Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 Introduction

Mesoamerica is a term used sometimes in cultural context, but in this article we are using it to name the land bridge between North and South America, made up of the Southern Mexico States (Chiapas, Quintana Roo, Yucatán, Campeche y Tabasco), Guatemala, Belize, El Salvador, Honduras, Nicaragua, Costa Rica, and Panama. With an area of approximately 755,000 km2, it is one of the most heterogeneous regions of the world in terms of elevation, land forms, climate, natural ecosystems, and human populations.

With an estimated population of 58 million inhabitants by 2015, of which close to one third are under 15 years of age, the population is expected to grow to 69 million inhabitants by 2030 (Celade 2004, 2014; CONAPO 2014). Population growth dynamics vary from country to country with extremes in Guatemala, which is projected to have between 2015 and 2030 a 1.87 %/year increase in population and El Salvador that is projected to have 0.39 %/year increase for the same period. Ethnic diversity is another characteristic of human populations in Mesoamerica, with over 60 different ethnic/linguistic groups (Center for Support of Native Lands, National Geographic 2002). Around one third of the population lives with less than US$4/day and therefore they are considered poor, although the trend appears to be the reduction of this number (World Bank 2013).

The region has also been the scene of violent conflicts, including civil wars in Guatemala, El Salvador and Nicaragua that concluded by the end of the 1990s decade. More recently, and in connection with common crime, gangs, and organized crime, four of the eight countries in Mesoamerica (Honduras, El Salvador, Belize, and Guatemala) have been included in the top ten ranking of murder rates per 100,000/inhabitants (UNODC 2013). The aggregate effects of poverty and violence are the believed cause of massive migrations, highlighted recently by the crises of unaccompanied children migrating from Central America and Mexico into the U.S. (Pierce 2015).

Mesoamerica is one of the original proposed global biodiversity hotspots and contains 24,000 species of plants, 2859 species of vertebrates and is one of the leading regions of the world regarding the presence of endemic species (Myers et al. 2000). The northern part of the region is one of the eight centers of origin of domesticated plants, for world crops like maize, beans, peppers, tomato, cotton, and cacao (Vavilov and Freier 1951). Around 28 % of all the terrestrial area is under some kind of formal protection (UNEP-WCMC 2014) but there is concern regarding the limitations of the effectiveness of protected areas in conserving biodiversity, preventing deforestation and degradation(Blackman et al. 2015), which is often interpreted as the results of economical and social issues and the general scarcity of resources to manage and administer protected areas.

Deforestation is one of the biggest threats for biodiversity and protected areas in Mesoamerica. An estimated 5 million hectares of forest have been lost between 2000 and 2013 in the eight countries of the region (Hansen et al. 2013). Deforestation has been historically linked to the promotion of colonization by governments, spontaneous migration and more recently to the increasing importance of agro-industrial crops like oil palm (Rival and Levang 2015). Fire in the wild-lands is closely related to agricultural expansion because of the almost universal use of fire as a clearing tool and is the other big threat to biodiversity. Severe fire events occurred in 1998, 2003, and 2005 when hundreds of thousands of hectares of forests and savannas burned for several months and smoke of those burnings reached the U.S. (Wang et al. 2006).

Mesoamerica is also a region vulnerable to disasters that cause periodically important human and material losses. Statistics between 1990 and 2014 indicate that there have been 36 earthquakes, 104 floods, 17 landslides, 82 storms, and 21 volcanic events. These disasters have caused more than 28 thousand deaths, left homeless close to one million people and caused US$50 billion in damages (Guha-Sapir et al. 2009).

In the general context given, the potential of Earth Observation (EO) to assist the management of natural resources, biodiversity, and disasters in the region is clear. In this chapter, we discuss the current state of EO applications and future perspectives related to Land-use Change, Ecosystem Dynamics and Biodiversity, and Solid-Earth Hazards. We hope that this contribution can identify current and future challenges related to obtaining the biggest societal benefit of EO and suggest actions to take advantage of anticipated innovations and data availability.

2 Current State

2.1 Land-Use Change, Ecosystem Dynamics, and Biodiversity

Wide availability and free access of remote sensing data along with advances in communications technology have made possible, relatively recently, an exponential growth of the use of EO data. The turning point in this growth of applications was probably the decision to make freely available all Landsat data in the U.S. Geological Survey (USGS) archive by 2008, when distribution went from a few thousand scenes by year to millions by year (Turner et al. 2015). Mesoamerica is no exception to that trend and increasingly more uses and applications are present in government, private, and academic contexts.

To characterize the recent progress and current state in the use of EO data in applications related to land use change, ecosystems, and biodiversity carried on in Mesoamerica we selected examples at national and regional scales. Information on 19 cases has been collected, 11 of them focused at the national level, and 8 are regional or multicountry level. A summary of these cases is presented in Table 1.

Table 1 Selected EO applications in Mesoamerica on land-use change, ecosystem dynamics, and biodiversity
Full size table

The earliest analysis included in Table 1 (Vreugdenhil et al. 2002) was the Map of Ecosystems of Central America and had the primary objective to describe the ecosystems to establish a baseline of biodiversity in the region and serve as a basis for biological monitoring programs. The map was also intended to be used as a reference in the design of the Mesoamerican Biological Corridor. The map was derived using visual interpretation of Landsat scenes with local experts in each country, helping to delineate ecosystems with a strong emphasis on field verification and ecosystem data collection. The map was perhaps the first time EO data was used at the regional scale, and still today is used even if the acquisition date of the Landsat scenes ranged between 1997 and 1999.

As expected, the most often used EO data comes from the Landsat Project (63 %), followed by the Moderate Resolution Imaging Spectroradiometer (MODIS) instrument (42 %), RapidEye and Advanced Very High Resolution Radiometer (AVHRR) (both with 16 %). Other data used included SPOT Vegetation, Medium Resolution Imaging Spectrometer (MERIS), Phased Array type L-band Synthetic Aperture Radar (PALSAR), Shuttle Radar Topography (SRTM), Tropical Rainfall Measuring Mission (TRMM), aerial photography, SPOT, Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER), Worldview and Airborne Light detection and ranging (LiDAR). More than half of the applications (53 %) used multi-sensor, multi-resolution data, with two cases in which at least four sensors were integrated to obtain a final product (Cherrington et al. 2011; Blackman 2013).

The dominant application found was land use and land-use change mapping (53 %), often explicitly coupled with REDD+ support (42 %), followed by fire management mapping and support (21 %). Other applications included ecosystem mapping (16 %), biomass mapping (10 %), ocean monitoring (10 %) and only in one case (Anderson et al. 2008), biodiversity modeling was a primary objective of the application (4 %).

Applications working in an operational mode (21 %) include the production of fire monitoring outputs (Ressl et al. 2009), ocean monitoring (Graves et al. 2007; Cerdeira López 2011), and a REDD decision support system (Blackman 2013). Applications that do not produce outputs uninterruptedly, but that will likely do so periodically on an annual or multi-annual basis (37 %) are mostly related to REDD+ support, fire management, mapping and land use and land-use change mapping (CONAP, INAB, CONRED, MARN 2010; Cherrington et al. 2010; INAB, CONAP, UVG, URL 2012; Rodríguez-Zúñiga et al. 2013; Jimenez 2013; Gebhardt et al. 2014 and Agresta, Dimap, Universidad de Costa Rica, Universidad Politécnica de Madrid 2015).

Government institutions participated in the majority of the applications (58 %), followed by regional or multilateral institutions (42 %) and academic institutions (26 %). CATHALAC, SERVIR and CONABIO were involved in almost half (47 %) of the applications and maintain all the ones considered operational.

With this general overview obtained from the selected applications reviewed, we can highlight several findings:

  1. (a)

    Landsat and MODIS are the most used sensors in the region and are critical for several current and future applications, including those related to REDD+ and fire monitoring. It appears that the use of high resolution data, in particular RapidEye, is gaining traction in the region, in part because the need of more detailed products (for example to map forest degradation) and also because the REDD-CCAD/GIZ Program provided all the Central America countries with a wall to wall coverage year? Is notable the limited use of SAR data in a region even though there are zones (the Caribbean coasts of Honduras, Nicaragua, Costa Rica, and Panama) where cloud coverage makes the acquisition of useful optical data very difficult.

  2. (b)

    With only one example of an application explicitly aimed to biodiversity issues, it appears to be necessary to promote a more active involvement of the conservation and protected area communities in the systematic use of EO data in the region. Several frameworks that propose specific ways in which EO can help biodiversity and conservation have emerged recently (Secades et al. 2014; Rose et al. 2015; Skidmore et al. 2015) and they should be examined to give direction to regional and national efforts.

  3. (c)

    REDD+ is an emerging driving factor in the development of EO applications. All or almost all the countries in Mesoamerica have joined the Forest Carbon Partnership Facility (FCPF) and the UN-REDD Program. That implies that, to go forward with REDD, there is an obligation to build robust and transparent national forest monitoring systems that will, at least, produce baseline and consistent frequent information related to land use, land-use change, forest biomass, and forest degradation. That represents clear opportunities to develop EO applications, that will need to persist in the long term, and that may have additional uses and benefit fields like ecosystem and biodiversity monitoring.

  4. (d)

    The role of SERVIR and CATHALAC has been crucial in the advances in the region and both deserve the credit for substantial progress in the transfer of knowledge, the training of human resources in the region and the development of relevant applications. However, the dependence on projects with a limited life-span that, with a few exemptions, wrap up after the project has finished and the funding has depleted is an issue that has to be addressed. If long-term applications are necessary and desirable, a bigger share of the costs of implementation have to be absorbed by the local governments and other local stakeholders with a clear long-term vision. Public institutions involvement in the selected examples is relatively high, but, they often act as cooperation recipients, and that needs to transition to full adoption of applications, included the costs that may entail.

  5. (e)

    Mexico is, clearly, the regional leader in terms of systematic use and high-level of sophistication of EO applications used in land-use monitoring, biodiversity, and ecosystems. The experience and capabilities of Mexican institutions are already being used to transfer technology and knowledge to the other countries in the region trough the “South-South cooperation to exchange experiences and capacities on MRV systems and REDD+” project (http://mrv.cnf.gob.mx/index.php/en/) and also through the provision of EO-derived data to users in the region (Ressl et al. 2009).

2.2 Solid-Earth Hazards, Resources, and Dynamics

The unique location of Mesoamerica as an isthmus between North and South America and in the Inter-tropical Convergence Zone positions this region as one the most biodiverse territories on Earth. On the other hand, this unique location also makes the region more prone to a range of natural hazards, such as volcanic eruptions, earthquakes, hurricanes, heavy rains, floods, landslides, droughts, and storm surges, to mention some (REDLAC 2008). The high presence of hazards in combination with socioeconomical and political issues creates high vulnerability conditions for the population. More than half of the population in Central America and the Caribbean are at risk from multiple hazards (Dilley 2005).

This region has a high mortality risk from geophysical (earthquakes and volcanoes) and hydrological (floods, cyclone, and landslides) hazards (Dilley 2005). El Salvador, Costa Rica, Guatemala, Dominican Republic, Nicaragua, and Honduras rank among the top 15 countries at relatively high mortality risk from multiple hazards (Dilley 2005).

Further, according to the Climate Risk Index, the top ten countries more affected by disasters between 1997 and 2006 were led by Honduras and Nicaragua at a global level (Harmeling 2007; REDLAC 2008).

According to the EM-DAT International Disaster Database (Guha-Sapir et al. 2009) from 1960 to 2015 there have been 497 natural disasters in Mesoamerica and Dominican Republic that include drought, earthquake, extreme temperature, flood, landslide, mass movement (dry), storms, volcanic activity, and wildfire. Two hundred of these disasters occurred between 2001 and 2015. This means that about 40 % of the disasters reported in the past half century have occurred in the past 15 years. The accuracy of these disaster estimations may very depending upon the methodology used to collect the data, but it shows a general trend of disasters on the rise in Mesoamerica, which aligns with the global trend. Among the most frequent disasters in Mesoamerica and Dominican Republic are floods and storms accounting for 34 and 27 %, respectively, of the total disasters reported between 1960 and 2015 (Fig. 1). The third most frequent disaster in the region is earthquakes. Accounting for 13 %. In the 2001–2015 period, the trend of disaster affecting the region is the same.

Fig. 1
figure 1

Types of natural disasters which affected Mesoamerica and Dominican Republic from 1960 to 2015. Data source EM-DAT The International Disaster Database (Guha-Sapir et al. 2009)

Full size image

The use of EO to provide actionable information in the case of disasters has been widely used in Mesoamerica and the Caribbean. There are several examples of the use of satellite remote sensing to assess the damages inflicted by disasters, to monitor weather events, and to provide valuable information for early warning (REDLAC 2008). This section will focus on examples in which EO data have been used primarily in post-disaster assessment, particularly for response and recovery purposes.

An example of the use of EO in disaster response is the International Charter Space and Major Disasters, (Charter for short). The Charter is “an international collaboration among EO mission owners/operators to provide space-based data and information in support of relief efforts during emergencies caused by major disasters” (Gevorgyan and Briggs 2014).

The Charter only supports the phase of immediate response to a disaster (Courteille 2015; Gevorgyan and Briggs 2014). It was created in 1999, after Hurricane Mitch had struck Mesoamerica, at the Third United Nations Conference on the Exploration and Peaceful Use of Outer Space (UNISPACE) in Vienna, where the European Space Agency (ESA) and the French space agency (CNES) proposed to supply satellite imagery to emergency responders (Rocchio 2014).

Since its activation in 2000 the Charter has covered more than 400 disasters in over 110 countries worldwide (Courteille 2015). The charter is now composed of 15 members from different space agencies and/or mission operators around the world. From the U.S., the National Oceanographic and Atmospheric Administration (NOAA), with its fleet of meteorological satellites, and the U.S. Geological Survey are authorized members of the Charter.

The Charter first provided satellite data to Mesoamerica in 2001 for an earthquake in El Salvador. Since then it has provided data for 27 different disasters covering Mesoamerica and the Hispaniola (Haiti and Dominican Republic). See Table 2.

Table 2 International charter space and major disaster activations for Central America and Hispaniola (2000–Oct 2015), organized by activation date
Full size table

The Charter has been activated mainly for hydrological-related disasters in Mesoamerica (Fig. 2). Hurricanes, tropical storms, heavy rains, and the resulting flooding and landslides account for more than 70 % of the major disasters in Mesoamerica and the Hispaniola that have triggered the Charter activation (Fig. 2).

Fig. 2
figure 2

Charter Activations by Disaster Type in Central America, South of Mexico, and Hispaniola from January 2001 to October 2015. Source compilation of data from: https://www.disasterscharter.org/, Visited on October 2015. Note Cyclone disasters include hurricanes, tropical storms, and tropical depressions

Full size image

Once the Charter has been activated, a project manager (PM) takes charge to quickly analyze the collected data and create value-added products for first responders (Rocchio 2014). Different types of satellite data are acquired through the Charter such as passive and active satellite remote sensing data, and from medium to high spatial resolution.

The Water Center for the Humid Tropics of Latin America and the Caribbean (CATHALAC), in Panama, has acted as the project manager of 9 out of the 27 Charter activations in Mesoamerica and Hispaniola. Refer to Table 2. CATHALAC first hosted SERVIR in 2005. SERVIR links EO and geospatial technologies with developmental decision-making in SERVIR regions. Proven success of SERVIR in Mesoamerica has resulted in expansion to other regions. Currently SERVIR actively works in Eastern and Southern Africa, the Hindu Kush-Himalaya region, and the Lower Mekong region. CATHALAC—in the context of SERVIR and acting as the Mesoamerican hub- has served to connect the scientific, analytical, and decision-making needs of its users in the region. In Mesoamerica, SERVIR provided rapid response support during disasters by acquiring expedited satellite images, and creating and sharing value-added products with end users (Flores et al. 2012). While it is no longer an active SERVIR hub, CATHALAC continues to meet the needs of its network of users by applying space data to disaster response support.

Flores et al. (2012) describes the type of EO data and information used in SERVIR-Mesoamerica from 2004 to 2011 to create value added products provided for monitoring, response, recovery and general disaster awareness.

Table 3 lists the EO-derived measurements and/or products used by SERVIR-Mesoamerica from 2004 to 2011 to generate added-value products for a variety of disasters.

Table 3 EO-derived measurement and/or product used by SERVIR-Mesoamerica from 2004 to 2011 to analyze different type of disasters
Full size table

Then, Table 4 lists the respective satellite sensors from which the derived EO measurement and/or product were generated, respectively.

Table 4 List of satellite sensors used to generate each derived measurement and/or product by SERVIR-Mesoamerica from 2004 to 2011
Full size table

Some Level 3 satellite-derived products such as land cover, and model products such as gridded population distribution were used to link the physical hazard (cyclone, heavy rain, flood, landslide, etc.) with the human or ecological factor at risk (Flores et al. 2012).

The majority of the satellite sensors used by SERVIR-Mesoamerica to provide added value products were from passive remote sensors that collect electromagnetic energy reflected or emitted by Earth’s surface. Using this type of sensors proves difficult when estimating flooded areas during heavy rain and high cloud coverage conditions. Active remote sensors, such as C-band SAR and L-band SAR on board of Radarsat 1 and 2 and ALOS, respectively, were used heavily in cases with high cloud coverage particularly to map flooded areas during disasters. Hyper spectral satellite remote sensing was used primarily for water quality assessment during cases of algal blooms and was also used for monitoring volcano activity.

Some of the satellite data used by SERVIR-Mesoamerica to address disasters was acquired through the Charter, but not all. SERVIR-Mesoamerica also provided value-added products to smaller disaster events that did not trigger the Charter activation.

SERVIR started in Mesoamerica when Landsat data was still not freely available, however, through SERVIR the region started getting access to such datasets and used for environmental monitoring and disaster response. As SERVIR-Mesoamerica started building up during its 5 years it also increased the generation of valued-added products for disasters to attend user needs in the region.

Today disaster response agencies across the region are heavily using EO data to address disaster response. As example national disaster institutions in Mesoamerica have become more involved in the Charter process to obtain satellite data, such is the case of SINAPROC in Panama, and CONRED in Guatemala. This last one has even acted as a project manager of the Charter.

3 Future Prospects

3.1 Land-Use Change, Ecosystem Dynamics, and Biodiversity

Despite the challenges, the future looks promising for EO in Mesoamerica. Access to free and open data is growing at a very fast pace. Beyond Landsat we can already use data of the Sentinel constellation that may solve problems related to cloud coverage (Sentinel-1) and provide a new and valuable resource for middle resolution optical data (Sentinel-2). Already free, or being put to public access are historical data of SPOT and ALOS PALSAR providing an additional opportunity to add more layers of EO data. Data continuity for middle resolution data appears to be solved in the mid-term, with for example, plans to build, and launch Landsat 9 underway.

We think that REDD+, fire and climate change monitoring are going to be pivotal for EO applications in the near future in Mesoamerica. REDD+ processes have started in all the countries in the region, and a great part of the success of the mechanism relies on the premise of being able to monitor forests continuously and with a level of detail that will need to resolve at a highest resolution possible deforestation and degradation of forests.

Fire, closely related to REDD+, because is a cause of forest degradation, is an increasing problem in the region and also a phenomenon of interest that can be monitored using EO. We will need to be able to go from just monitoring fire, to actually helping to modify significantly its management with early alert systems tied to regulations on the use of fire as an agricultural tool, for example.

REDD+ and fire applications have the potential to be designed in such a way, that products and outputs can be used in other fields, including biodiversity, ecosystem monitoring, climate change, water management, and agriculture. Because of that, communication and collaboration between remote sensing scientists, conservation practitioners and natural resource managers is fundamental.

3.2 Solid-Earth Hazards, Resources, and Dynamics

The current disaster trend indicates that disasters are increasing at the local and global scale (Guha-Sapir et al. 2009).Footnote 1 This is particularly dangerous for regions like Mesoamerica that have a high mortality risk from geophysical and hydrological hazards (Dilley 2005). The use of EO data has proven useful for disaster response and recovery and given the increasing positive trend of events it is expected that the role of EO in disaster management will intensify in years to come.

The current availability of new satellite data such as the Visible Infrared Imaging Radiometer Suite VIIRS on board of Suomi National Polar-orbiting Partnership (NPP), and Landsat 8 will allow continuity of products already used in Mesoamerica for disaster monitoring and response, such as the case of MODIS products used for fire monitoring, burned areas, and land cover to mention some. In addition, new missions such as the Soil Moisture Active Passive (SMAP) will generate new understanding of weather and climate patterns.

A pressing need of the technical community using satellite data for disaster assessment, particularly for response and recovery is the access to radar data to map floods under high cloud cover. The majority of the disasters in this region are related to hydrological hazards Figs. 1 and 2, which trigger other set of hazardous events, such as floods and landslides.

The new C-band SAR sensor, Sentinel 1 from the new European EO Programme, Copernicus, will be instrumental in providing information during high cloud coverage. The advantage of Sentinel 1 when compared to other radar data is that is freely available. This will open its use and application not only for disaster-related analyses but also for other applications.

NASA and Indian Space Research Organization (ISRO) SAR Mission (NISAR) also promises to become a source of valuable data for disaster risk management with the potential ability to measure deformation before and after geophysical hazard events.

4 Conclusion and Way Forward

There is no doubt about the fact that EO holds an immense potential in applications related to land use, ecosystems, biodiversity, and disaster risk management. We think that is also true that this potential has not been fully used in Mesoamerica (and elsewhere in the tropics). There are several challenges to improve on that:

  1. (a)

    Governments in the region that, in most cases, are the potential users with the power to modify policies and effectively generate change, invest very little in agencies in charge of natural resources, ecosystems, and biodiversity management. Therefore, these agencies often lack the resources to maintain technical units in charge of applications of EO.

  2. (b)

    EO data is often presented in formats that are not legible enough for the audiences that can influence and make decisions, including the general public. The message that EO delivers has to be clear and understandable to all.

  3. (c)

    There is often a disconnection between user needs and products. We have to examine carefully if the processes we use to choose what to do is backed by enough feedback from the people and entities that will actually use EO products to make decisions.

  4. (d)

    It is critical to understand that one of the main values of EO data relies on systematic, long-term observations that actually lead to information on changes and trends. Applications that emphasize research and demonstration obviously are critical for the development of EO, but the ones that can be used operationally and with a clear, practical purpose are those that function and provide societal benefits over a long time.

  5. (e)

    We have to aim, when possible, at the use of EO application products to support compliance, regulations, and law enforcement. Doing that will represent a technical and perhaps legal challenge, but it would be a significant contribution to governance and environmental justice.

  6. (f)

    At regional and local scale the use of EO data should be appropriately integrated into the operational processes of disaster risk management agencies to support all the phases of disaster risk management. The generation of geospatial-baseline information primarily in urban areas is critical to properly assess pre and post-event damage needed for immediate response and recovery. Such baseline information becomes more critical in this region that is undergoing population growth and is prone to natural hazards including earthquakes.