National ecosystems services priorities for planning carbon and water resource management in Colombia

Land Use Policy 42 (2015) 609–618

Contents lists available at ScienceDirect

Land Use Policy journal homepage: www.elsevier.com/locate/landusepol

National ecosystems services priorities for planning carbon and water resource management in Colombia N. Rodríguez a,∗ , D. Armenteras a , J. Retana b a Laboratorio de Ecología del Paisaje y Modelación de Ecosistemas, Departamento de Biología, Facultad de Ciencias, Universidad Nacional de Colombia, Carrera 30 No. 45-03, Bogotá, Colombia b Centre de Recerca Ecologica i Aplicaciones Forestals i Unitat d’Ecología, Universitat Autònoma de Barcelona, Campus de Bellaterra, 08193 Cerdanyola del Vallès, Bellaterra, Catalonia, Spain

a r t i c l e

i n f o

Article history: Received 18 July 2013 Received in revised form 29 May 2014 Accepted 17 September 2014 Keywords: Carbon storage Water provision Ecosystem services mapping Watershed Planning

a b s t r a c t The modelling and mapping of ecosystem services (ES) are important components of any programme of land management planning, as they help evaluate the potential benefits that ecosystems provide to society. The objective of this paper is to evaluate ES for setting priorities and planning carbon and water resource management in Colombia. By using information related to provision and regulation services for water, carbon storage and protection services against extreme events such as landslides, we have evaluated the spatial distribution of ES and identified geographical hotspots. The results are presented for two levels of analyses: (1) natural regions and (2) watersheds. We found differences in the distribution and range of values for ES and observed that each region and watershed tends to maximise one or two services, with the exception of the Caribbean region, which presents low values for most services. The services of water resources provision, regulation of water flow and carbon storage in the above-ground biomass presented high correlations among them, with the Pacific and Amazonian regions presenting the highest average values for these ES. The Andean region was important for the prevention of landslides and the amount of carbon in the soil. At the watershed level, the Amazon watershed and those associated with transition areas (piedmont) between the Andes and the lowlands of the Amazonian, Orinoquia and Pacific regions were the areas where the greatest number of hotspots was concentrated. These results provide valuable information on how better use official institutional information to quickly define and prioritise ES, to guide management actions within the country’s recent policies on integrated water resources management and on biodiversity and ES. © 2014 Elsevier Ltd. All rights reserved.

Introduction The concept of ecosystem services (ES), defined as the benefits that human beings obtain from ecosystems (MA, 2005), has become a key tool for different stakeholders interested in linking natural, human and economic systems (Armsworth et al., 2007; Bryan et al., 2010; Muradian and Rival, 2012). There is an increasing interest in incorporating ES into environmental planning policies (de Groot et al., 2010; Muradian and Rival, 2012; Viglizzo et al., 2012) and into the design of objectives and strategies for landscape management, with the aim of improving the provision of these services to society

∗ Corresponding author at: Universidad Nacional de Colombia, Laboratorio de Ecología del Paisaje y Modelación de Ecosistemas. Carrera 30 No. 45-03, Edificio 421, Of. 223, Bogotá, D.C., Colombia. Tel.: +57 1 3165000x11333. E-mail addresses: [email protected], [email protected] (N. Rodríguez). http://dx.doi.org/10.1016/j.landusepol.2014.09.013 0264-8377/© 2014 Elsevier Ltd. All rights reserved.

(Daily and Matson, 2008; Farrell and Anderson, 2010; Kroll et al., 2012). In the past decade, many publications have synthesised discussions around the ES concept and their classification system (de Groot et al., 2002; Wallace, 2007; Costanza, 2008; Braat and de Groot, 2012), payment for ES (PES) (Van Hecken and Bastiaensen, 2010; Wünscher and Engel, 2012) or the link between ecosystem processes, biodiversity, climate change, land use and ES (Egoh et al., 2007; Fu et al., 2012; Armstrong et al., 2012; Mace et al., 2012). However, there is still a lack of information and empirical data on the distribution, service fluxes and trade-offs between different landscape functions and how these ES change over time and across different spatial scales (de Groot et al., 2010; Haines-Young et al., 2012; Mace et al., 2012). One of the requirements to apply the ES concept and decisionmaking is the quantification and mapping of services (Braat and de Groot, 2012; Burkhard et al., 2012; Maes et al., 2012a). Mapping ES is the initial step for the subsequent analyses of ecological, social

610

N. Rodríguez et al. / Land Use Policy 42 (2015) 609–618

and economic processes associated with ES (TEEB, 2010). Both processes help answer questions related to the state and trends of ES for society, the drivers that affect them and the priorities of conservation and restoration strategies (Maes et al., 2012b). The mapping and modelling methods are varied, depend on the research objectives, type of service and working scale, and are still undergoing development and validation (Nedkov and Burkhard, 2012). The current trends in mapping and modelling include research on biodiversity, species functions, habitat structure, ecosystem processes, ecological production functions and landscape function analyses (Lavorel and Grigulis, 2012; Maes et al., 2012a; Nedkov and Burkhard, 2012; Bastian, 2013). Generally, information on ES mapping and modelling focuses on the distribution of services, especially food and water provision and regulation (UNEP-WCMC, 2011), and the majority of this research consists of studies conducted at different spatial scales (Maes et al., 2012a). China, South Africa, USA, the Netherlands and Australia are the countries in which most of the research is conducted for this subject (Egoh et al., 2008, 2012; Costanza and Kubiszewski, 2012). Ecosystem functions, such as inputs for mapping, are generally linked through models or indicators from the primary data by relating these functions to maps of land use/cover, eco-regions or habitat maps (Burkhard et al., 2012; Haines-Young et al., 2012; Maes et al., 2012b). The capacity of the ecosystem capacity to supply certain ES and the actual usage by people varies considerably (Bastian, 2013) because ecosystem functions depend mainly on biophysical conditions and land use (de Groot et al., 2010; Burkhard et al., 2012). In Latin America (LA), between the years 2001 and 2004, the Millennium Ecosystem Assessment initiative catalysed the increase in studies related to ES. Chile, Costa Rica, Colombia, Perú, Argentina and Brazil have conducted sub-global evaluations (MA, 2005), with differences among the studies that reflect each country’s context, needs and pressures regarding their natural resources. In the region, studies have focused on quantifying carbon- and waterrelated services, and the payment for ecosystem services (PES) has received considerable attention (Corbera et al., 2007; Pagiola et al., 2007; Quintero et al., 2009), but trade-offs analyses are scarce (Balvanera et al., 2012). The review by Balvanera et al. (2012) on the state of knowledge on this subject is noteworthy, concluding that there are imbalances in the focus of interest and information availability for each country and that the diversity of ecosystems and people in LA should be taken into consideration for interventions and future scenarios. However, most countries have official information that can be the basis for a national inventory of their ES and prioritise ES conservation areas. In Colombia, research on ES is limited, and with the exception of the first evaluation of ES within the MA (2005) study (Armenteras et al., 2005), current interest revolves around PES related to hydrological services and biodiversity and pastoral systems (Murgueitio et al., 2011; Moreno-Sanchez et al., 2012). Studies focusing on mapping and trade-offs are also scarce in Colombia (but see Tallis et al., 2012). Recently, Colombia has incorporated ES in its biodiversity policy (MADS, 2012), the actions of which should be based on the knowledge and availability of spatial information on the state and trends of the ecosystems and ES at different scales for decision-making. Herein, we present the first national mapping of ES in Colombia, which is a country with high geographic heterogeneity and several sources of environmental information that are often underused by decision makers. The general objective of this study is to carry out an evaluation of ecosystem services in Colombia for setting priorities and planning carbon and water resource management. We have based this study on two levels of analysis: five natural regions and 41 watersheds. The study is interesting because it shows the suitability of using available official information and how to use it as a surrogate to quickly map ES in tropical countries where data are

often collected at different scales and are limited in its availability. Currently, attention on ES is focused on tropical countries. We have formulated the following questions: (1) What is the spatial distribution for ES? (2) To what extent are ES correlated on both national and watershed levels? (3) What geographical areas maximise the different ES production (i.e., what are the ES hotspots)? Using available public information related to water resources, forests, soils and the use of geographic information tools, we have evaluated the following five ES in Colombia: water provision, regulation of water flow, carbon storage in the above-ground biomass and in the soil and finally landslide prevention. Methods Study area Colombia is located in north-western South America and is considered to be a mega-diverse country, with 34 different biomes and 132 natural ecosystem types (IDEAM et al., 2007). The country covers approximately 1,142,000 km2 in continental area and has five natural regions that are associated with 41 watersheds (Fig. 1), which are used as units of analysis in this study: the Andean (including the three Andes ranges, the Inter-Andean valleys and the Magdalena-Cauca watersheds), the Caribbean, the Amazonian, the Pacific and the Orinoco. These regions have contrasting hydroclimatic, geomorphologic, topographic, edaphic and land use/cover conditions as well as different levels of socio-economic development (Poveda et al., 2011; Armenteras-Pascual et al., 2011). The country is characterised by a high water yield with fluvial discharge that varies from 100 mm per year into the Caribbean to over 6000 mm per year into the Pacific. The water volume in Colombia is 2084 km3 , which is distributed in five big hydrological regions that are divided into the 41 aforementioned watersheds also used as level of analysis and that are currently monitored (MADS, 2010). The population density of the country is 40 people/km2 , distributed irregularly, with ca. 85% of the population concentrated in urban centres of the Andean region, with 3500 people/km2 , in contrast with the southern and western parts of the country, with less than 1 person/km2 (http://www.dane.gov.co). More than one-third of the territory has been transformed, with the current predominant land use being agriculture, including pasture in the Andean and Caribbean regions, tropical humid forests in the Amazonian and Pacific regions and savannahs in the Orinoquia region. The natural ecosystems are diverse, and although forests covered approximately 57 million hectares in 2005, deforestation rate has been estimated to be 273,334 ha per year between 2000 and 2005 (Cabrera et al., 2011). Moreover, in the past decade Colombia has experienced a five-fold increase in foreign investment, especially in mineral extraction (e.g., oil, carbon and gold) (Banco de la República, 2012), a trend that is foreseen to be maintained and to affect some of the basic services for people because many of these projects are planned to be developed in high biodiversity areas that are strategic for the conservation of water resources. Large projects aiming to produce biofuels are also being developed, together with a continuous encroachment of forests due to the progression of the agricultural frontier in the Amazonian and Orinoquia regions. Methods and databases used to estimate the ES We evaluated the spatial distribution of five ES: water provision, regulation of water flow, carbon storage in the above-ground biomass and carbon in the soil, and landslide prevention. These ES were selected because of their relevance to environmental management and planning and also for the availability of their data for the whole country. Our approach uses the hierarchical classification of

N. Rodríguez et al. / Land Use Policy 42 (2015) 609–618

611

Fig. 1. Map locating the study area including watershed boundaries.

ES proposed by The Economics of Ecosystems and Biodiversity initiative (TEEB) (de Groot et al., 2010; UNEP-WCMC, 2011). According to this classification, the categories used were provisioning (water) and regulating (regulation of water flow, climate regulation including carbon storage and moderation of extreme events including landslide prevention) (Table 1). Water provision (WP) The provision of fresh water has been identified as a fundamental and non-replaceable service for human well-being, food production and economic development, as well as for the maintenance of biodiversity (de Groot et al., 2002, 2010; Pert et al., 2010; Nedkov and Burkhard, 2012). Additionally, the economies of many sectors, such as agriculture, industry and tourism, depend on this service (Van Jaarsveld et al., 2005; Naidoo et al., 2008). This service is highly dependent on the quantity and distribution of precipitation combined with a series of abiotic factors, such as regional climatic systems and topography (de Groot et al., 2002). We used a map of the water yield of Colombia for the mean

annual conditions (IDEAM, 2010). This map provided data at a spatial resolution of 2 km and represented the superficial water quantity per unit catchment surface area for a certain time scale (runoff = precipitation − real evapotranspiration). The range was 0.1737–318.922 L/s km2 .

Regulation of water flow (RWF) RWF refers to the influence of natural systems on the hydrological fluxes at the surface of the Earth (Egoh et al., 2008) and includes processes associated with irrigation maintenance, natural drainage and buffering stream flow extremes (de Groot et al., 2002). We used the information of the retention and hydrological regulation index – IRH (IDEAM, 2010) at a 2-km spatial resolution. This index (IRH = Vp/Vt) was estimated from the relationship between the area below the mean flow line (Vp), and the corresponding area below the daily flow duration curve (Vt), and its range was 0.39–0.89. It is an adimensional index.

612

N. Rodríguez et al. / Land Use Policy 42 (2015) 609–618

Table 1 Summary of the ecosystem services considered and dataset sources. Service category

Service types

Biophysical component or process

State indicator

Range of value of service

Provisioning

Water provision

Water availability (L/s km2 )

0.1737–318.922 L/s km2 National Water Study (IDEAM, 2010)

Regulating

Regulation of water flows Climate regulation – carbon storage in above-ground biomass

The hydrologic cycle (precipitation, evapotranspiration, vegetation, soil and plant-available water content) Role of forests in water infiltration and gradual release of water Influence of ecosystems on local and global climate. Capacity that different land covers have to fixate carbon in their structures

Water retention Index Above-ground biomass (t/ha)

0.39–0.89

Represents the process of soil formation linked to the accumulation of organic matter Influence of climate, soil type and geomorphology Role of vegetation has for slowing down the process of land degradation through soil retention Depends of topography, geomorphology, soil and land cover characteristics

% Soil organic carbon

<1.0% to >6.0%

Landslide susceptibility

5 categories: 0 (very low) to 5 (very high)

Climate regulation – carbon storage in soil

Landslide prevention

Landslide prevention (LP) This factor refers to the capacity that vegetation has for slowing down the process of land degradation through soil retention, and it evaluates the function of vegetation cover and erosion ability (Egoh et al., 2008). We used the national landslide susceptibility map as a proxy to evaluate this service (Sánchez et al., 2010). This map analysed geological, geomorphologic and edaphic variables and vegetation cover, identifying areas that should be managed for hazard prevention in the face of erosion and mass wasting processes (LP = 0.15L + 0.15Df + 0.1Tm + 0.1Dd + 0.1S + 0.15P + 0.15Ie + 0.1Cv/Np, where L is the lithology, Df is the density of fracturing, Tm is the morphology, Dd is the drainage density, S is the soil, P is the slope, Ie is the intensity of erosion, Cv is the land cover and Np is the number of parameters). In accordance with expert criterion, five degrees of susceptibility were established for this map: null, low, medium, high and very high; the map scale was 1:500,000.

Carbon storage in the above-ground biomass (CAGB) This factor refers to the capacity that different land covers have to fix carbon in their structures and remove CO2 from the atmosphere, which contributes to climate change (Bai et al., 2011). For the forest category, we used a map of above-ground biomass (AGB). The estimates were realised for 16 types of natural forest using allometric models of AGB (Phillips et al., 2011; Alvarez et al., 2012), with biomass ranging from 96.2 t/ha to 295.1 t/ha. The carbon-fraction is calculated by conversion factor of 0.5 AGB. For shrub and/or herbaceous vegetation, we used values provided in the literature (IDEAM, 2009; Etter et al., 2011), and for the rest of the land cover types, we used estimates reported for Colombia by Yepes et al. (2011). The map has a spatial resolution of 270 m.

Carbon storage in the soil (SoilC) This factor represents the process of soil formation linked to the accumulation of organic matter (Egoh et al., 2008). The map of organic carbon distribution in Colombian soils was used, which was based on the work on soils by the Agustín Codazzi Geographic Institute (Instituto Geografico Agustín Codazzi), modified by the IDEAM. This map incorporated information on geomorphology, climate, physical and chemical soil properties and soil taxonomic classes

3.5–147.5 tC/ha

Source

National Water Study (IDEAM, 2010) Map of carbon stored in above-ground biomass in the forest of Colombia (Phillips et al., 2011), Land Use Changes (1970–2020) and the Carbon Emissions in the Colombian Llanos (Etter et al., 2011) Report from Colombia. United Nations Framework Convention on Climate Change (UNFCC), (IDEAM, 2001) The National landslide susceptibility map (IDEAM, 2010)

for each soil unit of the country (IDEAM, 2001). The map scale was 1:500,000 and the values ranged from <0.1% to >6%. Land cover map We used a land cover map as a basis for quantifying and mapping ES, which is an appropriate approach for the national level and the data availability (Haines-Young et al., 2012). We also considered regions and watersheds as the central levels for analyses. In particular, we used the CORINE land cover map for 2005–2009 (scale of 1:100,000) that, for Colombia, was based on the interpretation of Landsat and Spot satellite images and on validation in the field (IDEAM et al., 2012). For the purposes of the present study, we focused on continental covers using the second hierarchical level that has 14 cover classes: urban fabric; industrial; commercial and transportation; mine, dumping and construction sites; artificial non-agricultural vegetated areas; arable land; permanent crops; pastures; heterogeneous agricultural areas; forest, shrub and/or herbaceous vegetation associations; open spaces with little or no vegetation; inland wetlands; coastal wetlands; and inland waters. For the forest category, we used a map of the aboveground biomass (AGB) (Phillips et al., 2011), the map was created from the forest/non-forest map from the national REDD Project combined with stratification of 16 natural forest types. The maps have a spatial resolution at 270 m. ES mapping One of the simplest methods for general ES mapping is to combine resource availability with an assessment of the vegetation associated with availability as a proxy for assessing the presence of ES rather than the service. This approach has been applied in other regions (Egoh et al., 2008, 2009) and has been frequently used to map ES when data availability is limited and the focus is on the presence of ES (Maes et al., 2012a). In this study we also used this approach given the fact that different land covers generate a gradient of conditions in terms of structures and processes necessary to provide ecosystem services and thus define the potential of delivering them (Maes et al., 2012b). We also assumed not only that natural ecosystems had good capacities to supply ES but also that as

N. Rodríguez et al. / Land Use Policy 42 (2015) 609–618 Table 2 Weights assigned to ecosystem services in relation with the different land cover classes. A value 0 = no relevant capacity to supply services and a value 1 = highest capacity to supply services. CLC classes

RWF

WP

LP

SoilC

Urban fabric Industrial, commercial and transportation Mine, dump and construction sites Artificial non-agricultural vegetated areas Arable land Permanent crops Pasture Heterogeneous agricultural areas Forest Shrub and/or herbaceous vegetation associations Open spaces with little or no vegetation Inland wetlands Coastal wetlands Inland waters

0 0

0 0

0 0

0 0

0 0.2

0 0.2

0 0.2

0 0.2

0.2 0.6 0.5 0.6 1 1

0.2 0.6 0.6 0.6 1 1

0.4 0.6 0.4 0.6 1 1

0.2 0.6 0.4 0.5 1 1

0

0

0.2

0.2

1 1 1

1 1 1

1 1 1

1 1 1

Indicator values assigned to every land cover class: regulation of water flow (RWF), landslide prevention (LP) derived from literature (Burkhard et al., 2012; Nedkov and Burkhard, 2012; Koschke et al., 2012). Water provision (WP) and carbon storage in the soil (SoilC) derived from expert criteria. For carbon storage in above-ground biomass (CAGB) the values of ecosystems services were reported of Colombia.

the extent of human impact increased, the services supply capacity of the land cover varied strongly and tended to decrease (de Groot et al., 2010; Haines-Young et al., 2012). We defined a scale ranging from 0 to 1 to relate each service studied here with the different land cover types based on previous studies of the ES that regulate water flow and prevent landslides (Burkhard et al., 2012; Nedkov and Burkhard, 2012; Koschke et al., 2012). We also used expert local knowledge to generate consistent values for water provision and carbon storage in the soil. A consensus was reached after three expert workshops with national experts on hydrology, soils and environmental issues (Table 2). A value of 0 indicates that there is no relevant capacity to supply services, and a value of 1 indicates the highest capacity to supply these services. We generated a map for each ecosystem service by multiplying the original value from the service’s dataset by the value given as a function of the land cover type (from 0 i.e., land cover not relevant for a service to 1 i.e., maximum contribution by a specific cover to maintain a service). The ecosystem service maps were rasterised and converted to a 270 m spatial resolution. As explained above, the CAGB map was a generated as combination of available biomass and carbon information for the entire country (Phillips et al., 2011). Analysis To generated results that can be used at different planning levels, we evaluated the ES using two analyses levels: (i) the five natural regions of the country and (ii) the 41 watersheds. To facilitate the statistical analyses, we divided the country into 10 × 10 km grid (for

613

a total of 11,537 cells) using the geographic information system GIS package ArcGIS 10.1 (ESRI) for spatial analyses. Each cell value represented the ecosystem service value (an average weighted value calculated using Spatial Analyst in ArcMap 10 (ESRI)) with respect to the land cover type and the area of the cell. We identified hotspots for each service using the following definition of hotspot: ‘an area that provides large components of a particular service’ and the criteria proposed by Bai et al. (2011), i.e., the richest 10% of grid cells, to obtain the hotspot maps. With a superposition process using individual maps, we obtained a total hotspot map. The relationships between pairs of ES at the watershed (41 cases) and national (11,537 quadrats) levels were conducted using Pearson correlation analyses. Two variables, CAGB and FM, were square root-transformed prior to these analyses to reach normality. A principal component analysis (PCA) was performed to evaluate the relationship among the different ES using the values from the 41 watersheds. The objective of this analysis was to reduce the number of variables and generate new axes that best explained the variance in the data and illustrated the relationship among the ES. The STATISTICA (StatSoft, 2001) software was used for all statistical analyses.

Results The results show that there were differences in the spatial distribution of the ES among regions and that each region was associated with maximising the supply of one or two ES. In general, the country had high mean values of water yield (62.5 L/s km2 ) and CAGB (69.3 tC/ha). The Pacific and Amazonian regions present the highest mean values for the services of water provision (125.8 and 71.9 L/s km2 , respectively), regulation of water flow (>0.72 for both) and CAGB (60.3 and 120.9 t/ha, respectively), with the Caquetá, Putumayo and Amazonas-direct watersheds standing out the most; the Andean region was important for LP (>2.5) and maintaining the percentage of carbon in the soil (>0.03%) (Table 3). The ecosystems from the Orinoquia region present medium value for landslide prevention, whereas the Caribbean region presented low values for all ES (Table 3). At the watershed level, all the watersheds from the Amazonian region stood out with the highest mean values for services related to water and CAGB, and the Apure (Orinoquia), San Juan and Pacific-direct (Pacific) watersheds were important for the services of soil organic carbon and LP. Fig. 2 illustrates the distribution of the five services studied, indicating the range of values at the national scale. The highest values for the regulation of water flow service (>0.53) are distributed irregularly across the majority of the country (except for the Caribbean and Orinoquia regions) (Fig. 2A), and the highest values for water provision (>90 L/s km2 ) were found in most of the Pacific region and in the upper Río Caquetá and Putumayo watersheds (Amazonian region) (Fig. 2B). The areas with the highest values for Landslide prevention (<3.7%) corresponded to the boundary (piedmont) of the Andean region with the lowlands (Fig. 2C), mainly the upper parts of the Caquetá, Putumayo, Guaviare (Amazonas), Meta, Casanare (Orinoquia), San

Table 3 Average values for six ecosystem services studied at the regional scale and percentage of each ecosystem service hotspot within the region. Region

Caribbean Pacific Andes Orinoco Amazon National average

RWF

WP

LP

CAGB

SoilC

Mean

% Hotspot

Mean

% Hotspot

Mean

% Hotspot

Mean

% Hotspot

Mean

% Hotspot

0.4 0.7 0.5 0.5 0.7 0.59

6.1 7.7 18.8 16.0 51.4

21.1 125.8 43.6 46.6 71.9 62.45

0.0 76.2 23.8 0.0 0.0

1.3 2.2 2.53 2.59 1.58 1.76

7.67 5.79 54.8 30.8 1.0

21.1 60.4 39.5 28.4 120.9 69.3

2.6 1.6 5.9 6.6 83.3

0.01 0.02 0.03 0.01 0.01 0.02

0.6 13.2 84.9 0.5 0.8

Regulation of water flow (RWF), Water provision (WP), Landslide prevention (LP), Carbon storage in above-ground biomass (CAGB) and Carbon storage in the soil (SoilC).

614

N. Rodríguez et al. / Land Use Policy 42 (2015) 609–618

Fig. 2. Distribution spatial of the five ecosystem services, indicating the range of values at the national scale. The black represents the hotspots.

Juán and Micay-direct (Pacific) watersheds. Important areas in terms of CAGB were identified, with the exception of the Amazonian region, as some areas of the Andes (values >132.05 tC/ha) that coincided with forest ecosystems within the National Park System of Colombia (Fig. 2D). Finally, the highest values for the service of carbon storage in the soil (>0.66%) were limited to the Andes and some areas of the Pacific region (Fig. 2E).

Between 39% and 43% of the country’s area contained hotspots associated with the RWF and carbon CAGB, respectively. These services were concentrated in the Pacific and Amazonian regions. The remaining ES have a lower area hotspots (<16%), where the hotspots corresponding to water provisions occupied the lower percentage (<1%), even though the mean values for this service were high across the country. The Andean region contained approximately 85% of

N. Rodríguez et al. / Land Use Policy 42 (2015) 609–618

615

Fig. 3. Map of ecosystem services hotspots overlap. Shown the number of overlapping ecosystem services in an analysis levels.

the hotspots corresponding to soil carbon content, and 85% of the LP hotspot was concentrated in these two regions. The Caribbean region had the lowest percentage of hotspot area (Table 3). For the case of the superposition of different hotspots for Colombia, we found that approximately 33% of the areas had no hotspots, 15.8% had one hotspot and approximately 46% contained areas where two ES converge (Fig. 3). The latter areas were mainly located in the watersheds of the Amazonian and Pacific regions, where the ES of water provision, RWF and CAGB were correlated. Areas where four ES were superimposed were scarce (less than 1% of the coun˜ try) and were located in the Andean (Alto Magdalena and Saldana watersheds) and Pacific (San Juán and Mira watersheds) regions. The correlation between the different services was variable (Table 4A and B), and higher correlations were found when the analysis was conducted at the national level. In general, we found that the highest correlations were between the ES of regulation of water flow and CAGB and between the former and water provision, with higher correlation values at the watershed level (Table 4A and B). Notably at the country level, the landslide prevention ES showed

Table 4 Correlation between the different ecosystems services (A) national scale and (B) watersheds scale. Values in bold indicate significance (p < 0.01). RWF (A) National scale 1 RWF 0.74 WP 0.13 LP 0.64 CAGB 0.05 CSUELO (B) Watersheds scale 1 RWF 0.92 WP 0.25 LP 0.77 CAGB 0.15 CSUELO

WP

LP

CAGB

SoilC

1 0.19 0.78 0.12

1 0.10 0.39

1 0.02

1

1 0.27 0.85 0.28

1 0.07 0.58

1 0.03

1

616

N. Rodríguez et al. / Land Use Policy 42 (2015) 609–618

Fig. 4. Principal Component Analyses at regional and watershed levels.

a positive correlation with all ES. At the watershed scale, soil carbon was related positively to LP. The PCA conducted to establish joint relationships among all ES showed that the first two components explained approximately 86% of the variance and a very clear distribution of the watersheds in the plane according to the natural regions to which they belonged. According to axis I (57.4% of the variance explained), we separated the watersheds belonging to the Amazonian and Pacific regions in the right corner of the axis, those from the Andean and Orinoquia regions towards the centre and those from the Caribbean region to the left. The right side of axis I is characterised by high values for CAGB, WRF and water provision (Fig. 4). According to axis II, we separated watersheds from the Pacific and Andean regions into the upper section and watersheds from the Amazonian and Orinoquia regions and, to lesser extent, those from the Caribbean region into the lower section. This separation was performed because high values of soil carbon and landslides were found in the upper section of the axis. In general terms, for the ES considered in this study, the country could be divided into three groups: Group I would consist of watersheds from the Amazonian region, where the services of CAGB, RWF and water provision were maximised; Group II contained the watersheds from the Pacific and some from the Andean regions, where the maximised ES were landslide prevention and soil carbon; and Group III would consist of the watersheds from the Orinoquia and Caribbean and some from the Andean regions, which have no specific associated ES. Discussion Colombia plays an important role at the regional level for the supply of ES related to water and carbon due to its geographic, topographic and hydroclimatic characteristics (Poveda et al., 2011) and because more than 50% of the territory is covered by tropical forests. The total water yield of the country (63 L/s km2 ) exceeds the mean global water yield (10 L/s km2 ) by approximately six times

(IAvH et al., 2011), and the CAGB for 2010 was estimated to be 7,144,861,815 tC (Yepes et al., 2011). There are differences in the distribution and values of the ES analysed that reflect, on the one hand, the natural capacity of the different ecosystems to supply services with particular hydrologic, climatic, topographic, edaphic and land cover conditions. On the other hand, these differences also reflect the degree of transformation in many regions, where the capacity to supply or regulate the analysed services is decreasing and other services, such as crop production, are relevant. This is the case of the Caribbean region and part of the Andean region. This finding is in accordance with the hypotheses of Braat and Brink (2008), Haines-Young (2009) and de Groot et al. (2010), who suggested that land use is a key factor in the loss of certain ES and that the gradient and intensity of use would favour other ES. The supply pattern of the CAGB and RWF services reflects the distribution of forest ecosystems in the country, concentrating, as expected, in the Amazonian and Pacific regions and more sparsely in areas of the Andean region where some National Parks are located (such as Paramillo and Sierra Nevada de Santa Marta). The spatial distribution of these services, particularly carbon, is in accordance with the findings at the global scale by Naidoo et al. (2008) and, at the country level, by Tallis et al. (2012), who also reported low values of carbon to the west of the Andes and for the Caribbean region, relating these values to population density. The watersheds from the Pacific region and those with headwaters in the Andes and that drain to the Amazon and Orinoquia rivers have large water provision values. This may be associated with the incidence of hydroclimatic conditions originated by atmospheric circulation patterns from the Pacific Ocean and Caribbean Sea, the dynamic hydroclimatic systems from the Amazonian and Orinoquia regions and topographic gradients (Poveda et al., 2011). The IDEAM (2010) reported that these areas were considered to have the highest water yields and to be providers of water for the country (ca. 70%) and that many of the streams are direct tributaries

N. Rodríguez et al. / Land Use Policy 42 (2015) 609–618

of the Amazon and Orinoco Rivers, responsible for the majority of the water in South America (Restrepo et al., 2006). In contrast, the Caribbean region presents low values for ES associated with water resources and is vulnerable to water shortages (IDEAM, 2010), with the associated social implications of such shortages. The piedmont areas and the Andean slopes, which are covered generally with forests, are important for the service of landslide prevention, and their spatial distribution coincides with that given in the work by Tallis et al. (2012). However, there are low to intermediate values for this service in the upper and mid-altitude areas of the watersheds that drain to the Orinoco and Amazon rivers and for watersheds that drain to the mid-section of the Magdalena River. These watersheds, similar to other tropical mountain zones of South America, are particularly sensitive to erosion, landslides (Vanacker et al., 2005) and excessive sediment transport (Restrepo et al., 2006; Molina et al., 2008). This is in part due to steep topographic gradients, high precipitation, land use changes and poor agricultural management practices. The Andean region also concentrates the highest values of ES associated with soil carbon, particularly the central mountain range, and this association may be related to the soil type, which is generally derived from volcanic ash, the climatic conditions and the lithology. The analysed ES showed a positive correlation in both analysis levels, such as among the ES of water provision, RWF and CAGB (r > 0.74), presenting a spatial overlap in the low areas of the Amazonian and Pacific regions and reflecting in part the distribution of the lowland forest ecosystems of the country. It is the advisable and possible to guarantee the permanence of these three ES with national and regional policies aimed at reducing deforestation and with planning tools such as the declaration of figures of protection at the regional scale. This correlation contrasts with the negative relationship observed and the lack of spatial overlap between the ES of LP (Andean), explained by the topographic gradient. The additional insight provided by the PCA corroborated the type of correlations showing that the ES may be grouped for their management into three groups, where various services concurrently occur spatially in one region and watershed, thereby guiding both ecosystem conservation and restoration actions. The high degree of spatial overlap between hotspots for the three ES that were correlated most strongly (WP, RFW and CAGB) and the low overlap between the hotspots for more than four ES (<1% of the country) suggest particular management strategies. The protection of areas with a high number of hotspots, such as transition areas (piedmonts) between the Andes and lowlands of the Amazonian, Orinoquia and Pacific regions and some areas of the Andean region (e.g., Colombian Massif and mountain tops), would be the most efficient and urgent management option. These areas have high ecosystem diversity and landscapes critical to the natural supply of ES but, at the same time, have been identified as hotspots for deforestation; under scenarios of land use change, these areas may reduce the supply of important ES, such as biodiversity and water and climate regulation (Rodríguez Eraso et al., 2012), and hydrological functioning may be affected in processes such as stream flow, evapotranspiration and infiltration (Molina et al., 2012). Nevertheless, regions such as the Orinoco and the Caribbean should be evaluated with another set of ES, as those related to flood control, where likely they will present the highest values. Finally and despite some of the limitations raised above, our study contributes to the identification and synergies of the most important areas for the conservation of ES using available national information. These results could be incorporated during the implementation of national policies such as the National Policy for the Management of Biodiversity and Ecosystem Services or future Reducing emissions from deforestation and forest degradation related policies. Further these results can help assess the spatial

617

consequences of different policy scenarios related to land use change, mining and large-scale agro-industrial developments currently being a hot topic in the country. Following the same trend, the watershed level results of our study may guide the implementation of the current very comprehensive policy on water resources management. Indeed our results give support to the prioritisation for watershed management in the country that has yet to be defined. Thus, the grouping of watersheds as a function of the services they supply may be used as a basis for environmental planning. Similarly, the fact that important areas for supply coincide with protected areas may be considered to be a win–win situation among conservation and ES and to present opportunities to revise Colombia’s conservation priorities. Acknowledgments The authors would like to acknowledge the support of the Institute of Hydrology, Meteorology and Environmental Studies (IDEAM) and the Geographic Institute Agustín Codazzi (IGAC) for facilitating access to the data used in the present study. We also acknowledge the collaboration of Luz Marina Arévalo, director of Ecosystems and Environmental Data within the IDEAM and all the professionals working under this section and the Hydrology section of the IDEAM in the establishment of the criteria to evaluate the ES. We thank the Vice-Rector of Research from the National University of Colombia for funding the study through the postdoctoral position of Nelly Rodríguez and Research Groups Strengthening Grants and Carol Franco for her help with the graphics. We also thank the CYTED network IBEROREDD+ for financial support of the research. References Alvarez, E., Duque, A., Saldarriaga, J., Cabrera, K., de las Salas, G., del Valle, I., Lema, A., Moreno, F., Orrego, S., Rodríguez, L., 2012. Tree above-ground biomass allometries for carbon stocks estimation in the natural forests of Colombia. Forest Ecol. Manag. 267, 297–308. Armenteras, D., Rincón, A., Ortiz, N., 2005. Ecological Function Assessment in the Colombian Andean Coffee-growing Region, Sub-global Assessment Report. Millennium Ecosystem Assessment. Armenteras-Pascual, D., Retana-Alumbreros, J., Molowny-Horas, R., Roman-Cuesta, R.M., Gonzalez-Alonso, F., Morales-Rivas, M., 2011. Characterising fire spatial pattern interactions with climate and vegetation in Colombia. Agric. Forest Meteorol. 151, 279–289. Armstrong, C.W., Foley, N.S., Tinch, R., Van den Hove, S., 2012. Services from the deep: steps towards valuation of deep sea goods and services. Ecosyst. Serv. 2, 2–13. Armsworth, P.R., Chan, K.M., Daily, G.C., Ehrlich, P.R., Kremen, C., Ricketts, T.H., Sanjayan, M., 2007. Ecosystem-service science and the way forward for conservation. Conserv. Biol. 21, 1383–1384. Bai, Y., Zhuang, C., Ouyang, Z., Zheng, H., Jiang, B., 2011. Spatial characteristics between biodiversity and ecosystem services in a human-dominated watershed. Ecol. Complex. 8, 177–183. ˜ Balvanera, P., Uriarte, M., Almeida-Lenero, L., Altesor, A., Declerck, F., Gardner, T., Hall, J., Lara, A., Laterra, P., et al., 2012. Ecosystem services research in Latin America: the state of the art. Ecosyst. Serv. 2, 56–70. Banco de la República, 2012. Informe de la Junta directiva al Congreso de La República , Bogotá, DC. Bastian, O., 2013. The role of biodiversity in supporting ecosystem services in Natura 2000 sites. Ecol. Indic. 24, 12–22. Braat, L., ten Brink, P. (Eds.), 2008. The Cost of Policy Inaction. The Case of Not Meeting the 2010 Biodiversity Target. Study for the European Commission, DG Environment, Wageningen. Braat, L.C., de Groot, R.S., 2012. The ecosystem services agenda: bridging the worlds of natural science and economics, conservation and development, and public and private policy. Ecosyst. Serv. 1, 4–15. Bryan, B., Grandgirard, A., Ward, J.R., 2010. Quantifying and exploring strategic regional priorities for managing natural capital and ecosystem services given multiple stakeholder perspectives. Ecosystems 13, 539–555. Burkhard, B., Kroll, F., Nedkov, S., Müller, F., 2012. Mapping ecosystem service supply, demand and budgets. Ecol. Indic. 21, 17–29. Cabrera, E., Vargas, D.M., Galindo, G., García, M.C., 2011. Memoria Técnica: Cuantificación de la tasa de Deforestación para Colombia, Periodo 1990–2000, 2000–2005. Instituto de Hidrología, Meteorología y Estudios Ambientales – IDEAM, Bogotá, DC.

618

N. Rodríguez et al. / Land Use Policy 42 (2015) 609–618

Corbera, E., Kosoy, N., Martinez, M., 2007. Equity implications of marketing ecosystem services in protected areas and rural communities: case studies from Meso-America. Global Environ. Change 17, 365–380. Costanza, R., 2008. Ecosystem services: multiple classification systems are needed. Biol. Conserv. 141, 350–352. Costanza, R., Kubiszewski, I., 2012. The authorship structure of ecosystem services as a transdisciplinary field of scholarship. Ecosyst. Serv. 1, 16–25. Daily, G.C., Matson, P., 2008. Ecosystem services: from theory to implementation. Proc. Natl. Acad. Sci. U. S. A. 105, 9455–9456. de Groot, R.S., Wilson, M., Boumans, R.M., 2002. A typology for the classification, description and valuation of ecosystem functions, goods and services. Ecol. Econ. 41, 393–408. de Groot, R.S., Alkemade, R., Braat, L., Hein, L., Willemen, L., 2010. Challenges in integrating the concept of ecosystem services and values in landscape planning, management and decision making. Ecol. Complex. 7, 260–272. Egoh, B., Rouget, M., Reyers, B., Knight, A.T., Cowling, R.M., Van Jaarsveld, A.S., Welz, A., 2007. Integrating ecosystem services into conservation assessments: a review. Ecol. Econ. 63, 714–721. Egoh, B., Reyers, B., Rouget, M., Richardson, D.M., Le Maitre, D.C., van Jaarsveld, A.S., 2008. Mapping ecosystem services for planning and management. Agric. Ecosyst. Environ. 127, 135–140. Egoh, B., Reyers, B., Rouget, M., Rouget, M., Richardson, D.M., 2009. Spatial congruence between biodiversity and ecosystem services in South Africa. Biol. Conserv. 142, 553–562. Egoh, B., O’Farrell, P.J., Charef, A., Gurney, L.J., Koellner, T., Nibam Abi, H., Ego, M., Willemen, L., 2012. An African account of ecosystem service provision: use, threats and policy options for sustainable livelihoods. Ecosyst. Serv. 2, 71–81. Etter, A., Sarmiento, A., Romero, M., 2011. Land use changes (1970–2020) and the carbon emissions in the Colombian Llanos. In: Hill, J.M., Hanan, N.P. (Eds.), Ecosystem Function in Savannas: Measurement and Modeling at Landscape to Global Scales. Taylor & Francis, CRC Press, pp. 383–402, http://www.crcpress.com/ product/isbn/9781439804704 Farrell, P.J.O., Anderson, P.M.L., 2010. Sustainable multifunctional landscapes: a review to implementation. Curr. Opin. Environ. Sustain. 2, 59–65. Fu, B., Wang, S., Su, C., Forsius, M., 2012. Linking ecosystem processes and ecosystem services. Curr. Opin. Environ. Sustain. 5, 1–7. Haines-Young, R., 2009. Land use and biodiversity relationships. Land Use Policy 26S, 178–186. Haines-Young, R., Potschin, M., Kienast, F., 2012. Indicators of ecosystem service potential at European scales: mapping marginal changes and trade-offs. Ecol. Indic. 21, 39–53. IAvH, IDEAM, IIAP, INVEMAR, SINCHI, 2011. Informe del Estado del Medio Ambiente y de los Recursos Naturales Renovables 2010. Instituto de Hidrología, Meteorología y Estudios Ambientales – IDEAM, Bogotá, DC. IDEAM, IGAC, IAvH, INVEMAR, SINCHI, IIAP, 2007. Ecosistemas continentales, costeros y marinos de Colombia. Bogotá DC. IDEAM, 2001. Mapa de la distribución del carbono orgánico en los suelos de Colombia. Instituto de Hidrología, Meteorología y Estudios Ambientales, Bogotá, DC. IDEAM, 2009. Caracterización preliminar del ciclo del carbono a partir de la información existente. Quinto Informe. Proyecto Piloto Nacional de Adaptación al ˜ Instituto de Hidrología, Cambio Climático–INAP. Componente B- Alta Montana. Meteorología y Estudios Ambientales, Bogotá, DC. IDEAM, 2010. Estudio Nacional del Agua. Instituto de Hidrología, Meteorología y Estudios Ambientales, Bogotá, DC. IDEAM, MADS, IGAC, IIAP, SINCHI, PNN, 2012. Capa Nacional de Cobertura de la Tierra (periodo 2005–2009): Metodología CORINE Land Cover adaptada para Colombia, escala 1:100.000 – V1.0. Bogotá DC. Koschke, L., Fürst, C., Frank, S., Makeschin, F., 2012. A multi-criteria approach for an integrated land-cover-based assessment of ecosystem services provision to support landscape planning. Ecol. Indic. 21, 54–66. Kroll, F., Müller, F., Haase, D., Fohrer, N., 2012. Rural–urban gradient analysis of ecosystem services supply and demand dynamics. Land Use Policy 29, 521–535. Lavorel, S., Grigulis, K., 2012. How fundamental plant functional trait relationships scale-up to trade-offs and synergies in ecosystem services. J. Ecol. 100, 128–140. MA, 2005. Ecosystems and Human Well-Being: Synthesis Millennium Ecosystem Assessment. Island Press, Washington, DC. Mace, G.M., Norris, K., Fitter, A.H., 2012. Biodiversity and ecosystem services: a multilayered relationship. Trends Ecol. Evol. 27, 10–26. MADS, 2010. Política Nacional para la Gestión Integral del Recurso Hídrico. Ministerio de Ambiente y Desarrollo Sostenible, Bogotá, DC. MADS, 2012. Política Nacional para la Gestión Integral de la Biodiversidad y sus Servicios Ecosistémicos (PNGIBSE). Ministerio de Ambiente y Desarrollo Sostenible, Bogotá, DC. Maes, J., Egoh, B., Willemen, L., Liquete, C., Vihervaara, P., Grizzetti, B., Drakou, E.G., Zulian, G., Philipp, J., et al., 2012a. Mapping ecosystem services for policy support and decision making in the European Union. Ecosyst. Serv. 1, 31–39. Maes, J., Paracchini, M.L., Zulian, G., Dunbar, M.B., Alkemade, R., et al., 2012b. Synergies and trade-offs between ecosystem service supply, biodiversity, and habitat conservation status in Europe. Biol. Conserv. 155, 1–12.

Molina, A., Govers, G., Poesen, J., Van Hemelryck, H., De Bièvre, B., Vanacker, V., 2008. Environmental factors controlling spatial variation in sediment yield in a central Andean mountain area. Geomorphology 98, 176–186. Molina, A., Vanacker, V., Balthazar, V., Mora, D., Govers, G., 2012. Complex land cover change, water and sediment yield in a degraded Andean environment. J. Hydrol. 472–473, 25–35. Moreno-Sanchez, R., Maldonado, J.H., Wunder, S., Borda-Almanza, C., 2012. Heterogeneous users and willingness to pay in an ongoing payment for watershed protection initiative in the Colombian Andes. Ecol. Econ. 75, 126–134. Muradian, R., Rival, L., 2012. Between markets and hierarchies: the challenge of governing ecosystem services. Ecosyst. Serv. 1, 93–100. Murgueitio, E., Calle, Z., Uribe, F., Calle, A., Solorio, B., 2011. Native trees and shrubs for the productive rehabilitation of tropical cattle ranching lands. Forest Ecol. Manag. 261, 1654–1663. Naidoo, R., Balmford, A., Costanza, R., Fisher, B., Green, R.E., Lehner, B., Malcolm, T.R., Ricketts, T.H., 2008. Global mapping of ecosystem services and conservation priorities. Proc. Natl. Acad. Sci. U. S. A. 105, 9495–9500. Nedkov, S., Burkhard, B., 2012. Flood regulating ecosystem services – mapping supply and demand, in the Etropole municipality, Bulgaria. Ecol. Indic. 21, 67–79. Pagiola, S., Ramírez, E., Gobbi, J., Haan, C., Ibrahim, M., Murgueitio, E., Ruíz, J.P., 2007. Paying for the environmental services of silvopastoral practices in Nicaragua. Ecol. Econ. 64, 374–385. Pert, P.L., Butler, J.R., Brodie, J.E., Bruce, C., Honzák, M., Kroon, F.J., Metcalfe, D., Mitchell, D., Wong, G., 2010. A catchment-based approach to mapping hydrological ecosystem services using riparian habitat: a case study from the Wet Tropics, Australia. Ecol. Complex. 7, 378–388. Phillips, J.F., Duque, A.J., Yepes, A.P., Cabrera, K.R., García, M.C., Navarrete, D.A., Álvarez, E., Cárdenas, D., 2011. Estimación de las reservas actuales (2010) de carbono almacenadas en la biomasa aérea en bosques naturales de Colombia. Estratificación, alometría y métodos análiticos. Instituto de Hidrología, Meteorología, y Estudios Ambientales, Bogotá, DC. Poveda, G., Álvarez, D.M., Rueda, Ó., 2011. Hydro-climatic variability over the Andes of Colombia associated with ENSO: a review of climatic processes and their impact on one of the Earth’s most important biodiversity hotspots. Clim. Dyn. 36, 2233–2249. Quintero, M., Wunder, S., Estrada, R.D., 2009. Forest ecology and management for services rendered? Modeling hydrology and livelihoods in Andean payments for environmental services schemes. Forest Ecol. Manag. 258, 1871–1880. Restrepo, J.D., Kjerfve, B., Hermelin, M., Restrepo, J.C., 2006. Factors controlling sediment yield in a major South American drainage basin: the Magdalena River, Colombia. J. Hydrol. 316, 213–232. Rodríguez Eraso, N., Armenteras-Pascual, D., Retana Alumbreros, J., 2012. Land use and land cover change in the Colombian Andes: dynamics and future scenarios. J. Land Use Sci., 1–21, http://dx.doi.org/10.1080/1747423X.2011.650228. Sánchez, R., Urrego, L., Mayorga, R., Vargas, G., 2010. Metodología para la zonificación de susceptibilidad general del terreno a los movimientos en masa. Instituto de Hidrología, Meteorología, y Estudios Ambientales, Bogotá, DC. StatSoft, Inc., 2001. STATISTICA for Windows. StatSoft, Inc, Tulsa, OK. Tallis, H., Polasky, S., Lozano, J.S., Wolny, S., 2012. Inclusive wealth accounting for regulating ecosystem services. In: UNU-IHDP, UNEP (Eds.), Inclusive Wealth Report 2012. Measuring Progress Towards Sustainability. Cambridge University Press, http://www.unep.org/pdf/IWR 2012.pdf TEEB, 2010. In: Kumar, P. (Ed.), The Economics of Ecosystems and Biodiversity. The Ecological and Economic Foundations, London. UNEP-WCMC, 2011. Developing Ecosystem Service Indicators: Experiences and Lessons Learned from Sub-global Assessments and other Initiatives. Technical Series No. 58. Secretariat of the Convention on Biological Diversity, Montréal. Vanacker, V., Molina, A., Govers, G., Poesen, J., Dercon, G., Deckers, S., 2005. River channel response to short-term human-induced change in landscape connectivity in Andean ecosystems. Geomorphology 72, 340–353. Van Hecken, G., Bastiaensen, J., 2010. Payments for ecosystem services: justified or not? A political view. Environ. Sci. Policy 13, 785–792. Van Jaarsveld, A.S., Biggs, R., Scholes, R.J., Bohensky, E., Reyers, B., Lynam, T., 2005. Measuring conditions and trends in ecosystem services at multiple scales: the Southern African Millennium Ecosystem Assessment (SAfMA) experience. Phil. Trans. R. Soc. B 360, 425–441. Viglizzo, E.F., Paruelo, J.M., Laterra, P., Jobbagy, E.G., 2012. Ecosystem service evaluation to support land-use policy. Agric. Ecosyst. Environ. 154, 78–84. Wallace, K.J., 2007. Classification of ecosystem services: problems and solutions. Biol. Conserv. 1, 235–246. Wünscher, T., Engel, S., 2012. International payments for biodiversity services: review and evaluation of conservation targeting approaches. Biol. Conserv. 152, 222–230. Yepes, A., Navarrete, D.A., Phillips, J.F., Duque, A.J., Cabrera, E., Galindo, G., Vargas, ˜ D., García, M., Ordonez, M.F., 2011. Estimación de las emisiones de dióxido de carbono generadas por deforestación durante el periodo 2005–2010. Instituto de Hidrología, Meteorología, y Estudios Ambientales-IDEAM, Bogotá, DC.