Open access peer-reviewed chapter

Identification of Karst Forms Using LiDAR Technology: Cozumel Island, Mexico

Written By

Oscar Frausto-Martínez, Norma Angelica Zapi-Salazar and Orlando Colin-Olivares

Submitted: 01 March 2018 Reviewed: 29 May 2018 Published: 05 November 2018

DOI: 10.5772/intechopen.79196

Chapter metrics overview

1,077 Chapter Downloads

View Full Metrics


Morphological relief analysis allows the identification of geomorphological forms and cartographic-environmental studies make extensive use of the medium (1:50,000) and large scale (1:250,000), where the topographical contrast is evident. However, at a detailed scale (<1:20,000) and for territories where the contrast of relief does not exceed 10 m in height, the morphological analyses must be adapted accordingly, because they contribute information to altimetry studies and to the topographic configuration of units. Thus, through visual interpretation and manipulation of high-resolution topographical LiDAR data from Cozumel Island, a relief analysis is presented at a detailed scale for the purpose of recognizing the geomorphological units of karst origin, using altimetry and slope cartography, digital models of elevation, and shading that permits the identification of 109 new exokarstic doline and uvala formations.


  • GIS
  • karst
  • relief
  • island
  • Caribbean

1. Introduction

Geologically, the Island of Cozumel’s basement is calcareous from the upper Tertiary period and reef from the Holocene period, which, accompanied by tropical climatic elements, give rise to dissolution karst formations. The study uses records of uvalas and cenotes in the area of study, which were obtained through a record from touristic service providers and from the community, as well as from scientific articles [1, 2]; however, there is no updated and complete database. For the purpose of contributing to the identification of karst forms (dolines, uvalas, and poljes‑locally called “cenotes” and “rejolladas”), altitude mapping and digital models of elevation, shading, and slope generated with LiDAR (light detection and ranging) data were used for identification and cartography.

The word LiDAR is an acronym for the term LIght Detection And Ranging, that is to say, detection and measurement of light. This technique is currently becoming a basic tool in studies based on topographical analysis and precision of the information base [3, 4, 5]. The use of LiDAR products has greatly improved and has a significant influence in the earth science disciplines [6, 7]. Its usefulness is emergent in shallow reliefs and those with little altitudinal difference.

The Yucatan Peninsula has an area of 39,340 km2 and is located in southeast Mexico’s. The most outstanding structural features of Yucatán are the sinkhole region and the aligned islands of Cozumel and Mujeres. The altitude of Yucatán not exceeding 300 m elevation dominates. Previous studies have recognized that different types of karstic depressions abound in the vast plateaus of the northern and eastern Yucatan Peninsula, and there are also extensive systems of caves and caverns in the entire landscape [8]. Climatic subtypes are warm and humid with summer rains and warm and humid with summer and winter rains [9]. Cozumel is part of the Yucatan peninsula with sedimentary rocks formed on a wide platform. The core drillings indicate that the island is formed from reef sediment with a thickness of 100 m or more, which dates from the Oligocene and Quaternary Epochs [10].

Cozumel Island pertains to the state of Quintana Roo, located at 20° 28′ N, 86° 55′ W (Figure 1). Cozumel Island faces the coast of the Yucatan Peninsula in the Caribbean Sea, approximately 16.5 km to the east of the Yucatan Peninsula, in the zone of the Northwest Caribbean [11]. The climate is hot and humid with abundant rainfall in the summer. The average annual temperature is 25.5°C and precipitation reaches 1504 mm per year. Cyclones have an important effect, increasing the amount of rainfall in the summer [12].

Figure 1.

Location of Cozumel Island in Mexico.

The relief of the island does not exceed 15 m of altitude above the sea level and the cartographic representations have been developed with data at 1:75,000 and 1:50,000 scale [1, 13]. The presence of dolines and uvalas with exokarstic forms has been reported, but the majority have been characterized over time, due to the fact that they do not exceed 50 m in diameter [14, 15, 16], which makes it necessary to recognize other techniques that aid the identification of exokarstic forms. For that reason, this study employs LiDAR modeling for the purpose of recognizing doline and uvala forms, enriching the list of these relief forms, and indicating the areas with higher density of exokarstic forms.


2. Methodology

To achieve the objectives, the following were necessary (Figure 2):

Figure 2.

Flow diagram of the methodology.

Phase 1. Revision of the inventory of exokarstic forms (caverns, dolines, uvalas, and poljes – “cenotes” and “rejolladas”) reported in Cozumel Island [1, 2, 13, 14, 15, 16, 17], as well as the reports from the association of speleologists MAYAB AC [18] were consulted, 37 of the reported karst units were recognized in these studies, and a georeferenced database was also generated for the purpose of making the information available for reference.

Phase 2. Construction of a unified LiDAR mosaic of the terrain of Cozumel Island. To do so, the data from point clouds in 32 TIFF information files generated by the ALS 40 system and an information area of 62,556.27 km2 [19] were revised and corrected; the data resolution is 5 m for the X and Y axes and 15–20 cm for the Z axis. The resolution scale is 1:10,000 for each cartographic model.

Phase 3. The application of filters for the elaboration of morphometric digital models of elevation, slope, and shadows. The digital model of elevation was derived from the simple method of nearest neighbor interpolation, rated at 0.05 cm elevation, recognizing up to 16 m of altitude in the area of study [20]. For the terrain shading method, an azimuth of 272° and an elevation of 30° was used, and for the slope map, eight categories of classification were set manually, where the highest category is >40° [21].

Phase 4. Analysis of the distribution of units in the inventory and their contrast with the new models of identified forms. In each one of the models, 109 dolines and uvalas were visually identified, using the high resolution models and the contrast of the altimetric data, slope, and shading. Likewise, the information was contrasted with the 37 units reported in the previous studies. Furthermore, field trips were carried out during the months of June and July 2014 to confirm the occurrence of the dolines and uvalas, where the cartographic prospection at 1:10,000 scale was verified.


3. Results and discussions

In the identification of the exokarstic forms, the primary input used was a mosaic of LiDAR images of terrain with a resolution of 5 horizontal m and 0.50 vertical cm, to which later a shade filter with azimuth 272° and height 30° is added (Figure 3). Here, dolines and uvalas with diameters between 20 and 125 m and up to 9 m of depth can be identified, as well as basic linear details of the relief. In the mosaic, one can see the geological structure of the quaternary period (old coastal mountain ranges, dunes, and marine terraces).

Figure 3.

Elevation map with a vertical difference of 0.50 cm and terrain shading model (azimuth 227° and elevation 30°) with details about the known karst formations. Prepared by author based on LiDAR data [8].

By applying an altimetric filter with a vertical difference of 0.50 cm, karstic formations are distinguishable. For said identification, the distribution of pixels based on their altitude values was taken into account; groups that presented values in ascending order from the center to the shores were sought out, since this characteristic indicates that there are depressions. Furthermore, a visual interpretation was done in which the identification criteria was the geometric form of said groups. In this case, the semicircular form is associated with the dolines and the irregular forms with processes of the formation of uvalas. The dimensions are congruent with that which was identified in the shading model.

Finally, with the contrast of the slope (see Figure 4), where the association of the semicircular forms and forms with a gradient greater than 25° – corresponding, in the majority of cases, to the borders of the uvalas or collapse cenotes – the altimetric difference was up to 8 m of depth. The units with lower gradients (<25°) and altitude contrast (up to 1 m in height) are related to dissolution dolines.

Figure 4.

Slope map with six gradient categories. Prepared by author based on LiDAR data [8].

In the records of dolines and uvalas in the region of study, the existence of 37 karstic forms are reported, all of which have been reported as points (with latitude and longitude coordinates), the majority of which measure less than 10 m in diameter [1, 17]; with the interpretation of the models derived from the LiDAR data, 109 forms were able to be identified, with their diameter and depth. The cartography of all of the sites is shown in Figure 5.

Figure 5.

Map of point distribution of the karst forms in Cozumel Island. Prepared by author based on data reported [12, 13, 14, 15, 16, 17, 18, 19] and this study.

Furthermore, the density of dolines and uvalas in square kilometers is shown in Figure 6. The concentration can be explained by the fact that the area of the highest relief (>10 masl), which corresponds to a marine terrace in which the karstification processes are more evident than in the periphery as well as other morphological genesis (dunes, coastal mountain ranges, lagoons, and shoals), located along the coast and the north of the island, and whose formations can be buried or subterranean (such as caverns and grottoes).

Figure 6.

Density model of karstic forms by km2 in Cozumel. Prepared by author based on the identification of dolines and uvalas using the LiDAR model [8].

Although there is a new distribution model of dolines and uvalas, as well a proposal for density of karstic forms, it must be considered that in this study, the interpretation and identification of exokarstic forms follows a traditional process of relief form analysis [3, 5], based on the visual morphological differences (crests, slopes, surface, and background), known for their morphometric elements (height, slope, and depth).

The advantage of using LiDAR models, which are free for the public to access through the National Institute of Statistics, Geography and Informatics of Mexico, allows the processing of detailed data at a detailed horizontal scale (5 × 5 m) and at 0.05 cm of vertical height. The aforementioned was not possible until the year 2010, when LiDAR flights were carried out in this area of study, characterized by low altimetric contrast (<10 m) and by a base cartography at 1:50,000 scale. For this reason, it is not possible to identify karstic forms with dimensions less than 50 m, where the cost would be excessive, as pointed out by others authors [10, 14, 19].

New dolines and cenotes have been identified, along with the areas of greater density. However, an accurate characterization of each identified unit is needed to increase the understanding of the types of dolines (dissolution, collapse, and suffusion) and uvalas (first, second or third generation), their relation with the creation (structural or climatic) and with the type and intensity of the process of karstification.


4. Conclusions

The use of LiDAR data is not new to the study of karstic reliefs in tropical areas, where data has been manipulated by filters to highlight forms in the relief; among the most common are the hypsometric models, field studies, and slope.

In the area of study, there are no reports about the use of LiDAR technology for the recognition of dolines and uvalas, and consequently the use of LiDAR data is a proposal to identify and enrich the doline and uvala inventories in karst areas with little altimetry contrast. Even though in this study, new dolines and uvalas are reported, it lacks the complete verification and error estimation in the data interpretation and its concurrence with the reports in the literature.

The density map of dolines and uvalas have a resolution of 5 × 5 m, being a detailed-scale map which serves to orientate new searches for the calibration of LiDAR data and to be able to orientate its use in the entirety of the northeastern zone of the Yucatan Peninsula, where LiDAR data are open and free to access for scientific and academic study, with the potential for application to and studies about quaternary geology, the evolution of the landscape, the evidence of the coastal dynamic, and the transformation of the landscape in the area inhabited by the Maya of the Yucatan.



Thanks are due to the project Water and extreme hydrometeorological phenomenon in the Yucatan peninsula by CONACYT – REDESCLIM for their help. To the summer of scientific research for the help of student Angie Zapi and the trainee program at UQROO – University of Applied Sciences, Jena, Germany, for the help of student Christian Koch.


  1. 1. Mejía-Ortíz G, Yáñez ML-M, Zarza-González E. Cenotes (anchialine caves) on Cozumel Island, Quintana Roo, México. Journal of Cave and Karst Studies. 2007;69(2):250-255
  2. 2. Frausto O, Ihl T, Giese S, Cervantes A, Gutierrez M. Vulnerabilidad a la inundación en las formas exocarsticas del noreste de la península de Yucatán. Memorias del VI seminario latino – Americano de geografía Física, Universidad de Coimbra, maio de 2010. 2010. Available from:
  3. 3. Nayengandhi A, Brock J. Assessment of coastal vegetation habitats using LiDAR. In: Yang X, editor. Lecture Notes in Geoinformation and Cartography – Remote Sensing and Geoespatial Technologies for Coastal Ecosystem Assessment and Management. Heidelberg, Germany: Springer Publications; 2008. pp. 365-389
  4. 4. Mckean J, Nagel D, Tonina D, Bailey P, Wright C, Bohn C, Nayengandhi A. Remote sensing of channels and riparian zones with a narrow – beam aquatic – terrestrial LiDAR. Remote Sensing. 2009;1(4):1065-1096
  5. 5. Zhu Y, Taylor TP, Currens JC, Crawford MM. Improved karst sinkhole mapping in Kentucky using LiDAR techniques: A pilot study in Floyds Fork Watershed. Journal of Cave and Karst Studies;76(3):207-216. DOI: 10.4311/2013ES0135
  6. 6. Brock C, Purkis S. The emerging role of LiDAR remote sensing in Coastal research and resource management. Journal of Coastal Research SI. 2009;53:1-5
  7. 7. Bayer JM, Scheid JL, editors. PNAMP Special Publication: Remote Sensing Applications for Aquatic Resource Monitoring, Pacific Northwest Aquatic Monitoring Partnership. Washington: Cook; 2009. 100 p
  8. 8. Aguilar FB, Mendoza ME, Frausto O, Ihl T. Density of karst depressions in Yucatán state, Mexico. Journal of Cave and Karst Studies. 2016;78(2):51-60. DOI: 10.4311/2015ES0124
  9. 9. Delgado-Carranza C, Bautista F, Orellana-Lanza R, Reyes-Hernández H. Classification and Agroclimatic Zoning Using the Relationship between Precipitation and Evapotranspiration in the State of Yucatán. Vol. 75. Mexico: Investigaciones Geográficas, Boletín del Instituto de Geografía, UNAM; 2011. pp. 51-60. DOI: 10.14350/rig.29795
  10. 10. Wurl J, Giese S. Ground water quality research on Cozumel Island, State of Quintana Roo, Mexico. In: Frausto Martínez O, editor. Desarrollo Sustentable: Turismo, costas y educación. México: Universidad de Quintana Roo; 2005. pp. 171-176
  11. 11. INEGI. Modelo digital de elevación LiDAR. México: Instituto Nacional de Estadística, Geografía e Informática; 2011. Available from:
  12. 12. Orellana R, Nava F, Espadas C. El Clima de Cozumel y la Riviera Maya. In: Mejía-Ortíz LM, editor. Biodiversidad acuática de la isla de Cozumel. México, D. F.: Universidad de Quintana Roo – Plaza y Valdés; 2007
  13. 13. Fragoso-Servón P, Pereira A, Frausto O, Bautista F. Geodiversity of a tropical karst zone in South-East Mexico. In: Andreo B, Carrasco F, Durán J, Jiménez P, LaMoreaux J, editors. Hydrogeological and Environmental Investigations in Karst Systems. Environmental Earth Sciences. Vol. 1. Berlin, Heidelberg: Springer; 2015
  14. 14. Coronado-Álvarez L, Gutiérrez-Aguirre M, Cervantes-Martínez A. Water quality in wells from Cozumel island, Mexico. Tropical and Subtropical Agroecosystems. 2011;13(2):233-241
  15. 15. Fragoso-Servón P, Pereira Corona A, Bautista Zúñiga F, Gonzalo de Jesús Zapata B. Digital Soil Map of Quintana Roo, Mexico. Journal of Maps. 2017;13(2):449-456. DOI: 10.1080/17445647.2017.1328317
  16. 16. Cervantes-Martínez A, Gutiérrez-Aguirre M, Álvarez-Legorreta T. Indicadores de calidad del agua en lagunas insulares costeras con influencia turística: Cozumel e Isla Mujeres. Quintana Roo, México: Teoría y Praxis; 2015. pp. 60-83
  17. 17. Fragoso-Servón P, Bautista F, Frausto O, Pereira A. Caracterización de las depresiones kársticas (forma, tamaño y densidad) a escala 1:50,000 y sus tipos de inundación en el Estado de Quintana Roo, México. Revista Mexicana de Ciencias Geológicas. 2014;31(1):127-137
  18. 18. Yañez G. Sinkhole Database of Cozumel. Cozumel, Mexico: Mayab AC; 2015. Available from:
  19. 19. Frausto-Martínez O, Ihl T, Rojas López J. Risk Atlas of Cozumel Island, Mexico. Teoría y Praxis [en linea]. 2016 (Octubre-Sin mes). Availble from:
  20. 20. Weishampel J, Hightower J, Chase A, Chase D, Patrick R. Detection and morphologic analysis of potential below-canopy cave openings in the karst landscape around the Maya polity of Caracol using airborne LiDAR. Journal of Cave and Karst Studies. 2011;73:187-196. DOI: 10.4311/2010EX0179R1
  21. 21. Gesch DB. Analysis of LiDAR elevation data for improved identification and delineation of lands vulnerable to sea-level rise. Journal of Coastal Research: Special Issue. 2009;53:49-58

Written By

Oscar Frausto-Martínez, Norma Angelica Zapi-Salazar and Orlando Colin-Olivares

Submitted: 01 March 2018 Reviewed: 29 May 2018 Published: 05 November 2018