andean cloud forest tree lines

BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors nonprofit publishers academic institutions researchlibraries and research funders in the common goal of maximizing access to critical research

Andean Cloud Forest Tree LinesAuthor(s) Fausto O Sarmiento and Larry M FrolichSource Mountain Research and Development 22(3)278-287Published By International Mountain SocietyDOI httpdxdoiorg1016590276-4741(2002)022[0278ACFTL]20CO2URL httpwwwbiooneorgdoifull1016590276-4741282002290225B02783AACFTL5D20CO3B2

BioOne (wwwbiooneorg) is a nonprofit online aggregation of core research in the biological ecological andenvironmental sciences BioOne provides a sustainable online platform for over 170 journals and books publishedby nonprofit societies associations museums institutions and presses

Your use of this PDF the BioOne Web site and all posted and associated content indicates your acceptance ofBioOnersquos Terms of Use available at wwwbiooneorgpageterms_of_use

Usage of BioOne content is strictly limited to personal educational and non-commercial use Commercial inquiriesor rights and permissions requests should be directed to the individual publisher as copyright holder

Introduction

Alpine tree line concepts have historically fallen withinthe domain of wild mountain ecosystem ecology andfocused on temperate environmental factors that influ-ence the growth of montane forests and limit produc-tion forestry in northern latitudes with alpine environ-ments Tree line and ldquotimberlinerdquo are used interchange-ably but the term ldquoforest linerdquo is preferred as asynonym for ldquoforest limitrdquo (Sveinbjoumlrnsson 2000) Plantecologists often use timberline in association withclosed-canopy mountain forests whereas tree line isapplied to open-canopy forests (Koumlrner 1998) In thebiogeographical literature the altitudinal forest borderof mountains is usually referred to as the tree linewhereas the latitudinal continental ecotone is the tim-berline reflecting the border effect between forests andgrasslands (Armand 1992)

In the tropics harsh environmental conditions areviewed as the limiting factor for tree growth above a giv-en altitude mostly correlated with the threshold of10degC mean July isotherm (Jeniacutek 1997) or the locationof the frost line (Sarmiento 1986) At more northernlatitudes and on north-facing slopes the tree lineoccurs at a lower altitude (Ives 1978) A subalpine tran-sitional zone or krummholz line characterized by increas-ingly stunted and dwarfed trees is the norm in circum-polar and northern temperate zone mountains (Kull-man 1998) Vertical zoning thus results from therelationship between the vegetation type and the zonalclimate where belts are considered to be locally influ-enced variations dependent on temperate mountainconditions (Whittaker 1952) A combination of eleva-tion latitude and local climate defines where the for-estndashgrassland border occurs (Troll 1973) Hence physi-cally driven tree line presence is the norm at higher lat-itudes (Stevens and Fox 1991) In this lightSveinbjoumlrnsson (2000) points out that in North Americaand Europe 2 major themes emerged (1) the ldquopres-encendashabsence themerdquo with new tree line limits as aresponse to catastrophic events and (2) the ldquostarvationthemerdquo with tree line formation in response to nutrientavailability and resource allocations These themes arenot mutually exclusive and a combination of bothexplains locally relevant tree lines in the north

Although this concept has been considered validfor tropical montane cloud forests (TMCFs) we pro-pose an alternative hypothesis that historical andhuman factors may in fact have a far more importantinfluence on tree line presence in the tropics thanphysical factors do We suggest that a ldquohuman driverthemerdquo appears to be the most important one in tropi-cal America The driving forces of tree line formationin the Tropandean (tropical Andean) ecoregion arepast and present human influences including thedirect effects of mountain agriculture and grazing byboth introduced and native herbivores as well as burn-ing regimes and several other indirect effects

Unlike northern high-montane regions that are vir-tually uninhabited tropical high-altitude regions havehistorically served as human population centers (West-erberg and Christiansson 1999) New evidence points tothe anthropogenic nature of the ericaceous belt in theAfroalpine belt relating the patchy appearance of thetree line to the significance of fire and the role of her-bivores (Wesche et al 2000) In the tropical Andesmontane valleys and slopes as well as highlandplateaus offer a dependable temperate climate withlong mild growing seasons steady rainfall and absenceof infectious diseases (such as malaria and yellow fevercommon in the lowland tropics) Thus these areas havebeen preferential targets for human settlement sinceancient times (Flenley 1979 Gade 1999) People have

We challenge the para-digm of natural treeline formation in theequatorial Andes withan alternate view thatincorporates thehuman dimension Wepresent both directand indirect evidenceof anthropogenic influ-ences on cloud forest

tree line location these human influences should beincorporated in definitions of Andean tree lines as (1) theextensification of grassland for grazing and potato cultiva-tion for the upper limit and (2) the intensification of theagricultural frontier fuelwood gathering and timberextraction for the lower limit Although we do not claim tofully debunk the prevalent paradigm of tree line dynamics(which is descriptive depends on the natural sciencesand sees tree lines as physically controlled) we hope toachieve increased recognition for a challenging alternateview of tropical tree lines as functional resource-usedependent and human-driven Management plans andoverall tropical montane cloud forest conservation strate-gies need to consider this new perspective and incorpo-rate a proactive and assertive approach toward restora-tion of Andean forests in a way that will encourage land-scape diversity in tropical mountain ecoregions

Keywords Tree line mountain agriculture biodiversitydrivers restoration ecology Tropandean landscapespaacuteramo tropical Andes Ecuador

Peer reviewed February 2001 Accepted September2001

Andean Cloud Forest Tree LinesNaturalness Agriculture and the Human Dimension

Fausto O Sarmiento and Larry M Frolich

278

Mountain Research and Development Vol 22 No 3 August 2002 278ndash287

Research

279

lived for millennia in the high mountains of theneotropics (Budowski 1968 Ellenberg 1979) whereasthey have never lived (with the exception of thenomadic Saami in Scandinavian Lapland) in high-tem-perate and northern mountain areas

In the Tropandean ecoregion (Figure 1) themountain environment includes the human imprintimmersed in the matrix of cultural landscape despite 4axiomatic assumptions namely (1) the false ldquopristine-nessrdquo of the Transandean montane forests (Sarmiento1995a) (2) the seeming ldquonaturalnessrdquo of the verdantCisandean forests (Sarmiento 1998) (3) the ldquoirre-versibilityrdquo of the degraded Interandean forests (Hard-en 1999) and (4) the ldquofollyrdquo of restoration practicesfor conversion to healthy forest ecosystems in the equa-torial Andes (Sarmiento 1995b Chepstow-Lusty et al1998)

All the above are manifestations of the hiddenresponses (Stern 1983 Flenley 1992) to long-termhuman impact on tree borderlines in the mountainouslandscape of tropical America These and other moun-tain ecological principles need elucidation for applica-tion in conservation and sustainable developmentmodels because they underlie questions to beaddressed in future decision making affecting Andeancommunities

Tree lines The scientific rationale

Controversy over the origin of highland grasslands inforested mountains is still unsolved (Mark 1958 Billings1973 Luteyn 1999) For example temperate mountainsin North America harbor grassy areas locally known asldquobaldsrdquo or openings in the otherwise continuous forestcanopy in the highlands Although balds may have orig-inated at the time of prehistoric megafauna humanactivity has expanded them in mountain temperateforests (Weigl and Knowles 1995)

The long-term presence of humans in the moun-tainous tropics requires us to address a set of otherpotential determining factors for tropical tree linedynamics Fires set by humans agricultural clearinggrazing and trade routes may be of considerableimportance in the establishment and maintenance oftropical balds particularly paacuteramos (Ellenberg 1958Janzen 1973) In the equatorial Andes the presence offorest remnants with tall trees (such as those in Table 1)at elevations above 3600 m in remote coves of isolatedranges may be evidence of previous continuous forestformation Conversely the presence of both grass baldsor pajonal (Calamagrostis spp Festuca spp Cortaderia spp)and heath balds or chapparral (Hypericum spp Rubusspp Baccharis spp) on highland plateaus and in areas

FIGURE 1 A map of the 3 geo-graphical regions in the tropicalAndean (Tropandean) ecoregionin Ecuador 2 longitudinalcordilleras (eastern and west-ern) form the Interandean val-leys whose catchment basins(hoyas) are separated by trans-verse ranges (nudos) Adaptedfrom Joslashrgensen and Ulloa-Ulloa(1994) and Sarmiento (2000)


Mountain Research and Development Vol 22 No 3 August 2002

280TABLE 1 List of some of the tree species found in the study area represent-ing existing potential forest formation in the Interandean cloud forest of north-ern Ecuador (Based on Sarmiento 1988 Palacios and Tipan 1996)

Scientific name Local name Family taxon Life form Author taxon

Clusia flaviflora Guandera Clusiaceae Tree Engler cf

Weinmannia auriculifera Encenillo Cunoniaceae Tree Hieron ex Char

Weinmannia brachystachya Encenillo Cunoniaceae Tree Willd

Aequatorium campii Peludo Asteraceae Tree Nordenst

Weinmannia pinnata Encenillo Cunoniaceae Tree L

Valea stipularis Sacha capuli Elaeocarpaceae Tree L

Dicksonia sellowiana Helecho Dicksoniaceae Treeshrub Hook

Cyathea caracasana Helecho Cyatheaceae Tree (Klotzsch) Domin

Aexina macrophylla Cafecillo Melastomataceae Treeshrub (Naudin) Irlana

Brachyotum ledifolium Colca Melastomataceae Tree (Desr) Irlana

Brachyotum lindenii Colca Melastomataceae Tree Cogn

Saurauia bullosa Sacha yuyo Actinidaceae Treeshrub Wawra

Hyeronima macrocarpa Cuacho Euphorbiaceae Tree Mueel Arg

Meriania sanguinea Cafecillos Melastomataceae Treeshrub Wurdock

Miconia corymbiformis Colca Melastomataceae Tree Cogn

Miconia ochracea Colca Melastomataceae Treeshrub Triana

Cedrela montana Cedro Meliaceae Tree J Moritz ex Turcz

Ruagea hirsuta Cedrillo Meliaceae Treeshrub (CDC) Harms

Ilex spp Guayusa Aquifoliaceae Tree

Ocotea infrafoveolato Aguacatillo Laureaceae Tree H vd Werlf

Ocotea stuebelii Aguacatillo Laureaceae Tree Mez aff

Oreopanax micronulotus Puma maqui Araliaceae Tree Harms

Oreopanax palomophyllus Puma maqui Araliaceae Tree Harms

Gynoxis fuliginosa Huaiquero Asteraceae Tree (Kunlin) Cassini

Cybianthus marginatus Samal Myrsinaceae Treeshrub (Benthorn) Pipoly

Myrcianthes rhopaloides Arrayancillo Myrtaceae Tree (HBK) McVaugh

Piper carpunya Cordoncillo Piperaceae Tree R amp P

Hesperomeles lanuginosa Rosaceae Rosaceae Tree (R amp P) Hook

Polelepis sericea Yagual Rosaceae Tree Weddell

Senecio canescens Senecio Asteraceae Treeshrub HBK

Senecio lotiflorus Senecio Asteraceae Treeshrub Wedd

Alnus acuminata Aliso Betulaceae Tree HBK

Brunellia pauciflora Falso aluvillo Brunelliaceae Tree Cuatrec amp Orozco

Prunus rugosa Capuli Rosaceae Tree Kohene

Palicourea amethystina Cafeacute drsquo monte Rubiaceae Tree (R amp P) DC

Viburnum triphyllum Sombrilla Caprifoliaceae Tree Benlham

Cestrum peruvianum Hierba santa Solanaceae Tree Wildenow

table continues on next page

Research

281

formerly cultivated or grazed indicates previous humaninfluence (Grubb 1970)

The principal assumption on which temperate treeline concepts are basedmdashthat altitude is the determin-ing factormdashmay have little or no validity in the tropicalAndes at present as exemplified by remnant patches ofmature Andean forest above the paacuteramo areas More-over contemporary pine (Pinus radiata) or eucalypt(Eucalyptus globulus) plantations in the highland paacuteramodemonstrate the potential for viable tree-form vegeta-tion well above the tree line The Andean tree lineoften looks very similar to forest borders with agricul-tural lands where recent clear-cutting is known to havetaken place According to Laeliggaard (1992) ldquothe forestpatches border directly on grass paacuteramo vegetation andthe forest line is nearly always sharply cutrdquo Moreoverthere is no local terminology for the highland grass-land The Spanish word paacuteramo is not vernacular and isopen to a variety of interpretations (Luteyn 1999)

Our research findings some proxy testimony andexperience in the equatorial Andes offer evidence thatgeological archaeological and agricultural historymust be considered as the principal framework forunderstanding where the Andean cloud forest edgeoccurs We propose that the current TMCF landscapefeatures 2 distinctive tree lines as shown in Figure 2

bull An upper tree line that correlates with extensifica-tion of the human impact in the more climatically

demanding areas of the higher reaches shorteningthe extent of TMCFs from above

bull A lower tree line that correlates with intensificationof agriculture and fuelwood consumption in the lesschallenging topography of the foothills piedmontand interior highland plateaus shortening theextent of TMCFs from below

Study area

A significant 40000-ha strip of Interandean cloud foreststill stands on the eastern central valley slope in theprovinces of Imbabura and Carchi in northern EcuadorWith a distinctly unique and highly endemic Interan-dean flora and fauna this is perhaps the largest standof relatively pristine and yet nonprotected forest of thistype to be found in the equatorial Andes The water-shed is an extensive catchment basin distributed among2 main mountain ranges of the eastern Cordillera flow-ing toward the Interandean valley Pristine TMCF patch-es are found as relicts of an earlier widespread forestcover The valleys of San Gabriel and El Angel to thenorth and the valleys of Pimampiro Sigsipamba andMariano-Acosta to the south present an agriculturalmatrix of heavily fragmented forest patches hedgerowsand open areas covered with crops (mainly potato andcorn plantations) or pasture (Figure 3) In thePuruhanta area (located in the northwest corner of theCayambe-Coca Ecological Reserve) mature Andean


Sessea corimbosa Tomatillo Solanaceae Tree Goudof

Sessea crassivenosa Tomatillo Solanaceae Treeshrub Biller

Symplocos quitensis Shungi Symplocaceae Tree Brand

Aegiphila monticola Quesilulo Verbenaceae Tree Moldenke

Freziera canescens Zapatillo Theaceae Tree Bonpland

Escallonia myrtilloides Tibar Saxifragaceae Tree L

Escallonia paniculata Tibar Saxifragaceae Treeshrub (R amp P) Roem amp S

Clethra ferruginea Rojillo Clethraceae Tree (R amp P) Link

Clethra ovalifolia Rojillo Clethraceae Tree Turcz

Baccharis macrantha Chilca Asteraceae Treeshrub HBK

Barnadesia arborea Paloespino Asteraceae Treeshrub HBK

Diplostephium floribundum Alcotana Asteraceae Tree (Benth) Wedd

Diplostephium glandulosum Alcotana Asteraceae Tree Hieron

Diplostephium rhododendroides Alcotana Asteraceae Tree Hieron

Gynoxis buxifolia Huaiquero Asteraceae Tree Cassini

Schefflera spp Chefleras Araliaceae Treeshrub



282

forests still survive in isolation above 3800 m (Sarmien-to 1988) around Chique Lake The climate pattern typ-ical of the tropics is characterized by generally uniformtemperatures the year round (mean temperature 12˚C)and 2 distinctive moisture regimesmdash-a rainy season withmean precipitation of 2700 mmyear and a dry seasonwith mean precipitation of 520 mmyear (Cantildeadas-Cruz1983) Isolated ldquoislandsrdquo of paacuteramo display typical grass-

land vegetation including serotonic ferns and tallerherbs such as frailejones (Espeletia pycnophylla) orejas deconejo (Senecium spp) and achupallas (Puya clava-herculis)

The expanding population and opening of newagricultural lands do not appear to have greatly affect-ed the paacuteramo border in the eastern range within thestudy site A continuous 50-km stretch of intact forest

FIGURE 3 A panoramic view of therural landscape showing the hetero-geneity of land uses and the promi-nence of potato and corn cultivationNote native trees surviving along thehedgerows steep brooks and otherareas unsuitable for agriculture orgrazing (Photo by Larry Frolich)

FIGURE 2 A view of the rural land-scape of Carchi province wherehuman activity encroaches on theTMCFs and expands the agriculturalfrontier thus raising the lower limit ofthe tree line Past activity has beenmore important in forming thepaacuteramo and the upper tree line Pastand current pressures make protec-tion of cloud forest remnants crucialin this area (Photo by T Sulser)

Research

283

has prevented extensive use of the paacuteramo which is cur-rently uninhabited The main human impacts haveprobably remained the same if not diminished sincepre-Hispanic times

Trade routes are now along the Pan-AmericanHighway in the valley floor as opposed to ancient culun-cos which followed the ridge tops on the paacuteramondashforesttree line Contemporary travel in the paacuteramo is alongsmall trails principally to and from fishing rivers Thepaacuteramo is purposely burned during long dry periodsunder the belief that this will induce rainfall and willbring greener grazing grounds But it could be arguedthat frequent heat lightning especially experienced atnight may also lead to natural combustion under dryconditions albeit this phenomenon has not been docu-mented in the area As has been argued elsewhere fortropical mountains (Wesche et al 2000) ldquothe over-whelming majority of fires in tropical mountains areman-made with natural causes contributing at most aminor fractionrdquo Man-made and man-aged highlandenvironments are 2 considerations of the new para-digms of montology (Sarmiento 2000)

Direct evidence of tree line formation

Mountain agricultureHow different is the paacuteramo tree line from the agriculturalfrontier

Cash cropping with heavy external chemical inputsprincipally of potatoes but also of wheat barley andcorn is a common practice on the eastern Andeanslope in the study site Long enduring customs of ter-racing strip contour cultivation alley cropping andother traditional techniques such as wachu rozado (Frol-ich et al 1999) are observed As pointed out by Frolichand Guevara (1999) in Carchi province the agriculturalfrontier is continuously pushing the lower limits of thetree line upward because of the need for fresh fertilevolcanic soils to improve potato yields for an ever-expanding market The agricultural frontier tree line isnow high above the central valley floor because steepslopes have been cleared and converted to farmland forpotato cash monocropping Meanwhile most of theagricultural biodiversity and especially heirloom potatovarieties has been lost Worldwide the introduction ofindustrialized agricultural practices has undoubtedlydone more to affect cloud forest tree line dynamicsthan any other single factor

With time farm sites become exhausted of nutri-ents and biologically unbalanced soils are infested bynematodes and other pests that force farmers toincrease their use of synthetic pesticides artificial fertil-izers and manure Insecurity of land tenure andunclear colonization rules make intensification of agri-cultural production the driving force in claiming new

lands for cultivation from forest areas (Frolich and Gue-vara 1999) Production is rarely subsistence orientedand the bulk of the harvest is always hauled away to thecities of Tulcaacuten Ibarra and Quito This economicprogress has not occurred without severe collateralenvironmental and social consequences (Crissman et al1998)

Mountain grazingHow different is the paacuteramo tree line from the pasture frontier

New research trends identify megafauna herbivoresas the initial and principal creative force behind thebalds in temperate forested mountains (Weigl andKnowles 1995) In the equatorial Andes the presenceof fossils of megaherbivores such as mastodonts(Mastodon cuvierii M carchensis) giant sloths (Oreomy-lodon orcessi) Andean horses (Equus andium) and theprecursor of modern llamas (Paleolama spp) suggeststhat these Pleistocene foragers may have been instru-mental in clearing portions of TMCFs (Owen-Smith1987) thereby explaining the appearance of grass baldsin contemporary tropical Andean forests

Evidence of precarious animal husbandry in thearea can be found in old pottery depicting ungulatesand other zoomorphic styles found on ceramic artifactsMoreover several cultural centers including PastoCapuliacute and Quillasingas exhibited clear sedentary pat-terns of cultivation of crops and animal rearing Somesuch as Pimampiro developed into impressive tradesites (Molestina 1985) But Carchi was never subjectedto Inca rule and only dubious evidence of the presenceof camelid husbandry exists With the Spanish con-quest vast areas of the high mountain forests werecleared to introduce livestock mainly sheep and cattle

In modern times governmental subsidies offered todairy industry prompted the replacement of forest withpasture To this day cattle grazing is the principal activi-ty for large landholders of the central valley floor Pro-duction of milk cheese and other derivatives is moreprofitable than most crops In many places good agri-cultural soils have been converted to pasture using thetenacious introduced African grass kikuyo (Panicum clan-destinum) Other improved species are also planted forgrazing purposes particularly ryegrass and crabgrassMost farmers and their families are involved in dairyfarming either for subsistence milk production or tocomplement household income During the dry seasonlarge herds are sometimes left to graze within the monte(or secondary forest within the cloud forest patches)Most often however grazing is contained within highpasture areas where controlled burning is routinelyexercised all year long especially during the dry seasonSmoke and fire regimes have prompted a species com-position favoring smoke-stimulated flowering of groundbromeliads pyrophytic grasses and many serotonic



284

forbs and shrubs that resemble the paacuteramo assemblageof many fire-tolerant and even fire-dependent species

Burning regimeHow different is the tree line from the fire line

The ancient custom of burning the forest bordersfor charcoal making is still practiced by carboneros whocurrently sell their loads of timber-based charcoal inthe markets of nearby towns and cities In-situ produc-tion of charcoal is easily identified by cairns or kilnsites which leave permanent marks in the soilrsquos struc-ture and chemical composition Burning intenselyenough to produce charcoal kills the mycorrhizal com-ponents needed to jumpstart succession creating acrust of burned soil unsuitable for recalcitrant nativeforest seeds Thus recolonization of the cleared usual-ly round-shaped kiln site by native forest species isdelayed and species composition is altered perhapsthereby preventing full forest succession Kiln sitesmaintain the classic encroachment of graminoids andserotonic shrubs typical of a burning regime and con-tinued burning halts any ecological succession toward aclimax forest

Fire scars and charcoal deposits are interspersedalong the forestndashpaacuteramo tree line providing strong evi-dence that humans created the abrupt straight line sep-arating the forest from the grassland (Laeliggaard 1992)The presence of the round-shaped balds or paacuteramoislands near the TMCF border zone suggests that acces-sible border forest was used as an easy source ofexportable charcoal and firewood Firewood gatheringin and of itself cannot account for the widespreadnature of forest island burning Pasture managementand large-scale charcoal production on the other handbenefit big landowners or the middleman and usuallyresult in clearing large patches of forest

Indirect evidence of tree line formation

Paleoecological data and biogeographyPaleobiogeography in the tropics is understudied rela-tive to northern countries primarily because of scarceresources and the lack of strong scientific and govern-mental infrastructure for pursuing such studiesDetailed geological maps based on in-depth pollen andsoil analyses of the kind available for virtually any UScounty have not been completed for most of the trop-ics Limited data are thus often interpreted on the basisof corollaries with temperate climate studies Hencethe prevalent view of forestndashalpine tree line dynamics isthat because we are witnessing glacial recession at theend of an ice age the tree line is progressively advanc-ing to higher altitudesmdashas the computerized modelselucidating mountain climate change also predict

But where detailed analyses have been done in the

tropics (eg El Junco in the Galaacutepagos islands Sabanade Bogotaacute Cajas lakes in Ecuador a crater floor inPanama and a bog near Cuzco Peru etc [see Colin-vaux et al 1997 and Wille et al 2002]) they reveal thatthe situation is not so simple and methodologies as wellas paradigms should be revised Introduced multiple-use trees accompanied human population forays intohigh altitude as demonstrated by alders (Alnus spp) inCosta Rica (Kappelle 1996) and in Peru and Mesoamer-ica (Chepstow-Lusty et al 1996)

Palynological history is constrained by the difficultyof finding species-specific markers for some familiesthat may have both shrubs and tree forms It is difficultto specify the geological history of tree line dynamicsThis forces palynologists to generalize the comparisonof morphotypes for ldquoforestrdquo versus ldquoprairiesrdquo Colinvauxet al (1997) concluded that ldquoAndean vegetation did notrespond to glacial cooling or Holocene warming bymovement in beltsrdquo and proposed that taxa were reas-sorted according to temperature tolerance ldquoVegetationdid not move up and down slope as belts Rather plantassociations were reformed as temperature-sensitivespecies found different centers of distribution withchanging temperaturerdquo

The presence of paacuteramo endemics (ie Espeletia pyc-nophylla) is still enigmatic and suggests that vicarianceplays an important role The use of the caulirosula life-form (sensu Cuatrecasas 1957) as an indicator of highpaacuteramo is particularly relevant for the northern Andesof Colombia or Venezuela where several species ofEspeletia exist But in the equatorial Andes only 1 Espele-tia species is encountered in 2 distinct paacuteramos (El Angelin the north and Llanganatis in central Ecuador) mak-ing it ambiguous as a good paacuteramo indicator

Archaeological artifactsWhen the Spaniards came to the tropical Andes theydiscovered an area already widely cultivated and thusaffected by extensive human impact (Denevan 1992Kessler and Driesch 1993) In the equatorial Andesthey encountered impressive remains of previoushuman intervention in the landscape including aban-doned cangahua pyramids terraces and irrigation chan-nels Pimampiro at the southern end of our study sitewas well established as an important commercial centerfor exchange of goods from the coastal plain (as evi-denced by Spondilus shells) and from the Amazonregion (as evidenced by tropical feathers) In fact sev-eral different types of pottery pieces worked using dif-ferent techniques from various distant lowland cultureshave been found (Molestina 1985) The Mira watershed(across the central valley from our study site) was heavi-ly populated by different cultures such as Capuliacute Pas-tos and Quillasingas during the Formative and LateFormative periods Their artifacts are now the principal

Research

285

tools used to analyze the ancient human occupation ofthe Mira-Chota watershed (Knapp 1991)

Irrigation channels and trade routesThe Interandean basin in our study site is crossed byancient terracing systems and channels built on slopedterrain bringing water toward the area of PimampiroThe scale of this channel and irrigation system suggestsa population in the order of several tens of thousandsof people who are likely to have lived in the area (Moth-es 1987)

The commercial center in Pimampiro and the con-necting networks of trade routes to the south andnorth as well as to the lower tropical plains of the Ama-zon (via Puruhanta Lake) and to the Coast (via de MiraCanyon) made the Mira-Chota basin an important cul-tural center in pre-Hispanic times Satellite imagery hasdemonstrated the strategic location of pucaras or fortsinterspersed along the key points in the Interandeanvalley

Historical recordsThe extent of TMCFs within the basin in the colonialera is evidenced by the construction of big colonialhouses and churches in nearby cities particularly Ibar-ra Tulcaacuten San Gabriel and El Angel using old growthtimber and native lumber from the area Records keptat the catholic Bishoprsquos office in Ibarra for instanceattest to the wealth of forest products being exportedfrom the Chota valley and surrounding hills Today theChota valley is void of any significant native forest cov-er and most of the surrounding hills are denuded The

overall kingdom of Quito with all its chiefdoms to thenorth including those of the Chota-Mira watershed isreported to have included vast expanses of forests(Salomon 1986)

FisheriesHuman impact on tree lines can also be discerned infreshwater bodies that are surrounded by pristineTMCFs For instance Lake Puruhanta (or Chique) inthe study site an otherwise pristine and apparently nat-ural setting was significantly modified as a result ofwidespread efforts to develop fishing farms throughoutthe Ecuadorian highlands and especially in Imbaburaprovince considered the countryrsquos lake district Despitethe apparently pristine surroundings the area ofPuruhanta was significantly altered when inoculation ofthe exotic trout (Oncorhynchus mykiss Salmo gairdnerii)resulted in the extinction of local native prentildeadillas(Astroblepus spp) and bagrecitos (Rhamdia spp) At thesite of a fishing campground the forest was cut andburned up to the lakeshore with paacuteramo tussockregrowth appearing in the highly restricted site clearlyindicating human intervention (Sarmiento 1988)

Conclusions

The notion of anthropogenic community compositionlandscape structure and ecosystem function is a viablesuccinct explanation for Andean tree line dynamicsboth in the upper and lower boundaries of the nativeTMCFs (Figure 4) The simplified temperate-basedview of altitudinal vegetation belts moving up and down

FIGURE 4 Distinctive anthropogenic bound-ary between the forest and the paacuteramo inNueva America Note the straight line ofthe slope cut and the avoidance of ravinesand brooks not suitable for grazing or agri-culture Often semicircular shapes areobserved as evidence of previous kilns forcharcoal extraction (Photo by L Frolich)



286

with climate change has been confronted by a moreplausible explanation consistent with ecological theoryin which individual species are ephemera of habitat cli-mate and opportunity (Prentice et al 1991) both intemperate and in tropical ecosystems Instead of a sink-ing tree line a redo of assemblages with cold tolerantspecies might have kept the equatorial Andes forestedthroughout the last Ice Age

Evidence favoring the need to integrate the human

dimension in Holocene tree line dynamics in the Amer-ican tropics comes from different sources (but seeLauer and Klaus 1975) The equatorial Andes with along prehistorical presence of humans in the mountainlandscape thus emerge as a classic example of theanthropogenic nature of tropical montane culturallandscapes including supposedly pristine primaryforests and paacuteramos

REFERENCES

Armand AD 1992 Sharp and gradual mountain timberlines as a result ofspecies interaction In Hansen AJ Di Castri F editors Landscape Bound-aries New York Springer pp 360ndash378Billings WD 1973 Arctic and alpine vegetations similarities differencesand susceptibility to disturbance Bioscience 23697ndash704Budowski G 1968 Mountains of tropical America In Troll C editor Geo-ecology of the Mountainous Region of the Tropical Americas Proceedings ofthe UNESCO Mexico Symposium 1966 Colloquium Geographicum Volume9 Bonn Ferdinand Duumlmmlers VerlagCantildeadas-Cruz L 1983 El mapa bioclimaacutetico y ecoloacutegico del Ecuador QuitoEcuador MAG-PRONAREGChepstow-Lusty A Bennett K Fjeldsaring J Kendall A Galiano W Tupayachi-Herrera A 1998 Tracing 4000 years of environmental history in the Cuzcoarea Peru from the pollen record Mountain Research and Development18(2)159ndash172Chepstow-Lusty A Bennett K Switsur V Kendall A 1996 4000 years ofhuman impact and vegetation change in the central Peruvian Andesndash-withevents paralleling the Maya record Antiquity 70(270)824ndash833Colinvaux PA Bush M Steinitz-Kannan M Miller MC 1997 Glacial andpostglacial pollen records from the Ecuadorian Andes and Amazon Quater-nary Research 4869ndash78Crissman CC Espinosa PA Ducrot CEH Cole DC Carpio F 1998 The casestudy site physical health and potato farming systems in Carchi provinceIn Crissman CC Antle JM Capalbo SM editors Economic Environmentaland Health Trade-offs in Agriculture Pesticides and the Sustainability ofAndean Potato Production Dordrecht Netherlands Kluwer Academic Pub-lishers pp 89ndash117Cuatrecasas J 1957 A sketch of the vegetation of the North Andean province[summary] In Jonshon R editor Proceedings of the VIIIth Pacific Science Con-gress Volume 4 Quezon City The University of the Philippines pp 167ndash173Denevan WM 1992 The pristine myth the landscape of the Americas in1492 Annals of Association of American Geographers 82(3)369ndash385

Ellenberg H 1958 Wald oder Steppe Die natuumlrliche Pflanzendecke derAnden Perus Umschau 1958645ndash681Ellenberg H 1979 Manrsquos influence on tropical mountain ecosystems ofSouth America Journal of Ecology 67401ndash416Flenley JR 1979 The Equatorial Rain Forest A Geological History LondonButterworthFlenley JR 1992 UV-B insolation and the altitudinal forest limit In FurleyPA Proctor J Ratter JA editors Nature and Dynamics of ForestndashSavannaBoundaries London Chapman and Hall pp 273ndash282Frolich LM Guevara E 1999 The role of family-based agricultural innova-tion in conserving tropical montane cloud forest the Guandera Project innorthern Ecuador In Sarmiento F Hidalgo J editors Entendiendo las Inter-faces Ecoloacutegicas para la Gestioacuten de Paisajes Culturales en los Andes Memo-rias del Tercer Simposio Internacional sobre Desarrollo Sustentable en losAndes Quito Ecuador Corporacioacuten Editora Nacional pp 45ndash50Frolich LM Sherwood S Hemphill A Guevara E 1999 ldquoEco-Papasrdquothrough potato conservation towards agroecology Leisa 3ndash444ndash46Gade D 1999 Nature and Culture in the Andes Madison WI University ofWisconsin PressGrubb PJ 1970 The impact of man on the paacuteramo of Cerro AntisanaEcuador Journal of Applied Ecology 77ndash8Harden C 1999 Conecciones sobre la tierra Recomendaciones de mane-jo de un anaacutelisis escala cuenca de la erosioacuten del suelo en los Andes ecua-torianos In Sarmiento F Hidalgo J editors Entendiendo las InterfacesEcoloacutegicas para la Gestioacuten de Paisajes Culturales en los Andes Memoriasdel Tercer Simposio Internacional sobre Desarrollo Sustentable en los AndesQuito Ecuador Corporacioacuten Editora Nacional pp 117ndash122Hustich I 1983 Treeline and tree growth studies during 50 years somesubjective observations Nordicana 47181ndash188Ives JD 1978 Remarks on the stability of timberline In Troll C Lauer Weditors Geoecological Relations Between the Southern Temperate Zone andthe Tropical Mountains Wiesbaden Germany Steiner Verlag pp 313ndash318

AUTHORS

Fausto O SarmientoDepartment of Geography and Office for International Education 200 Bar-row Hall University of Georgia Athens GA 30602 USAfsarmienugaedu

Larry M FrolichDepartment of Ecology and Evolutionary Biology Brown University Provi-dence RI 02912 USAlarry_frolichbrownedu

ACKNOWLEDGMENTS

Many thanks for the help and insights gained from families who live in thenumerous communities within the study site including Mariscal Sucre Nue-va America Mariano Acosta Monte Olivo El Chamizo and El Playoacuten deSan Francisco The Ecuadorian Museum of Natural Sciences facilitatedwork around Puruhanta Lake and Fundacioacuten Jatun Sacha provided housingin the Guandera Reserve Ideas on the shifting paradigm of Andean treelines were presented at the 3rd International Symposium on SustainableMountain Development of the Andean Mountain Association (AMA) inQuito Ecuador and at the Global Mountain Biodiversity Assessment(GMBA) Conference in Rigi Switzerland FS was funded by the UGArsquos Inter-national Academic Faculty Development Program LF was funded by theSchool of Agricultural and Environmental Sciences of PUCE-I and the BrownUniversityrsquos Department of Ecology and Evolutionary Biology Publicationsupport came from the UGA Center for Latin American and Caribbean Stud-ies (CLACS) through a matching grant from the Italian Committee for theInternational Year of Mountains (IYM2002) to the Andean Mountain Associ-ation (AMA) and the National Science Foundation (NSF) award 9971024

Research

287

Janzen DH 1973 Rate of regeneration after a tropical high elevation fireBiotropica 5(2)117ndash122Jeniacutek J 1997 The diversity of mountain life In Messerli B Ives J editorsMountains of the World A Global Priority London The Parthenon PublishingGroup pp 199ndash230Joslashrgensen PM Ulloa-Ulloa C 1994 Seed plants of the high Andes ofEcuadorndash-a checklist AAU Reports 341ndash58Kappelle M 1996 Los Bosques de Roble (Quercus) de la Cordillera de Tala-manca Costa Rica Biodiversidad Ecologiacutea Conservacioacuten y Desarrollo Ams-terdam University of Amsterdam Santo Domingo de Heredia Costa Rica INBioKessler M Driesch P 1993 Causas e historia de la destruccioacuten de losbosques altoandinos en Bolivia Ecologiacutea en Bolivia 211ndash18Knapp G 1991 Andean Ecology Adaptive Dynamics in Ecuador BoulderCO Westview PressKoumlrner C 1998 A re-assessment of high elevation treeline positions andtheir explanation Oecologia 115445ndash459Kullman L 1998 Tree-limits and montane forests in the Swedish Scandessensitive biomonitors of climate change and variability Ambio 27(4)312ndash321Lauer W Klaus D 1975 Geoecological investigations on the timberline ofPico de Orizaba Mexico Arctic and Alpine Research 7315ndash330Laeliggaard S 1992 Influence of fire in the grass paacuteramo vegetation ofEcuador In Balslev H Luteyn JL editors Paacuteramo An Andean EcosystemUnder Human Influence San Diego CA Academic Press pp 151ndash170Luteyn JL 1999 Paacuteramos A Checklist of Plant Diversity Geographical Dis-tribution and Botanical Literature Memoirs of the New York Botanical Gar-den Volume 84 Bronx NY The New York Botanical GardenMark AF 1958 The ecology of the southern Appalachian grass balds Eco-logical Monographs 28293ndash336Molestina MC 1985 Investigaciones arqueoloacutegicas en la zona negativodel Carchi o Capuliacute Revista Cultura 7(21a)31ndash82Mothes P 1987 La Acequia de Pimampiro riego tradicional en el Nortedel Ecuador Ecuador Debate 1469ndash86Owen-Smith N 1987 Pleistocene extinctions the pivotal role of mega-her-bivores Paleobiology 13351ndash362Palacios W Tipaz G 1995 Un bosque remanente de altura (Reserva Guan-dera) Estructura y composicion floristica Revista Geografica 3739ndash60Prentice IC Bartlein PJ Webb T 1991 Vegetation and climate change ineastern North America since the last glacial maximum Ecology722038ndash2056Salomon F 1986 Native Lords of Quito in the Age of the Incas The PoliticalEconomy of North Andean Chiefdoms Cambridge Cambridge University Press

Sarmiento F 1995a Human impacts on the cloud forests of the upperGuayllabamba river basin Ecuador and suggested management respons-es Ecological Studies 110284ndash295Sarmiento F 1995b Restoration the challenge for conservation of tropan-dean landscapes In Churchill S Balslev H Forero E Luteyn J editors Bio-diversity and Conservation of Neotropical Montane Forests Bronx NY TheNew York Botanical Garden pp 637ndash651Sarmiento F 1988 Expedicioacuten cientiacutefica de reconocimiento ecoloacutegico dela laguna de Puruhanta Revista Geograacutefica 257ndash28Sarmiento F 1998 The Quijos river valley a protected landscape categoryas best management practice for conservation and development in tropan-dean Ecuador The George Wright Forum 14(3)59ndash66Sarmiento F 2000 Breaking mountain paradigms ecological effects onhuman impacts in man-aged tropandean landscapes Ambio29(7)423ndash431Sarmiento G 1986 Ecological features of climate in high tropical moun-tains In Viulleumier F Monasterio M editors High Altitude Tropical Bio-geography New York Oxford University Press pp 11ndash45Stern R 1983 Human impact on tree border lines In Holzner W WergerMJA Ikusima A editors Manrsquos Impact on Vegetation The Hague Nether-lands Junk Publishers pp 227ndash236Stevens GC Fox JF 1991 The causes of treeline Annual Review of Ecolo-gy and Systematics 22177ndash191Sveinbjoumlrnsson B 2000 North American and European treelines externalforces and internal processes controlling position Ambio 29(7)388ndash395Troll C editor 1968 Geo-ecology of the Mountainous Region of the TropicalAmericas Bonn Ferdinand Duumlmmlers VerlagTroll C 1973 The upper timberlines in different climatic zones Arctic andAlpine Research 5(3)3ndash18Weigl PD Knowles TW 1995 Megaherbivores and southern Appalachiangrass balds Growth and Change 26365ndash382Wesche K Miehe G Kaeppeli M 2000 The significance of fire forAfroalpine ericaceous vegetation Mountain Research and Development20(4)340ndash347Westerberg L Christiansson C 1999 Highlands in East Africa unstableslopes unstable environments Ambio 28(5)419ndash429Whittaker RH 1952 Vegetation of the Great Smokey mountains Ecologi-cal Monographs 261ndash80Wille M Hooghiemstra H Hofstede R Fehse J Sevink J 2002 Upper for-est line reconstruction in a deforested area in northern Ecuador based onpollen and vegetation analysis Journal of Tropical Ecology 18409ndash440

Introduction

Alpine tree line concepts have historically fallen withinthe domain of wild mountain ecosystem ecology andfocused on temperate environmental factors that influ-ence the growth of montane forests and limit produc-tion forestry in northern latitudes with alpine environ-ments Tree line and ldquotimberlinerdquo are used interchange-ably but the term ldquoforest linerdquo is preferred as asynonym for ldquoforest limitrdquo (Sveinbjoumlrnsson 2000) Plantecologists often use timberline in association withclosed-canopy mountain forests whereas tree line isapplied to open-canopy forests (Koumlrner 1998) In thebiogeographical literature the altitudinal forest borderof mountains is usually referred to as the tree linewhereas the latitudinal continental ecotone is the tim-berline reflecting the border effect between forests andgrasslands (Armand 1992)

In the tropics harsh environmental conditions areviewed as the limiting factor for tree growth above a giv-en altitude mostly correlated with the threshold of10degC mean July isotherm (Jeniacutek 1997) or the locationof the frost line (Sarmiento 1986) At more northernlatitudes and on north-facing slopes the tree lineoccurs at a lower altitude (Ives 1978) A subalpine tran-sitional zone or krummholz line characterized by increas-ingly stunted and dwarfed trees is the norm in circum-polar and northern temperate zone mountains (Kull-man 1998) Vertical zoning thus results from therelationship between the vegetation type and the zonalclimate where belts are considered to be locally influ-enced variations dependent on temperate mountainconditions (Whittaker 1952) A combination of eleva-tion latitude and local climate defines where the for-estndashgrassland border occurs (Troll 1973) Hence physi-cally driven tree line presence is the norm at higher lat-itudes (Stevens and Fox 1991) In this lightSveinbjoumlrnsson (2000) points out that in North Americaand Europe 2 major themes emerged (1) the ldquopres-encendashabsence themerdquo with new tree line limits as aresponse to catastrophic events and (2) the ldquostarvationthemerdquo with tree line formation in response to nutrientavailability and resource allocations These themes arenot mutually exclusive and a combination of bothexplains locally relevant tree lines in the north

Although this concept has been considered validfor tropical montane cloud forests (TMCFs) we pro-pose an alternative hypothesis that historical andhuman factors may in fact have a far more importantinfluence on tree line presence in the tropics thanphysical factors do We suggest that a ldquohuman driverthemerdquo appears to be the most important one in tropi-cal America The driving forces of tree line formationin the Tropandean (tropical Andean) ecoregion arepast and present human influences including thedirect effects of mountain agriculture and grazing byboth introduced and native herbivores as well as burn-ing regimes and several other indirect effects

Unlike northern high-montane regions that are vir-tually uninhabited tropical high-altitude regions havehistorically served as human population centers (West-erberg and Christiansson 1999) New evidence points tothe anthropogenic nature of the ericaceous belt in theAfroalpine belt relating the patchy appearance of thetree line to the significance of fire and the role of her-bivores (Wesche et al 2000) In the tropical Andesmontane valleys and slopes as well as highlandplateaus offer a dependable temperate climate withlong mild growing seasons steady rainfall and absenceof infectious diseases (such as malaria and yellow fevercommon in the lowland tropics) Thus these areas havebeen preferential targets for human settlement sinceancient times (Flenley 1979 Gade 1999) People have

We challenge the para-digm of natural treeline formation in theequatorial Andes withan alternate view thatincorporates thehuman dimension Wepresent both directand indirect evidenceof anthropogenic influ-ences on cloud forest

tree line location these human influences should beincorporated in definitions of Andean tree lines as (1) theextensification of grassland for grazing and potato cultiva-tion for the upper limit and (2) the intensification of theagricultural frontier fuelwood gathering and timberextraction for the lower limit Although we do not claim tofully debunk the prevalent paradigm of tree line dynamics(which is descriptive depends on the natural sciencesand sees tree lines as physically controlled) we hope toachieve increased recognition for a challenging alternateview of tropical tree lines as functional resource-usedependent and human-driven Management plans andoverall tropical montane cloud forest conservation strate-gies need to consider this new perspective and incorpo-rate a proactive and assertive approach toward restora-tion of Andean forests in a way that will encourage land-scape diversity in tropical mountain ecoregions

Keywords Tree line mountain agriculture biodiversitydrivers restoration ecology Tropandean landscapespaacuteramo tropical Andes Ecuador

Peer reviewed February 2001 Accepted September2001

Andean Cloud Forest Tree LinesNaturalness Agriculture and the Human Dimension


278

Mountain Research and Development Vol 22 No 3 August 2002 278ndash287

Research

279


















































Research

281







Study area





















282






Research

283














284












Research

285







Conclusions





286




REFERENCES



AUTHORS



ACKNOWLEDGMENTS


Research

287



Research

279


















































Research

281







Study area





















282






Research

283














284












Research

285







Conclusions





286




REFERENCES



AUTHORS



ACKNOWLEDGMENTS


Research

287













































Research

281







Study area





















282






Research

283














284












Research

285







Conclusions





286




REFERENCES



AUTHORS



ACKNOWLEDGMENTS


Research

287





282






Research

283














284












Research

285







Conclusions





286




REFERENCES



AUTHORS



ACKNOWLEDGMENTS


Research

287



Research

283














284












Research

285







Conclusions





286




REFERENCES



AUTHORS



ACKNOWLEDGMENTS


Research

287





284












Research

285







Conclusions





286




REFERENCES



AUTHORS



ACKNOWLEDGMENTS


Research

287



Research

285







Conclusions





286




REFERENCES



AUTHORS



ACKNOWLEDGMENTS


Research

287





286




REFERENCES



AUTHORS



ACKNOWLEDGMENTS


Research

287



Research

287



andean cloud forest tree lines

Documents