (Ed.), Coastal Hazards. Springer, New York, pp. 367–402. Aguirre-Rubi, J., Luna-Acosta, A., Zarragoitia, M., Zaldibar, B., Izagirre, U., Marigomez, I., 2018. Assessment of ecosystem health disturbance in mangrove-lined Caribbean coastal systems using the oyster Crassostrea rhizophorae as sentinel species. Sci. Total Environ. 618, 718–735. Alberico, I., Amato, V., Aucelli, P., D’Argenio, B., Di Paola, G., Pappone, G., 2012. Historical shoreline change of the sele plain (southern Italy): the 1870–2009 time window. J. Coast. Res. 28 (6), 1638–1647. Andrade, C., Ferrero-Ronquillo, A., Leon, H., Mora-Paez, H., Carvajal-Perico, H., 2017. Sobre cambios en la línea de costa entre 1735 y 2011 y la subsidencia en la Bahía de Cartagena de Indias, Colombia. Rev. Acad. Colomb. Cienc. Ex Fis. Nat. 41 (158), 94–106. Anfuso, G., 2002. Morfologia y dinamica sedimentaria del litoral gaditano entre Chipiona y Rota. Servicio Publicaciones. Universidad de Cadiz, Cadiz, Spain. Anfuso, G., Gracia, F.J., Battocletti, G., 2013. Determination of cliffed coastline sensitivity and associated risk for human structures: a methodological approach. J. Coastal Res. 29 (6), 1292–1296. Aybulatov, N.A., Artyukhin, Y.V., 1993. Geo-ecology of the World Ocean’s Shelf and Coasts. Hydrometeo Publishing, Leningrad. Barragan, J.M., Andreis, M., 2015. Analysis and trends of the world’s coastal cities and agglomerations. Ocean Coast Manag. 114, 11–20. Bell, P.S., 1999. Shallow water bathymetry derived from an analysis of X-band marine radar images of waves. Coast. Eng. 37, 513–527. Benassai, G., Chirico, F., Corsini, S., 2009. Una metodologia per la definizione del rischio da inondazionecostiera. Studi Costieri 16, 51–72. Bertoni, D., Sarti, G., Alquini, F., Ciccarelli, D., 2019. Implementing a coastal dune vulnerability index (CDVI) to support coastal management in different settings (Brazil and Italy). Ocean Coast Manag. 180, 104916. Bieniawski, Z.T., 1989. Engineering Rock Mass Classifications. Wiley, New York. Birkmann, J., 2007. Risk and vulnerability indicators at different scales: applicability, usefulness and policy implications. Environ. Hazards 7 (1), 20–31. Blanco, J., Narvaez Barandica, J., Viloria, E., 2006. ENSO and interannual salinity changes in the Ci�enaga Grande de Santa Marta coastal lagoon system, colombian Caribbean. Estuar. Coast Shelf Sci. 66, 157–167. Boak, E., Turner, I., 2005. Shoreline definition and detection: a review. J. Coast. Res. 21 (4), 688–703. Bonetti, J., Woodroffe, C., 2017. Spatial analysis for coastal vulnerability assessment. In: Bartlett, D., Celliers, L. (Eds.), Geoinformatics for Marine and Coastal Management. CRC Press, Boca Raton, pp. 367–395. British Geological Survey, 2012. Coastal Erosion. UK Geohazard Note. NERC, pp. 1–5. Brundtland, H., 1987. Report of the World Commission on Environment and Development “Our Common Future” (United Nations, New York). Bush, D.M., Pilkey, O.H., Neal, W.J., 1996. Living by the Rules of the Sea. Duke University Press, Durham. Cardona, O.D., 2007. Indicators of Disaster Risk and Risk Management: Program for Latin America and the Caribbean. Inter-American Development Bank, Washington. Carter, T.R., Parry, M.L., Nishioka, S., Harasawa, H., 1994. Technical Guidelines for Assessing Climate Change Impacts and Adaptation. University College and Centre for Global Environmental Research, Tskuba. Cazenave, A., 2014. Sea level rise and its coastal impacts. Coast Future 2 (2), 15–34. Contreras, D., Kienberger, S., 2011. D.4.2 Handbook of Vulnerability Assessment in Europe MOVE, Methods for the Improvement of Vulnerability Assessment in Europe. European Commission of Environment, Salzburg. Cooper, J.A.G., McLaughlin, S., 1998. Contemporary multidisciplinary approaches to coastal classification and environmental risk analysis. J. Coast. Res. 14, 512–524. Cooper, J.A.G., Pilkey, O., 2012. Pitfalls of Shoreline Stabilization. Springer, Berlin. Cooper, J.A.G., Jackson, D.T.W., 2019. Coasts in Peril? A shoreline health perspective. Front. Earth Sci. 7, 260. Correa, I.D., Alc� antara, J., 2005. Historical and recent shore erosion along the Colombian Caribbean coast. J. Coast. Res. 49, 52–57. Correa, I.D., Rios, A., Gonzales, D., Toro, M., Ojeda, G., Restrepo, I., 2005. Erosi� on litoral entre Arboletes y Punta San Bernardo, Costa Caribe Colombiana. Bol. Geol. 29 (2), 117–128. Costas, S., Ferreira, O., Martinez, G., 2015. Why do we decide to live with risk at the coast? Ocean Coast Manag. 118 (A), 1–11. Crowell, M., Honeycutt, M., Hatheway, D., 1999. Coastal erosion hazards study: phase one mapping. J. Coast. Res. 28, 10–20.

Coastal erosion has substantial negative impacts on human activities and structures as well as on social and political concerns of human life, ecology and culture. Further, even if coastal erosion is due to natural processes, the associated impacts are increased by human interventions/ actions such as poor or misguided management practices, and inap­ propriate or non-existent coastal planning. Coastal erosion can produce critical economic losses. In Cartagena and Colombia, coastal erosion is responsible for significant financial losses annually in terms of coastal property loss, including damage to structures and loss of land. To miti­ gate coastal erosion, the local and national government spends millions of dollars every year on ineffective hard structures that contribute to the problem. In addition to beach erosion, several coastal ecosystems are damaged and lost annually. Exacerbation of existing coastal erosion hazards by climate change will create new and broader scale risks, not previously experienced in the Cartagena coastal area. CEM, particularly through applied risk assessment analyses, has implications for making decisions today, especially for those activities and assets that have long lifetimes in the coastal area, and that will be affected by climate change-related impacts. Also, responses to the more intense impacts will affect broader community values at the coast, such as public amenities, recreation, tourism, coastal habitats, and other environmental aspects. Adverse ef­ fects can thus arise from both the hazards themselves, as well as human responses to the hazards. The capacity to adapt will be different for different groups in the community, according to their vulnerability. While the need to protect and preserve the functioning of natural ecosystems is one of the aims of spatial planning, coastal planning also attempts to improve the economic and social well-being of coastal zones, and to help them develop their full potential for their human communities. In this sense, coastal erosion must be addressed on a proactive basis, in order to be able to adapt to and minimize the expected risks of erosion. A proactive approach in this context refers to knowledge of the process as a starting point for a policy of anticipating risk by imple­ menting spatial planning and technical measures for coastline management. The methodological approach presented in this paper provides a straightforward and uniform erosion hazard identification method that can be used for this framework in Colombia, and could also be applied to steer future coastal developments at erosion hotspots in other areas. Data presented in this work underline the importance and interest of further research on such topics in order to fully understand the natural process and evaluate coastal erosion hazard. Since the goal of this study was to use a general methodological approach accurately applicable at different areas, the evaluation of the coastal erosion risk was based on an objective and quantitative methodology to remove uncertainty and subjectivity. Additional enhancements in the development of the indices can be made, for example, by including more variables in the assessment of hazard, vulnerability, and risk related to coastal erosion. The selection and addition of different variables must avoid redundancy and ambi­ guity. Further, the increase in the number of variables requires the 13

N. Rangel-Buitrago et al.

Ocean and Coastal Management 186 (2020) 105099

De Andr�es, M., Barragan, J.M., Garcia Sanabria, J., 2018. Ecosystem services and urban development in coastal Social-Ecological Systems: the Bay of C� adiz case study. Ocean Coast Manag. 154, 155–167. de Jonge, V., 2009. From a defensive to and integrated approach. In: Reinhard, S., Folmer, H. (Eds.), Water Policy in the Netherlands: Integrated Management in a Densely Populated Delta. Earthscan, Washington, pp. 17–46. de Ruig, J., 1998. Coastline management in The Netherlands: human use versus natural dynamics. J. Coast. Conserv. Manag. 4, 127–134. De Serio, F., Armenio, E., Mossa, M., Petrillo, A., 2018. How to define priorities in coastal vulnerability assessment. Geosciences 8 (11), 415–435. Del Rio, L., Gracia, F., 2009. Erosion risk assessment of active coastal cliffs in temperate environments. Geomorphology 112, 82–95. Di Paola, G., Iglesias, J., Rodriguez, G., Benassai, G., Aucelli, P., Pappone, G., 2011. Estimating coastal vulnerability in a meso-tidal beach by means of quantitative and semi-quantitative methodologies. J. Coast. Res. 61, 303–308. Drejza, S., Bernatchez, P., Marie, G., Friesinger, S., 2019. Quantifying road vulnerability to coastal hazards: development of a synthetic index. Ocean Coast Manag. 181, 104894. ESCAP/UNISDR, 2012. Reducing vulnerability and exposure to disasters: the Asia-Pacific disaster report. ESCAP/UNISDR, Bangkok. FAO, 2007. The Role of Coastal Forests and Trees in Protecting against Coastal Erosion. Food and Agriculture Organization of the United Nations, Khao Lak. García Mora, M.R., Gallego, J.B., Williams, A.T., García Novo, F., 2001. A coastal dune vulnerability classification. A case study of the SW Iberian Peninsula. J. Coast. Res. 17 (4), 802–811. Geiser, C., Currents, B., 2017. Impediments to inland resettlement under conditions of accelerated sea level rise. Land Use Policy 66, 322–330. Gerkema, T., Duran-Matute, M., 2017. Interannual variability of mean sea level and its sensitivity to wind climate in an inter-tidal basin. Earth Syst. Dynam. 8, 1223–1235. Gornitz, V., 1991. Global coastal hazards from future sea level rise. Palaeogeogr Palaeocl 89, 379–398. Gornitz, V.M., Daniels, R.C., White, T.W., Birdwell, K.R., 1994. The development of a coastal risk assessment database: vulnerability to sea-level rise in the U.S. Southeast. J. Coast. Res. 12, 327–338. Gracia, F.J., Sanjaume, E., Hern� andez, L., Hern� andez, A.I., Flor, G., G� omez-Serrano, M. A., 2009. Dunas marítimas y continentals, Bases ecol� ogicas preliminares para la conservaci� on de los tipos de h� abitat de inter� es comunitario en Espa~ na. Ministerio de Medio Ambiente y Medio Rural y Marino, Madrid. Gracia, A., Rangel-Buitrago, N., Oakley, J., Williams, A.T.W., 2018. Use of ecosystems in coastal erosion management. Ocean Coast Manag. 156, 277–289. Himmelstoss, E.A., 2009. DSAS 4.0 installation instructions and user guide. In: Thieler, E. R., Himmelstoss, E.A., Zichichi, J.L. (Eds.), Digital Shoreline Analysis System (DSAS) Version 4.0 - an ArcGIS Extension for Calculating Shoreline Change. U.S. Geological Survey, Charleston. Hoogland, M.L.P., Hofman, C.L., 2015. Archaeological assessment in compliance with the Valetta Treaty. Spanish water, Curacao. In: Hofman, C.L., Haviser, J.B. (Eds.), Managing Our Past into the Future: Archaeological Heritage Management in the Dutch Caribbean. Sidestone Press, Leiden, pp. 183–194. IPCC (Intergovernmental Panel on Climate Change), 1990. Climate Change, the IPCC Scientific Assessment. Cambridge University Press, New York. IPCC (Intergovernmental Panel on Climate Change), 2014. Climate Change 2014: the Physical Science Basis. Cambridge University Press, New York. IPCC (Intergovernmental Panel on Climate Change), 2018. Climate Change 2018: the Physical Science Basis. Cambridge University Press, New York. IUCN, 2008. red list - State of the World’s species. IUCN, Gland. Jenks, G.F., Caspal, F.C., 1971. Error on choropletic maps: definition, measurement, reduction. Ann. Assoc. Am. Geogr. 61 (2), 217–244. Jones, A., Phillips, M., 2011. Disappearing Destinations. CABI, London. Kiat, Ng, Borges, P., Phillips, M., Medeiros, A., Calado, H., 2019. An integrated coastal vulnerability approach to small islands: the Azores case. Sci. Total Environ. 481, 142–156. Lentz, E.E., Hapke, C.J., 2011. Geologic framework influences on the geomorphology of an anthropogenically modified barrier island: assessment of dune/beach changes at Fire Island, New York. Geomorphology 126, 82–96. Li, K., Li, G.S., 2011. Vulnerability assessment of storm surges in the coastal area of Guangdong Province. Nat. Hazards Earth Syst. Sci. 11, 2003–201. Martinez, C., Contreras-Lopez, M., Winckler, P., Hidalgo, H., Agredano, H., 2018. Coastal erosion in Central Chile: a new hazard? Ocean Coast Manag. 156, 141–155. Mattei, G., Rizzo, A., Anfuso, G., Aucelli, P., Gracia, J., 2019. A tool for evaluating the archaeological heritage vulnerability to coastal processes: the case study of Naples Gulf (southern Italy). Ocean Coast Manag. 179, 104876. Mcfadden, L., 2010. Coastal Hazards and Vulnerability. Earthscan, London. McLaughlin, S., 2002. Assessment and Development of a Coastal Vulnerability Index for Northern Ireland Employing GIS Tecniques. Unpublished Ph.D. Thesis. University of Ulster, Ireland. McLaughlin, S., Cooper, J.A.G., 2010. A multi-scale coastal vulnerability index: a tool for coastal managers? Environ. Hazards 9 (3), 233–248. McLaughlin, S., McKenna, J., Cooper, J.A.G., 2002. Socio-economic data in coastal vulnerability indices: constraints and opportunities. J. Coast. Res. 36, 487–497. Menoni, S., Margottini, C., 2011. Inside Risk: a Strategy for Sustainable Risk Mitigation. Springer-Verlag Italia Srl, Milano. Micallef, S., Micallef, A., Galdies, C., 2018. Application of the coastal hazard wheel to assess erosion on the Maltese coast. Ocean Coast Manag. 156, 141–155. Molares-Babra, R., Mestres-Ridge, M., 2012. Efectos de la descarga estacional del Canal del Dique en el mecanismo de intercambio de aguas de una bahía semicerrada y micromareal: Bahía de Cartagena, Colombia. Bol. Cientifico CIOH 30, 53–74.

Mora, H., Carvajal, J., Ferrero, A., Leon, H., Andrade, C., 2018. On natural gas venting and preliminary results on subsidence in Cartagena bay (Colombia). Bol. Cientifico CIOH 38, 35–51. Morim, J., Hemer, M., Wang, X.L., et al., 2019. Robustness and uncertainties in global multivariate wind-wave climate projections. Nat. Clim. Chang. 9, 711–718. Neal, W.J., Pilkey, O.H., Kelley, J.T., 2007. Atlantic Coast Beaches: A Guide to Ripples, Dunes, and Other Natural Features of the Seashore. Mountain Press, Missoula. Neal, W.J., Pilkey, O.H., Cooper, J.A.G., Longo, N., 2018. Why coastal regulations fail. Ocean Coast Manag. 156, 21–34. Ng, K., Phillips, M.R., Borges, P., Thomas, T., August, P., Calado, H., Veloso-Gomes, F., 2014. Maintaining a way of life for S~ ao Miguel Island (the Azores archipelago, Portugal): an assessment of coastal processes and protection. Sci. Total Environ. 481, 142–156. Ortiz, J.C., Otero, L.J., Restrepo, J.C., Ruiz, J., Cadena, M., 2013. Cold fronts in the Colombian Caribbean Sea and their relationship to extreme wave events. Nat. Hazards 13, 2797–2804. Orejarena-Rondon, A.F., Sayol, J., Marcos, M., Otero, L., Restrepo, J.C., HernandezCarrasco, I., Orfila, A., 2019. Coastal impacts driven by Sea-level rise in Cartagena de Indias. Front. Earth Sci. 6, 614. Parker, A., Ollier, C.D., 2017. Short-term tide gauge records from one location are inadequate to infer global sea-level acceleration. Earth Syst. Environ. 1 (2), 1–17. Perez, A., Ortiz, J., Bejarano, L., Otero, L., Restrepo, J., Franco, A., 2018. Sea breeze in the Colombian Caribbean coast. Atm� osfera 31 (4), 389–406. Pilkey, O.H., Cooper, A.G., 2014. The Last Beach. Duke University Press, Durham. Pilkey, O., Neal, W.J., Kelley, J., Cooper, A., 2011. The World’s Beaches: a Global Guide to the Science of the Shoreline. University of California Press, Berkley. Pilkey, O., Pilkey, O., 2019. Sea Level Rise: A Slow Tsunami on America’s Shores. Duke University Press, Durham. Pranzini, E., Williams, A.T., 2013. Coastal Erosion and Protection in Europe. Routledge/ Earthscan, London. Pranzini, E., Wetzel, L., Williams, A.T., 2015. Aspects of coastal erosion and protection in Europe. J. Coast. Conserv. Manag. 19 (4), 445–459. Rangel-Buitrago, N., Anfuso, G., 2011. An application of Dolan and Davis (1992) classification to coastal storms in SW Spanish littoral. J. Coast. Res. 64, 1891–1895. Rangel-Buitrago, N., Stancheva, M., Anfuso, G., 2011. Effects of coastal armouring in the Bolivar department (Caribbean sea of Colombia). Prob. Geog. 1–2, 97–106. Rangel-Buitrago, N., Posada, B., 2013. Determinaci� on de la vulnerabilidad y el riesgo costero mediante la aplicaci� on de herramientas SIG y m�etodos multicriterio en la línea de costa. Caribe Colombiano. Intropica 8 (1), 31–44. Rangel-Buitrago, N., Anfuso, G., 2015. Risk Assessment of Storms in Coastal Zones: Case Studies from Cartagena (Colombia) and Cadiz (Spain). Springer, New York. Rangel-Buitrago, N., Anfuso, G., Williams, A.T., 2015. Coastal erosion along the Caribbean coast of Colombia: magnitudes, causes and management. Ocean Coast Manag. 114, 129–144. Rangel-Buitrago, N., Neal, W.J., 2019. Coastal erosion management. In: Finkl, C.W., Makowski, C. (Eds.), Encyclopedia of Coastal Science. Springer Nature, Switzerland. Rangel-Buitrago, N., Contreras, M., Martinez, C., Williams, A.T.W., 2018a. Can coastal scenery be managed? The Valparaiso region, Chile as a case study. Ocean Coast Manag. 163, 383–400. Rangel-Buitrago, N., De Jonge, V., Neal, W.J., 2018b. How to make integrated coastal erosion management a reality. Ocean Coast Manag. 156, 290–299. Rangel-Buitrago, N., Williams, A., Anfuso, G., 2018c. Hard protection structures as a principal coastal erosion management strategy along the Caribbean coast of Colombia. A chronicle of pitfalls. Ocean Coast Manag. 156, 58–75. Rangel-Buitrago, N., 2019. Coastal Scenery: Evaluation and Management. Springer, Amsterdam. Reimann, L., Vafeidis, A., Brown, S., Hinkel, J., Tol, R., 2018. Mediterranean UNESCO World Heritage at risk from coastal flooding and erosion due to sea-level rise. Nat. Commun. 9, 4161. Restrepo, J., Escobar, H., 2018. Sediment load trends in the Magdalena River basin (1980–2010): anthropogenic and climate-induced causes. Geomorphology 302, 76–91. Restrepo, J., L� opez, 2008. Morphodynamics of the Pacific and Caribbean deltas of Colombia, South America. J. South Am. Earth Sci. 25 (1), 1–21. Restrepo, J., Escobar, R., Tosic, M., 2018. Fluvial fluxes from the Magdalena River into Cartagena Bay, Caribbean Colombia: trends, future scenarios, and connections with upstream human impacts. Geomorphology 302, 92–105. Rizzo, A., Aucelli, P., Gracia, J., Anfuso, G., 2018. A novelty coastal susceptibility assessment method: application to Valdelagrana area (SW Spain). J. Coast. Conserv. 22 (5), 973–987. Salik, K.M., Jahangir, S., Wu, Z., Su, H., 2015. Climate change vulnerability and adaptation options for the coastal communities of Pakistan. Ocean Coast Manag. 112, 61–73. Silvera, Y., Bonetti, J., 2019. Assessment of the physical vulnerability to erosion and flooding in a sheltered coastal sector: Florian� opolis Bay, Brazil. J. Coast. Conserv. 23 (2), 303–314. Skilodimou, H.D., Bathrellos, G.D., Chousianitis, K., 2019. Multi-hazard assessment modeling via multi-criteria analysis and GIS: a case study. Environ. Earth Sci. 78, 47. Small, C., Nicholls, R.J., 2003. A global analysis of human settlement in coastal zones. J. Coast. Res. 19, 584–599. Stanchev, H., Stancheva, M., Young, R., Palazov, A., 2018. Analysis of shoreline changes and cliff retreat to support marine spatial planning in shabla municipality, Northeast Bulgaria. Ocean Coast Manag. 165, 127–140. Stancheva, M., Rangel-Buitrago, N., Anfuso, G., Palazov, A., Stanchev, H., Correa, I., 2011. Expanding level of coastal armouring: case studies from different countries. J. Coast. Res. 64, 1815–1819.

14

N. Rangel-Buitrago et al.

Ocean and Coastal Management 186 (2020) 105099 UNESCO, 1972. Convention Concerning the Protection of the World Cultural and Natural Heritage. UNESCO World Heritage Centre, Paris. UNISDR, 2009. UNISDR Terminology on Risk Reduction. ISDR, Paris. van Heuvel, T., Hillen, R., 1994. Coastline Management with GIS in the Netherlands. EARSeL, Delft. Veersalu, T., 2011. The Best Practices of Coastal Zone Protection and Conservation in Spatial Planning. Technical Report, Baltic Greenbelt. Viles, H., Spence, T., 1995. Coastal Problems: Geomorphology, Ecology, and Society. Edward Arnold, London. Williams, A.T.W., Rangel-Buitrago, N., Pranzini, N., Anfuso, G., 2018. The management of coastal erosion. Ocean Coast Manag. 165, 156–169. Zemp, M., Huss, M., Thibert, N., Eckert, N., Mcnabb, R., Huber, J., Barandun, M., Machguth, H., 2019. Global glacier mass changes and their contributions to sea-level rise from 1961 to 2016. ON Nat. 568, 382–386.

Stancioff, C.E., Vermeer, J., Mukhopadhyay, A., Ruiter, S., Hofman, C.L., 2018. Predicting coastal erosion in St. Kitts: collaborating for nature and culture. Ocean Coast Manag. 165, 156–169. Stronkhorst, J., Levering, A., Hendriksen, G., Rangel-Buitrago, N., Rosendahl, L., 2018. Regional coastal erosion assessment based on global open access data: a case study for Colombia. J. Coast. Conserv. 22 (4), 787. Sunamura, T., 1992. Geomorphology of Rocky Coasts. John Wiley & Sons, New York. Torres, R., Tsimplis, M., 2013. Sea-level trends and interannual variability in the Caribbean Sea. JRG Oceans 118 (6), 2934–2947. Thieler, E.R., Himmelstoss, E.A., Zichichi, J.L., Miller, T.L., 2005. Digital Shoreline Analysis System (DSAS) version 3.0 - an ArcGIS extension for calculating shoreline change. US Geological Survey, Woods Hole. UNEP, 2009. The Role of Ecosystem Management in Climate Change Adaptation and Disaster Risk Reduction. UNEP, Copenhagen.

15