Related Biodiversity Vulnerability Assessment in the Tropical Andes

Home , Mountain research, Pacific Slope, Tropical Andes

Science Peaks

Five-tiered integrated climate- related biodiversity vulnerability assessment in the Tropical Andes

Daniel Ruiz Carrascal, Sebastian K. Herzog, Peter M. Jørgensen, Trond H. Larsen, Rodney Martínez, Juan José Nieto, Susan V. Poats, Marcella Ohira

The Tropical Andes are one of the top ing near-term climate change trends, risks for biodiversity at a local scale that biodiversity hotspots on Earth. Long- land-use patterns, biodiversity patterns could subsequently be expanded to other term climate change and rapid land- and gradients, the vulnerability of spe- strategic areas. The overall goal is to sup- use change are both threatening the cies and ecosystems to changes in his- port and guide adaptation measures and integrity and functioning of Andean torical climatic conditions, as well as lo- sustained conservation programs for key ecosystems and thereby the environ- cal perceptions of climate variability and tropical environments. Here we briefly mental goods and services they pro- change in two binational transboundary describe some recent advances of the cli- vide to humans (Herzog et al., 2011). mate analyses and summarize the objec- Whereas numerous global and re- “Thus, an integration of all tives of the main components (climate, gional climate models exist, climate this available information biodiversity, social, land use/cover, and change and vulnerability analyses at and approaches into local outreach and capacity building) of our a local scale – the scale most relevant ongoing research project[1] . to decision makers and land-use plan- scale analyses represents a ners – are virtually non-existent in the novel approach in the Andes. Andes.” The climate component The main objective of the climate com- With regard to biodiversity, large-scale study areas: on the Pacific slope of the ponent[2] is to develop knowledge on patterns and gradients of species rich- northern Andes in the border region be- ness are fairly well established in the re- tween Colombia (Nariño department) 1 The multidisciplinary project “Impacts of climate change on biodiversity in the Tropical Andes: climate- gion for a handful of selected taxonomic and Ecuador (Carchi province), and on related vulnerability assessments and improved groups, but vast knowledge gaps exist at the Amazonian slope of the central An- decision making processes for conservation and land use planning in two Andean biodiversity hotspots” is smaller spatial scales and for the great des in the border region between Bolivia funded by the John D. and Catherine T. MacArthur Foundation through a grant to the Inter-American majority of taxonomic groups. Local bio- and Peru, in the Madidi – Apolobamba Institute for Global Change Research. D. Ruiz is also logical inventory data, where available, – Bahuaja-Sonene – Tambopata protect- partially supported (as in-kind contribution) by the Department of Earth and Environmental Sciences at have not been integrated into multidisci- ed area complex. These regions are re- Columbia University in the City of New York (USA), plinary analyses. Furthermore, although nowned for their exceptional biodiversity the International Research Institute for Climate and Society at Lamont-Doherty Earth Observatory (USA), recent advances in the mapping and clas- and endemism and have been considered and the Antioquia School of Engineering (Colombia). sification of Andean ecosystems repre- key Andean biodiversity hotspots. Re- 2 The following scientists participate as co-PIs: Dr. sent immense progress, the vulnerability sults will then be integrated to pinpoint Mark Cane (Department of Earth and Environmental of these ecosystems to climate change high-risk areas and ecosystems that are Sciences at Columbia University in the City of New York, USA), Dr. Jorge Ignacio del Valle (Universi- can only be crudely guessed in the most particularly vulnerable to the synergistic dad Nacional de Colombia Sede Medellín), Dr. Laia general terms. Thus, an integration of all effects of long-term climatic changes and Andreu Hayles (The Tree Ring Laboratory, Lamont- Doherty Earth Observatory-Columbia University in this available information and approach- land-use change. Our interest is to assist the City of New York, USA), Ángel G. Muñoz (PhD student at Department of Earth and Environmental es into local-scale analyses represents a the four Tropical Andean countries (Bo- Sciences at Columbia University in the City of New novel approach in the Andes. livia, Colombia, Ecuador, and Peru) in York, USA), David Suarez Duque, MSc. (Corporación Grupo Randi Randi-Ecuador), and Remi Cousin (Staff the implementation of a standard meth- Associate at International Research Institute for Cli- In this five-tiered project we are study- odology for estimating climate change mate and Society, Columbia University in the City of

Mountain Research Initiative Newsletter no. 7, 2012 7 Five-tiered integrated climate-related biodiversity vulnerability assessment in the Tropical Andes

Figure 1. 1950-2010 long-term linear trends in ECHAM4.5 mean annual air temperatures (Roeckner et al., 1996) along the longitudinal axis of the Andes Cordillera (see pink 2.8125° grid cells in map panel on the left; location of the ecotransects under study indicated with horizontal arrows), for the latitudinal range of 15°N to 60°S, and for 9 pressure levels (1000, 950, 850, 700, 500, 400, 300, 200, and 100 mb; see x-axis of right panel). Air temperatures are obtained through ECHAM4.5 ensemble simulation runs. Trends are expressed in °C per decade; see color scale on the right. Only statistically significant (at α<0.05) long-term linear trends are displayed; i.e. nonsignificant trends are depicted by white boxes. Black triangles and crosses depict, respectively, the average and maximum altitudes (expressed in atmospheric pressures) of the NOAA NGDC GLOBE gridded 1-km, quality controlled global digital elevation model (Hastings and Dunbar, 1999) in each ECHAM4.5 model grid cell. Areas blocked in grey depict grid boxes below the ground surface. Analyses of ECHAM4.5 simulation runs indicate, in particular, that air temperatures have increased at all latitudes and pressure levels at a rate ranging from +0.03 to +0.40 °C per decade. Between 15°N and 15°S and at higher elevations [100-400 mb], air temperatures have increased at a maximum rate ranging from +0.27 to +0.40 °C per decade. This rate of warming in the upper troposphere in the 15°N to 15°S latitudinal range is 1.8 times greater than that simulated for the lower troposphere over the available 61-year historical period. Note also the differences above and below the tropopause, which is defined as 100 mb at the equator with a linear increase with latitude to 300 mb at the poles.

local climate gradients and to determine cant long-term trends as well as changes Los Nevados Natural Park in the Colom- short- to medium-term climate scenar- in the mean and the variance, and assess- bian Central Andean region (Ruiz et al., ios (10-20 years ahead) by combining ments of spatial and altitudinal patterns. 2012), climate-related risk assessments observed and projected climate trends Spatial scales of analysis include local of highly strategic Andean environments derived from computer modeling, cli- and regional conditions. Analyses of the at local scale are now covering three matological field stations (ground-truth former include case studies in the two bi- study sites in the northern and central data), climate change indices, local Andes. Analyses of regional conditions temperature and humidity records from “With these binational are based on evidence from near-term digital sensors, and reconstructions of regions (...) climate-related historical climate models’ simulation pre-instrumental periods through clas- risk assessments of highly runs (reanalysis data) and comprise the sical dendrochronological techniques full length of the Andes Cordillera and (tree-ring records). Statistical analyses of strategic Andean environments all pressure levels (Figure 1). The group observed and simulated data include Em- at local scale are now cov- is specifically studying long-term trends pirical Orthogonal Functions/Principal ering three study sites in the and changes in 1950-to present mean an- Component Analyses, observatory and northern and central Andes.” nual near-surface and free air tempera- confirmatory (hypothesis tests) analyses tures, environmental lapse rates, dew for the detection of statistically signifi- points, specific humidity, squared moist and dry Brunt-Väisälä frequencies, lift- New York, USA). The following students participate as national transboundary areas specifically ing condensation levels, and convective Graduate Research Assistants: David Andrés Herrera examining historical minimum tempera- available potential energies, all of them (MSc candidate at Universidad Nacional de Colombia Sede Medellín), Fabian Suntaxi (MSc candidate at tures, maximum temperatures, and daily suggested by ensemble simulation out- Escuela Politécnica del Litoral-ESPOL-Ecuador), and Segundo Chimbolema (Corporación Grupo Randi rainfall. With these binational regions, puts. Analyses of regional conditions are Randi-Ecuador). and with a similar ongoing initiative in complemented with the study of cloud

8 Mountain Research Initiative Newsletter no. 7, 2012 Colombian-Ecuadorian Altitudinal Ecotransect © Daniel Ruiz Carrascal characteristics suggested by satellite bian-Ecuadorian and Bolivian-Peruvian events) to multi-decadal phenomena records and monthly sea surface tem- transboundary regions, respectively. will provide elements for the analysis of perature anomalies observed in the spa- Data loggers have been deployed at el- long-term changes and for better simula- tial domains [30°S-30°N, 30°E-90°W] evational intervals of 500 meters across tion and validation efforts. Digital sensor of the tropical Indo-Pacific region and two ca. 4,500 m altitudinal gradients data are currently being combined with [30°S-30°N, 60°W-15°E] of the tropical (Figure 2) and in as many ecosystems weather station records, whose historical Atlantic Ocean over the period 1942-to as possible. Gathered data are improv- periods span over 50 years, to build a set present. ing our understanding of the physical of climate change indices and to assess At least 12 data loggers/digital sen- processes taking place along the altitu- near-term historical conditions of atmo- sors measuring temperature and rela- dinal transects such as conditions of at- tive humidity at hourly intervals have mospheric instability, and local seasonal “Digital sensor data are been installed in each of the study areas temperature and precipitation anomalies. currently being combined to complement the available hydrome- A better understanding of local mecha- with weather station records, teorological networks, which include 75 nisms and their relationships with inter- and 17 weather stations in the Colom- annual (El Niño-Southern Oscillation whose historical periods span over 50 years”

spheric instability and moist convection in the two study areas. The dendrochronological work, in turn, aims to recon- struct the past ca. 100-200 years along the upper portion of the elevational gradients. A long list of Andean tree species has been analyzed to identify key species with sufficiently wide elevational ranges, distinctive annual rings, and long life spans. The result, a short list of tree species with dendrochronological potential currently includes 5 species for the analysis of páramo environments and high-Andean forests (such as Polylepis incana and Weinmannia cochensis) and 5 species in upper cloud forests (such as Symplocos carmencitae, Ocotea sp. and

Figure 2. Vertical profile of the 4,500 m altitudinal ecotransect and location of the temperature/ Cedrela montana). The oldest individu- relative humidity data loggers on the Amazonian slope of the Central Andes, in the border region als of the selected tree species have been between Bolivia and Peru. georeferenced, initially in the Colombia- Ecuador border area, in relatively well

Mountain Research Initiative Newsletter no. 7, 2012 9 Five-tiered integrated climate-related biodiversity vulnerability assessment in the Tropical Andes

preserved environments. To date the such goods and services due to climate women. The initial results of this study group has sampled 64 increment cores of change, and on whether local land-use will be available in late 2012. W. cochensis, 139 of P. incana, 76 of S. practices and patterns have been already carmencitae, 65 of Ocotea, and 64 of C. adapted or changed in response to cli- The land-use component montana. Tree-ring width chronologies mate change. Within this component, the The land-use component aims to deter- are currently under construction and ra- Corporación Grupo Randi Randi, based mine land-use types and patterns based diocarbon analyses will be used to assess in Quito, Ecuador, conducted a survey on satellite imagery and local informa- periodicity (i.e. annual) of tree rings in tion, and to relate biodiversity and vul- different species. “Within this component, the nerability patterns/gradients to existing Corporación Grupo Randi climate gradients, climate change trends The biodiversity component Randi, based in Quito, and forecasts, land-use patterns, and pre- The main objectives of the biodiver- dicted changes in land-use due to climate sity component include: (i) to develop Ecuador, conducted a survey change using Geographic Information knowledge on current biodiversity pat- of 545 persons (...) to learn Systems. terns (by ecosystem) and gradients (by more about local perceptions The outreach/capacity elevation) using several higher taxo- of climate change (...).” nomic groups of plants (e.g., ferns, bro- building component meliads, palms) and two animal groups Finally, the main objectives of the out- (birds, dung beetles) as proxies for over- of 545 persons (50% women) during reach and capacity building component all diversity, based on existing species- 2011 to learn more about local percep- include: (a) to determine potential adap- locality data and on field inventories tions of climate change and how people tive management measures and actions where knowledge gaps exist; and (ii) to are making adaptations to changes per- to increase the resilience of high-risk evaluate the vulnerability of species and ceived in their local climates. The sur- biodiversity areas to climate change; (b) ecosystems (based on the pooled vulner- vey sites comprise a series of transects to provide capacity building on the devel- ability of their component species) to cli- across sections of the western flanks of oped tools and analysis to institutions in mate change using the NatureServe Cli- the Andes in the north (southern Colom- Andean countries including Ministries of mate Change Vulnerability Index (http:// bia and northern Ecuador), and sections Environment, National Climate Change www.natureserve.org/prodServices/cli- of the eastern flanks in the south (south- Adaptation Programs, meteorological matechange/ccvi.jsp). ern Peru and northern Bolivia) within services, universities, and non-govern- the two transboundary Andean areas mental organizations to ensure that the The social component defined for the larger study. The surveys approach can be replicated elsewhere; The main objective of the social compo- were complemented with key informant and (c) to disseminate the results and nent is to consult local communities on in-depth interviews in each community conclusions in order to facilitate their in- ecosystem goods and services of par- surveyed, together with participatory corporation into action plans of national ticular value to them, on their perception mapping exercises on vulnerability with and international institutions. of any changes in the provisioning of focus groups composed of local men and

Authors

Daniel Ruiz Carrascal1 Associate Professor, Escuela de Ingeniería de Antioquia – EIA, Colombia, [email protected], International Research Insti- tute for Climate and Society (IRI) – Columbia University in the City of New York, USA [email protected] Sebastian K. Herzog, Asociación Armonía, Bolivia Peter M. Jørgensen, Missouri Botanical Garden, USA Trond H. Larsen, Conservation International, USA Rodney Martínez, Centro Internacional para la Investigación del Fenómeno El Niño – CIIFEN, Ecuador Juan José Nieto, Centro Internacional para la Investigación del Fenómeno El Niño – CIIFEN, Ecuador Susan V. Poats, Corporación Grupo Randi Randi – CGRR, Ecuador Marcella Ohira, Inter-American Institute for Global Change Research – IAI, Brazil

10 Mountain Research Initiative Newsletter no. 7, 2012 References

Hastings, D.A., Dunbar, P.K., 1999. Global land one-kilometer base elevation (GLOBE) digital elevation model, Documentati- on, Volume 1.0. Key to Geophysical Records Documentation (KGRD) 34. National Oceanic and Atmospheric Administration, National Geophysical Data Center, 325 Broadway, Boulder, Colorado 80303, U.S.A.

Herzog, S.K., Martínez, R., Jørgensen, P.M., Tiessen, H., Eds., 2011. Climate change and biodiversity in the Tropical An- des. Inter-American Institute of Global Change Research (IAI) and Scientific Committee on Problems of the Environment (SCOPE), São José dos Campos and Paris, 348 pp., ISBN: 978-85-99875-05-6.

Roeckner, E., Arpe, K., Bengtsson, L., Christoph, M., Claussen, M., Dümenil, L., Esch, M., Giorgetta, M., Schlese, U., Schul- zweida, U., 1996. The atmospheric general circulation model ECHAM4: model description and simulation of present-day climate. Max-Planck-Institut für Meteorologie Rep. 218, Hamburg, Germany, 90 pp.

Ruiz, D., Martinson, D.G., Vergara, W., 2012. Trends, stability and stress in the Colombian Central Andes. Climatic Change 112 (3): 717-732.

Mountain Research Initiative Newsletter no. 7, 2012 11