Schulz Et Al. 2011 Phytogeographic Divisions, Climate Change and Plant Dieback Coastal Desert Northern Chile

8/4/2019 Schulz Et Al. 2011 Phytogeographic Divisions, Climate Change and Plant Dieback Coastal Desert Northern Chile

1/19

Vol. 65 No. 2 1691872011

DOI: 10.3112/erdkunde.2011.02.05 http://www.erdkunde.uni-bonn.deISSN 0014-0015

PHYTOGEOGRAPHIC DIVISIONS, CLIMATE CHANGE AND PLANT DIEBACK

ALONG THE COASTAL DESERT OF NORTHERN CHILE

Natalie Schulz, Patricio aceituNoand Michael richter

With 10 gures and 3 tables

Received 14. July 2010 Accepted 02. May 2011

Summary: Along the hyper-arid Chilean coastal desert between 30S and 18S the Loma vegetation undergoes a gradualtransition from open shrubland to small isolated areas of a scarce plant cover. Floristic and physiognomic features allowa differentiation of ve Loma formations, each of them characterized by a distinctive spectrum of plant communities.However, particularly in the northern section of the investigation area, numerous indications point to a strong vegetationdecline including a deterioration of plant cover, reduction of the vitality of various taxa, probably also a local loss of someperennial species, and even a dieback of specic populations. These signs of a retrogression, which coincide with a regionaldisappearance of Guanaco herds in the coastal area between 20S and 2330S, became apparent in the second half of thepast century and were most likely provoked by recent climate change in the arid coastal region. Especially the decrease ofrainfall frequency might have negative implications for the regeneration and preservation of plants. In addition, a strongreduction of cloudiness in the northernmost section affects plant growth due to further limitations in the water disposability.A projected sustained decline of rainfall is expected to continue endangering the surprisingly high oristic diversity of thesensitive ecosystem complexes in the coastal desert.

Zusammenfassung: Entlang der perariden chilenischen stenwste von 30S bis 18S verndert sich die Loma-e-Entlang der perariden chilenischen stenwste von 30S bis 18S verndert sich die Loma-e-getation von offenen Strauchbestnden hin zu isolierten Panzenvorkommen geringer Deckung. Floristische und physi -ognomische riterien erlauben eine Untergliederung in fnf Abschnitte, die sich jeweils durch ein eigenes Spektrum anPanzengesellschaften auszeichnen. Allerdings belegen im Norden des Untersuchungsgebietes zahlreiche Merkmale eineerminderung der italitt einzelner Sippen, den lokalen erlust einiger perenner Taxa und sogar den Niedergang ganzerPopulationen einzelner Arten. Dieser egetationsrckgang, der mit den erlust von Guanaco-Herden aus der stenregionzwischen 20S und 2330S einhergeht, zeichnet sich vor allem seit der zweiten Hlfte des letzten Jahrhunderts ab und ist mitgroer Wahrscheinlichkeit auf ernderungen im lima der ariden Region zurckzufhren. Insbesondere die Abnahme derohnehin bescheidenen Regenhugkeit drfte sich negativ auf die Regeneration und den Erhalt der Panzen auswirken. Zu-dem beeinusst ein Rckgang der Bewlkung im Norden der chilenischen stenwste die Lebensbedingungen der egeta-tion aufgrund weiterer Beschrnkungen im Wasserhaushalt negativ. Die absehbare weitere Abnahme der Niederschlge wird

die berraschend groe Panzendiversitt des hchst sensiblen Loma-kosystem-omplexes auch in Zukunft bedrohen

Keywords: Loma formations, oristic composure, plant dieback, fog, decreasing rainfalls

1 Introduction

For some 15 years, reports on a regionally dra-matic plant dieback have called attention to a dam-age of the unique ecosystems in the northern partof the coastal Atacama Desert (18S-30S), knownas one of the worlds driest regions. Although char-acterized by hyper-arid conditions, it locally sus-

tains exuberant Loma vegetation, which harboursa surprisingly high amount of endemic vascularplant species. Despite rain scarcity and harsh envi-ronmental conditions, it succeeds in subsisting onthe windward escarpment of the coastal cordillera,beneting from frequent fogs.

Meanwhile, an increasing number of indica-increasing number of indica-tions point to a decline of the Loma vegetation in

the northern part of the desert, probably associatedwith the loss of some species (e.g., FollMaNN 1995;richter 1995; ruNdel et al. 1997; dilloN andhoFFMaNN 1997; Grau 2000; Muoz-Schicket al.2001; PiNto et al. 2001; PiNto and kirberG 2005;PiNto 2007). Although some aspects of the vegeta-tion decline has already been approached in a fewstudies on the current conservation status of some

cactus species (belMoNte et al. 1998; PiNto andkirberG 2005; PiNto 2007), the actual dimensionsand origins of the deterioration processes remainso far unknown. Some authors have postulatedincreasing aridity in recent decades as a probablecause of the vegetation decline (e.g., richter1995;ruNdel et al. 1997; PiNto and kirberG 2005).However, the lack of studies on recent cl imate evo-


2/19

170 Vol. 65 No. 2

lution in this region, particularly on precipitation,cloudiness and fog development, has made the cor-roboration of this assumption difcult so far.

More detailed research on driving forces of the vegetation decline constitutes an important task,

particularly considering the singularity and highsensibility of the Loma ecosystems. In this context,the objective of our study is to assess vegetationchanges in the arid coastal region in the recent past.Furthermore, detailed analyses of climate develop-ment during the past century mayprovide evidencefor possible climate effects on vegetation retro-gression. This paper focuses on spatial patternsof vegetation in order to elaborate a comparativeframework of the plant formations and commu-nities based on detailed previous and own studiesof Chilean Lomas (i); on indices and evidence ofrecent vegetation changes compiled from varioussources (ii); and nally on analyses of climate varia-tions in the recent past (iii), which are expected tocontribute to the discussion on possible triggers ofthis vegetation decline.

2 Data and methods

egetation sampling was carr ied out in 15 Lomalocalities between 20S and 30S (2005, 2006, and2009). In each area, plant cover was studied alongone to three transects situated mostly on southern,

south-western or western sea-facing slopes, whichusually receive moisture by stratocumulus clouds(detailed information in Schulz 2009). Floristiclists were compiled for each one of the studiedLoma localities and additionally for 14 further lo-calities using information from different sources(for localities see Fig.1). They include data fromown eld trips and those recorded in former veg-etation studies and botanical collections (the latterfrom databases of three important Chilean her-baria: Santiago (SGO), Concepcin (CONC), andLa Serena (HULS). Floristic data from 29 Lomalocalities were used for cluster analyses in order to

differentiate the Loma vegetation. A hierarchicalagglomerative clustering was done as presence-ab-sence analysis of the total number of registered per-ennial native plants. The Srensen index was usedas an algorithm for the calculation of similarity andaverage group linkage as distance between the clus-ters. The species nomenclature follows hoFFMaNNand Walter (2004) for Cactaceae and Squeo et al.(2008), MarticoreNa et al. (1998 and 2001) for fur-ther families.

With regard to difculties in the application ofquantitative methods to analyse vegetation changes within the investigation areas, we opted for a quali-tative approach, since no vegetation releves prior to1972 were available. Due to insufcient spatial resolu-

tion and temporal coverage, satellite and areal imagesare inappropriate for detecting changes in the desert vegetation considered here. Hence, previous oristicpublications and further sources were revised to assessindications on the previous oristic composition andstate of plant cover. Since botanical research in some ofthe areas has been conducted continuously throughoutthe past nine decades, comparisons of oristic lists al-low a detection of general tendencies during the recentvegetation history.

Trend analyses were performed for precipitationand total cloud cover. Analyses of tendencies in fog, afurther important ecological factor in the study area,could not be included due to a series of difculties as-sociated with trend analyses of fog. Firstly, there areno suitable observational fog data series, since meteo-rological stations in the study area are situated mostlyin the proximity of airports, at lower, widely fog-freesites. Secondly, reliable continuous fog data derivablefrom fog precipitation measurements or satellite imagescover a relatively short period since 1997 and 1980 re-spectively. For the purposes of this study it is, however,crucial to assess longer trends from at least the mid-20thcentury or longer. And nally, tendencies identied forspecic sites are not easily transferable to larger coast-

al zones, as fog characteristics vary highly along thenorthern Chilean coast depending on the ocean prox-ocean prox-imity, altitude, local relief, exposure of the coastal rang-es (larraiN et al. 2002), as well as orographic fog inu-ence and other factors not studied so far. Nevertheless,some general conclusions about advection fog tenden-cies can be derived from stratocumulus trends, giventhe fact that the formation and persistence of advectionfog are highly inuenced by the presence of this type ofclouds (cereceda et al. 2002; Garreaud et al. 2008).

To evaluate long-term changes of precipitationand cloudiness regimes, series of annual and month-ly means from four coastal stations (Arica, Iquique,

Antofagasta, La Serena) and one station nearby thecoast (Copiap, s. Fig.1) were compiled from meteoro-logical annals published by the Chilean MeteorologicalService (DMC). Also, daily precipitation values wereprovided by the DMC. Few data gaps in the rainfallseries are completed by information obtained fromalMeyda (1948), who extracted the missing data fromoriginal reports of the respective sites. All climatic sta-tions considered here are located below the zone of di-rect fog inuence.


3/19

171N. Schulz et al.: Phytogeographic divisions, climate change and plant dieback along ...2011

Apart from Arica, which does not show any cor-relation with precipitation series of other stations, allrainfall series were submitted to a relative homogene-

ity test according to alexaNderSSoN (1986). Despitethe fact that all stations were relocated at least onceduring the last century, this test did not reveal any

R

oLoa

R

oC

op

iap

RoElqui

Arica

La Serena

CaRrizal Bajo

AR = Arica1820 S

7020 W

2049 S 7010 W

2116 S 7003 W

2329 S 7035 W2335 S 7027 W

2411 S 7033 W2419 S 7034 W2432 S 7033 W2438 S 7031 W2449 S 7031 W2456 S 7028 W

2506 S 7026 W2526 S 7025 W

2532 S 7030 W

2552 S 7039 W2602 S 7039 W

2810 S 7109 W

2632 S 7036 W2644 S 7039 W

2659 S 7043 W

2927 S 7115 W

= Obispo

= Quebrada Len

= El Tofo

2202 S 7010 W

2014 S 7004 W

2233 S 7014 W

2342 S 7028 W

2619 S 7037 W

2527 S 7026 W

2501 S 7026 W

2021 S 7007 W

2050 S 7009 W

= Antofagasta

= Chanral

= TalTal

= PaPoso

= Punta Gruesa

= Alto Patache

= Punta Lobos

= Alto Chipana

= Cerro Moreno= La Chimba

= El Cobre= Blanco Encalada= Miguel Diaz= La Plata= Medano= Rinconada

= Matancilla= Cerro Perales

= Cifuncho

= Esmeralda= Pan de Azucar

= Carrizal Bajo

= Flamenco

CPMTCP

CF

EMPA

CF

EMPA

CR

FLOB

QL

FLOB

QL

El Tofo ET

= Tocopilla

= Iquique

= Cobija

AP

PL PL

AC AC

CM CMLC LC

EC ECBE BE

MD MDLP LPMERC

MERC

MT

28

Salar

7274

TO

IQ

CB

AN

CH

TT

PP

PG PG

AP

Caldera

18

22

26

30

20

24

TalTal

PaPoso

CoBija

Vallenar

I. Regin

II. Regin

III. Regin

IV. Regin0 100 km

6870

Huasco

Calama

0 m

200 m

500 m

1000 m

2000 m

3000 m

4000 m

5000 m

Altitude:

CHanral

TOcopilla

La Serena

Copiap

ANtofagasta

IQuique

ARica

P E R U

B O L I V I A

A R G E N T I N A

NC

0

5

10

15

20

25

J F M A M J J A S O N D

18.9C 1.6 mmARICAmm

00

0.4

0.6

0.8

1.0

10

20

30

0.2

C18.4C 0.9 mmIQUIQUE

5

10

15

20

25


mm

0

0.4

0.6

0.8

1.0

10

20

30

0.2

0 0

C16.7C 3.0 mmANTOFAGASTA

5

10

15

20

25

mm

00

0.4

0.6

0.8

1.0

10

20

30

0.2

J F M A M J J A S O N D0

C13.7C 79.9 mmLA SERENA

5

10

15

20

25

mm

0 0

10

20

30


C15.2C 18.5 mmCOPIAP

5

10

15

20

25

mm

0

10

20

30

J F M A M J J A S O N D0

Fig. 1: Study area. Monthly precipitation and temperature means on the basis of data from 19712000 (note two scales forprecipitation). Localities considered by the oristic analysis (data from own collections and/or other studies) are indi-cated in and at the right of the map


4/19

172 Vol. 65 No. 2

signicant inhomogeneity. Data of total cloud coverobservations are compiled for the period 1950-2008.Since the interannual variability of cloudiness at all veried stations is widely coherent and the observa-tion methodology has not changed during the past

decades, no homogeneity tests were performed. Theannual and seasonal frequency of cloudy days (meandaily cloud cover in oktas 6/8) and cloudless days(mean daily cloud cover 2/8) were derived from themeteorological annals of the DMC.

3 Geo-ecological resources of the study area

Along its entire length, the narrow coastal strip ofnorthern Chile is bordered by a mostly steep moun-tain escarpment of up to 3000 m a.s.l. elevation, whichrises like a wall from the seaside in its northern sec-tion, while it is dissected by numerous dry valleys inthe southern part of the research area. Usually, theslopes bottom out gently into sedimentary fans of u-vial deposits and debris ows on a narrow shoreline(marine terraces) of some hundred meters up to fewkilometres wide (Fig. 2).

The coastal desert of northern Chile is part ofthe stretched extremely arid zone extending alongthe western rim of South America, primarily inu-enced by steady subsidence processes in the domainof the Southeast Pacic subtropical anticyclone (e.g.,Garreaud 2009). The climate is characterized by a

moderate temperature regime, high atmospheric hu-midity (annual average: 65 - 80%) and rare rainfalls(Fig.1). Annual precipitation ranges from approx. 1 mmin the northernmost part (Iquique, Arica) to about 80mm in the south of the arid section. Occasional rain-

fall episodes occur mainly during austral winter (Mayto September) and are mostly associated with north-eastward moving extratropical low-pressure systems(richter et al. 1993; Garreaud and ruttlaNt 1996;FueNzalida et al. 2005). In the northernmost part of

the area around Arica (18S) episodic summer rainsalso occur.

A quasi-persistent climatic feature of the aridcoastal region is the presence of high fog (camanchaca).It develops best during the cold season, resultingmostly from an adjacent stratocumulus cloud deck inthe upper part of the marine boundary layer below apronounced temperature inversion (Fig. 2 left). Apartfrom advection fog, locally orographic (upslope) fogcan develop as well (cereceda et al. 2002). A sharptemperature increase within the inversion layer inhibitsthe vertical development of stratocumuli. Dependingon the season and latitude, they are conned to abelt between 300-600 m a.s.l. and 800-1200 m a.s.l.(Garreaud et al. 2008). The steep coastal mountainchain largely impedes cloud movement inland.

The high frequency and strong intensity of fogsrepresent a decisive ecological factor, since they en-sure the existence and survival of the Loma vegetationalong the hyperarid Peruvian and northern Chileancoast (richter 1981). Dense fog deck protects plantcover from direct solar radiation and desiccation andalong with relatively low temperatures and high at-mospheric humidity it also reduces evapotranspirationrates. Since many plants can strip moisture from fog

(Fig. 3), it further acts as an important water resourcefor certain taxa, particularly in the northern part ofthe coastal desert. At night, dew forms frequentlyon the ground and may also represent an importantsource of water, however its role for Loma ecosystems

Fig. 2: The orographical fog-belt along the coastal mountains of Pan de Azucar (2602S; left) and scattered cacti of mostlydead Eulychnia iquiquensisnear Cobija (2233S; right). Note the sedimentary fans below the steep escarpment, a typicallandform for most of the study area


5/19


is not studied yet. On the other hand, even the rarerain episodes play a signicant role for the long-termsubsistence of the vegetation. They facilitate the re-production, dispersal and establishment of annualsand perennials being of crucial importance for the re-

plenishment of their seed banks (dilloN and ruNdel1990). The importance of rains as main water resourceincreases towards higher latitudes.

4 Results of a comparative oristic survey onthe Chilean coastal desert

4.1 Characteristics of the Chilean Lomavegeta-tion

Systematic investigation of the northern Chileancoastal vegetation started in the mid 19 th centurywith the rst and one of the most extensive works on

the ora of the Chilean Loma byrudolPh PhiliPPi(1860). It was continued during the 20 th century byseveral researchers such as reiche (1907),JohNStoN(1929),WerderMaNN (1931), ricardi(1957), ruNdelet al. (1991), Muoz-Schicket al. (2001), and dilloN

(2005), among others. The results of these botanicalefforts reveal that more than 700 species of vascu-lar plants are native to the coastal area between 18Sand 2930S, which according to ruNdel et al. (1991and 2007) corresponds to the distribution area ofthe Chilean Loma vegetation. About two-thirds ofthem consist of perennials, which in contrast to thePeruvian Loma vegetation are usually the predomi-nant life-forms of the Chilean Lomas. Furthermore,the ora is characterized by a high number of endem-ics with a total of around 40% of the vascular plantsbeing restricted to the coastal desert of Chile.

From its southern part at around 30S, the Lomavegetation stretches as a relatively closed formation

Fig. 3: Droplet-catching lichens (upper left) and spines (upper right) of a Eulychniacactus in Pan de Azucar, where cactiare covered by thick lichen-curtains (lower left). Fog dripping from cacti facilitates the formation of small gardens intheir surroundings with Euphorbia lactiuaat Pan de Azucar (lower left) and Lycium leiostemumat Cerro Moreno (lower right)


6/19

174 Vol. 65 No. 2

alongside the windward escarpment of the coastalmountain range towards the equator (ruNdel et al.2007). North of 26S it shows a rather scattered pat-tern, and north of 2130S it concentrates on smallvegetation patches on few sites with most abundant

fog, usually on southern to western slopes. At north-ern and eastern hillsides with marginal fog impactand more intense insolation plant cover is normallysparse or absent. The same is true for areas situatedabove the stratus layer from 1000 - 1200 m a.s.l. up-wards. However, on the oor of dry valleys or small-er creeks contracted but dense vegetation is usuallypresent, proting from the conuence of the sloperunoff and hence, better hydrological conditions.

In most cases, the Lomas show a vertical dif-ferentiation of various plant communities. Thisresults primarily from varying fog frequencies (in-tensities) along the altitudinal gradient (ruNdel andMahu 1976; richter1995), though further climaticforces may play an additional role (e.g., temperature,intensity of drizzle from stratocumulus clouds).The middle zone of the fog-inuenced belt usuallysustains a relatively dense plant cover formed bycolumnar cacti and diverse, mostly high-growingshrubs. Frequently they are densely covered by epi-phytes (mainly by beard lichens, rarely byTillandsia

geissei ). The dominance of nano- and micro-phaner-ophytes in this altitudinal belt reects higher waterinputs and consequently higher phytomass produc-tion. Greater moisture supply results primarily from

higher fog frequency and intensity in the middle andupper fog zone (see cereceda et al. 2008; correa1990). Particularly high-growing plants are capableof effective stripping of fog water due to their sizeand larger total surface (elleNberG 1959) and thuscreate their own favourable microclimate with tinygardens of a beneting micro-ecosystem (Fig 3). The lower and upper boundary fog zone, markedby sharply decreasing fog frequency, as well as areasbelow the fog layer, are usually populated, if at all,by open xerophytic communities composed mostlyof dwarf shrubs and low-growing spherical cacti.However, such a simple relationship between hydro-

logical characteristics of fog as, e.g., water contentand vegetation properties does not exist when seenat a larger horizontal scale. Closer and species-richervegetation can be found in zones with relative littlemeasurable horizontal fog precipitation (e.g., Paposo,Cerro Perales, Fig. 1) while sparser, species-poorervegetation inhabit sites where the highest values ofhorizontal precipitation were measured (e.g., CerroMoreno, Iquique). These patterns suggest that oth-er factors besides fog play a signicant role for the

large-scale vegetation differentiation. Here, the lati-tudinal precipitation (and temperature) gradient isprobably of decisive importance.

4.2 Loma formations

Based on oristic features derived from an analysisof all perennial species from 29 selected Loma locali-ties (Fig. 4) and on physiognomic differences in plantcover, a classication of the Chi lean Loma vegetationis suggested. At least ve different plant formationscan be distinguished along the Chilean Loma region(Fig. 5), each of them with its respective spectrum ofplant communities. This differentiation reects inpart classications mentioned in the literature (e.g.,ruNdel et al. 1991; GaJardo 1994). General charac-teristics of these formations are briey described inthe following subchapters (details in Schulz 2009).

a) The formation of isolated fog oases(1830S-2130S, group I)

The vegetation in this part of the desert is re-stricted to few relatively small areas mostly on south-ern to western slopes of larger headlands, recognizedas zones of intense coldwater upwelling and frequentformation of dense orographic fog (baharoNa andGalleGoS 2000; cereceda et al. 2002). In total, nineisolated fog oases have been identied until now(PiNto and luebert 2009, additionally Pabellon de

Pica, 2053S). Towards the lower latitudes, they be-come species-poorer and decrease in size while baresections between the oases increase (SielFeld et al.1995). Nevertheless, up to 70 species have been re-corded in some of these areas up to now (PiNto andluebert 2009).

Plant-covered areas (at least those formed by per-ennials) range from the 350-600 m a.s.l. level up tothe cliff edge with their lower limit rising generallytowards north. egetation is mostly dominated byshrubs and dwarf shrubs, smaller cacti and at someplaces by the columnar cactus Eulychnia iquiquensis.The latter forms extensive stands with up to several

hundred individuals as for example at Cerro Camaraca(1838S), Pabellon de Pica (2053S) and Alto Chipana(2116S) (see also PiNto 2007). In the few years thathave abundant rain, a multiplicity of annual plantsand geophytes also appear, totally altering the aspectof vegetation for several months (Muoz-Schick etal. 2001). Generally, the plant cover is characterizedby a differentiation in species-poor communities atlower elevations and denser as well as species-richercommunities within the zone of intense fog (example


7/19


in Fig. 6a). Within the marginal upper fog zone, alsomonospecic stands ofTillandsia(mainlyT. landbecki)were found further away from the coast, forming ex-tended elds at elevations between 900 and 1200 ma.s.l. (ruNdel et al. 1997; PiNto et al. 2006).

b) Eulychnia iquiquensis succulent formation

(2130S-2430S, groups II and III)The southward adjacent Eulychnia iquiquensisfor-

mation is represented by a relatively continuous, butwidely scattered plant cover stretching alongside theseaward slopes at elevations between approx. 300 ma.s.l. and around 1100 m a.s.l. Characteristic for thisregion are relatively species-rich communities on up-per hillsides, formed byEulychnia iquiquensisand a vari-ety of (dwarf) shrubs (Fig. 6b). At the end of relatively wet years, many annual and perennial herbs enrich

these habitats. During dry periods the herb vegetationis rather sparse, mostly restricted to valley bottomsand gullies or forming micro-sites around columnarcacti (Fig. 3) and larger rocks, beneting from favour-able microclimatic conditions.

The density of plant cover and species richnessdeclines gradually towards the drier foot area, where

xerophytic shrubs prevail. In the southern part of thissection, the lower as well as the upper boundary areaof the fog zone is frequently dominated by relativelydense populations of spherical Copiapoacacti, some-times forming monospecic stands.

A quite conspicuous exception in this formationis constituted by the vegetation of Cerro Moreno,northwest of Antofagasta (2330S). Apart from be-ing species-richer and considerably more vigorousdue to specic climatic conditions (richter 1995;

100 75 50 25 0

Alto Patache (AP)

Punta Lobos (PL)

Punta Gruesa (PG)

Alto Chipana (AC)

Iquique (IQ)

La Chimba (LC)

Antofagasta (AN)

Cerro Moreno (CM)

El Cobre (EC)

Blanco Encalada (BE)

Tocopilla (TC)

Cobija (CB)

Cifuncho (CF)

Flamenco (FL)

Chaaral (CH)

Queb. Len (QL)

Obispito (OB)

Esmeralda (EM)

Pan de Azucar (PA)

Paposo (PP)

Taltal (TT)

Cerro Perales (CP)

Rinconada (RC)

Matancilla (MT)

Medano (ME)

Miguel Diaz (MD)

La Plata (LP)

El Tofo (ET)

Carrizal Bajo (CR)

(I)

(II)

(III)

(IV)

(V)

(VI)

decreasing similarity

2013

20212050

2102

2116

2205

2236

2330

2335

2340

2433

2455

2500

2525

2525

2507

2552

2606

2443

2450

2540

2620

2657

2633

2644

2413

2421

2810

2927

Fig. 4: Dendrogram of oristic similarity for selected Loma-localities (distance: Srensens coefcient, cluster algorithm:unweighted group average; only native perennial species were considered). The decoupling of group III and the displace-ment of Cifuncho to group V are probably explained by a relative small number of species at these sites and thus a smaller

percentage of common species with other localities of the same plant formation


8/19

176 Vol. 65 No. 2

eSPeJo et al. 2001), it includes patches of relict veg-etation preserved at very steep and foggy precipiceson western and south-western sites of the massif.Surprisingly, taxa such as Colliguaja odoriferaand eventhe ferns Polystichumsp. and Elaphoglossum gayanumoc-cur here as disjunct populations, far away from theirprincipal distribution areas in central and southernChile.

c) Northern Euphorbia lactiua-Eulychniaspp. suc-culent shrub formation (2430S-2620S, group I)

This formation occupies the central part of thecoastal desert. Without any doubt the vegetation ofthis section, especially its northern part (vicinity ofPaposo, 2500S), is considered an archetype of theChilean Lomas. Distinguished by remarkable spe-cies richness for such an arid region, it harbours ahigh percentage of the Chilean Loma endemics. Thissaid, it seems reasonable to regard its formations asa core area of the Chilean Lomas. Concerning the vegetation, this part of the coast is dominated bytwo characteristic elements, the high-growing shrubEuphorbia lactiuaand the columnar cactus Eulychnia(E. iquiquensisnorth of approx. 2530S and E. brevi-

orasouth of it). The most vegetated areas of the fogzone are comprised of numerous shrubs, herbs andvarious species of cacti. In areas of most intense fogmoistening, which becomes obvious by dense cov-ers of lichens on twigs and stones, also the epiphyteTillandsia geisseigrows on branches ofEulychnia, tallershrubs, and sometimes even on the ground and onrocky faces of foggy steeps.

Above and below the fog belt, the plant coverpasses through series of communities compounded

by more xerophytic shrubs and dwarf shrubs, fre-quently accompanied by the bromeliad Deuterocohniachrysantha and smaller cacti (Eriosyce and Copiapoa,the latter showing its diversication centre here; Fig.6c). However, higher inow of slope runoff duringrain events allows the development of surprisingly vigorous and species-rich azonal vegetation withabundant Euphorbia and Eulychnia on alluvial fansand older mudow cones at the foot of the mountainescarpment.

d) Southern Euphorbia lactiua-Eulychniaspp. suc-culent shrub formation(2620S~2730S, group ) This formation occupies only a relatively

small section of the coast. Although EuphorbiaandEulychnia are still the dominant elements, speciesrichness declines considerably in comparison withthe northerly adjacent region. Due to more regu-lar and abundant rainfalls, the vegetation cover ismore homogenous. The community of Euphorbiaand Eulychnia is widely spread through all elevationlevels (except at the foot of the mountain) and expo-sures. Here, it is less dependent of a spatially vari-able fog moisture distribution, and thus no fog in-

duced oristic differentiation of plant cover can bedistinguished in this section (Fig. 6d). Nevertheless,the high abundance of epiphytic beard lichens indi-cates a continued important inuence of fog. Mostof the shoreline and the at area below about 150m a.s.l. are populated by dune and gravel commu-nities (detailed information in kohler 1970) withabundant annuals forming the famous desiertoorido (owering desert) during extraordinarilymoist periods.

19 20 21 22 23 24 25 26 27 28 29 30 S

fog oases Eulychnia - formation northernEuphorbia - Eulychnia - formation

shrub formationsouthern

500

0

1000

1500

2000

2500

m

asl

Fig. 5: Latitudinal position of the Loma formations with respective approximate altitude of the vegetation belt.


9/19


11

12

Ephedra breana - Cumulopuntia sphaerica -community

Nolana sedifolia - Frankenia chilensis - communityCopiapoa calderana ssp. atacamensis -community

Eulychnia iquiquensis-Lycium leiostemum - community

1

2 3

5

Euliychnia iquiquensis-Copiapoa calderana -community

4

Copiapoa calderana -community3

Haplopappus foliosus - Bahia ambrosioides- communityPleocarphus revolutus - Eulychnia acida -community

Oxalis virgosa - Balbisia peduncularia - community

Pleocarphus revolutus - Eulychnia acida -community

11

12

13

12

13

12

Euphorbia lactiflua - Eulychnia iquiquensis -community

Deuterocohnia chrysantha - Copiapoacinerascens - community

Heliotopium pycnophyllum -

Tetragonia maritima - community

Euphorbia lactiflua - Eulychnia breviflora - community

Skytanthus acutus - Heliotropium floridum- community7

8a

6

8b

9

8b

97

8a

6

1

2

3

4

5

3

a) b)

e) f)

Alto Patache Cerro Moreno (sw escarpment)

El Tofo northwestern escarpmentEl Tofo southwestern escarpment

FlamencoPan de Azucar

900

800

700

600

500

400

300

200

100

1000

1100 1100

1000

900

800

700

600

500400

300

200

100

0 0

m a.s.l. m a.s.l.

1100 1100

m a.s.l. m a.s.l.

1100 1100

m a.s.l. m a.s.l.

900

800

700

600

500

400

300

200

100

1000

900

800

700

600

500

400

300

200

100

1000

0 0

800

700

600

500

400

300

200

100

900

1000

900

800

700

600

500

400

300

200

100

1000

0 0

2 cacti; shrubs: Eulychnia iquiquensis(mainly dead); Nolana sedifolia,Frankenia chilensis, Lycium cf. deserti(mainly dead), Ephedra breanaherbs:Alstroemeria lutea, Oxalis bulbocastanum

1 cacti; shrubs:Cumulopuntia spaerica, Ephedra breana,

Lycium cf. deserti(mainly dead)

4 & 5 cacti; shrubs: Eulychnia iquiquensis(partly dead), Eriosyce recondita;Lycium leiostemum, Ephedra breana, Nolana cf. lachimbensis, N. sedifolia,

Heliotropium eremogenum, Ophryosporus triangularis,

Tetragonia angustifolia, Baccharis taltalensis

herbs: Cistanthe sp., Jarava sp., Nassella pungens,

Polyachyrus fuscus, Tigridia philippiana

3 cacti; shrubs: Copiapoa calderanassp.atacamensis;Heliotropium pycnophyllum, Nolana peruviana

2050 S 2329 S

2926 S 2929 S

2602 S 2632 Sc) d)

8a cacti; shrubs:Echinopsis deserticola, Eulychnia iquiquensis; Euphorbialactiflua, Frankenia chilensis,Lycium deserti, Nolana divaricata,N. incana, Oxalis gigantea, Puya boliviensis, Suaeda foliosa

annuals:Polyachyrus cinereus, Cistanthe grandiflora,Leucocoryne appendiculata, Senecio cachinalensis

6 & 7 cacti; shrubs: Copiapoa cinerascens; Atriplexsp., Bakerolimon plumosum,Deuterocohnia chrysantha, Heliotropium pycnophyllum, Nolana sedifolia,Tetragonia maritim

8b cacti; shrubs:Eulychnia breviflora, Eriosyce rodentiophila; Euphorbia lactiflua,

Lycium sp.,Nolanacf.divaricata, Oxalis giganteaherbs: Cistanthe longiscapa, Polyachyrussp.

9 shrubs: Encelia canescens, Heliotropium floridum, Skytanthus acutus,Tetragonia maritima

herbs: Oenothera coquimbensis, Tiquilia litoralis

11 cacti; shrubs:Echinopsis deserticola, Copiapoa coquimbana; Bahia

ambrosioides, Baccharis paniculata, Fuchsia lycioides, Haplopappus foliosus,

Lobelia polyphylla, Oxalis virgosa, Polyachyrus poeppigii

herbs: Calceolaria glandulosa, Nassella sp., Poa holciformis

12 cacti; shrubs:Eulychnia acida; Bahia ambrosioides, Balbisia peduncularis,

Frankenia chilensis, Heliotropium stenophyllum, Nolana coelestis,

Ophryosporus triangularis, Pleocarphus revolutus

herbs:Bromus berteroanus, Helenium urmenetae, Loasa tricolor,Triptilion gibossum, Cristaria aspera

13 cacti; shrubs:Echinopsis deserticola, Eulychnia breviflora; Bahia ambrosioides,

Balbisia peduncularis, Fuchsia lycioides, Haplopappus foliosus,

Heliotropium stenophyllum, Mycrianthes coquimbensis, Oxalis virgosa,

Senecio glabratus, Senna cumingii

herbs:Jarava plumosa, Nassella sp.12 cacti; shrubs:Eulychnia acida; Bahia ambrosioides, Balbisia peduncularis,

Frankenia chilensis, Heliotropium stenophyllum, Nolana coelestis,

Ophryosporus triangularis, Pleocarphus revolutus

herbs:Bromus berteroanus, Helenium urmenetae, Loasa tricolor,Triptilion gibossum, Cristaria aspera

Fig. 6: Vertical distribution of plant communities within the Loma-belt at selected localities. a) Alto Patache; b) CerroMoreno (only SW- and W-slopes); c) Pan de Azucar; d) Flamenco; e) and f) El Tofo (different slope orientation). Mostabundant species are indicated below the respective sketches. Letters in grey indicate that most of the character istic com-munity members are dead


10/19

178 Vol. 65 No. 2

e) arious shrub formations of the southernLoma region (2730S-2930S, group I)

Towards the south, the coastal desert harboursan increasing number of plant communities, de-scribed in detail among others bylailhacar (1986).

A common feature of this section is an increasedabundance of tall shrubs, mostly still accompaniedby succulents. The higher percentage of speciesfrom central Chile suggests a stronger oristic afn-ity to the vegetation of the semiarid and semihumidcoast. The community of Haplopappus angustifoliumdistributed in the strongly fog-inuenced summitarea of Cerro Negro (2810S) is a good example forthis fact since it includes several oristic elementsof the vegetation south of 30S (e.g., Gochnatia fo-liolosa, Baccharis vernalis, Senecio coquimbensis ). Similarto the ora of the famous National Park Fray Jorge(3030), the southern oral elements subsist here atclimatically favoured sites as relicts of a supposedformer vegetation community that extended fur-ther northward from moister parts of the country.Despite the considerably increased importance ofrelatively regular winter rains, the role of fog formicroclimate regulation and as an additional waterresource (araveNa et al. 1989) is still signicant.

In this southernmost section, differentiation inthe horizontal vegetation structure between N- andS-slopes is more pronounced than a vertical differ-entiation. Widely spread is the Oxalis spp.- Balbisia

peduncularis community with a high abundance of

various shrubby and herbaceous species. This plant

ensemble is found on more sun-exposed hillsides.Instead, the succulent-poor Haplopappus foliosuscom-munity prefers S- and SW-slopes in the southernpart of the region (El Tofo, 2927S; Fig. 6e and 6f).Once again, the foot zone of the escarpment is pop-

ulated by communities of dunes and marine terracescontaining numerous annual plants.

5 Vegetation change

Concerning the apparent vegetation declinein the northern part of the area, the dieback ofEulychnia iquiquensis is most frequently mentioned(krauS 1994; FollMaNN 1995; richter 1995;ruNdel et al. 1997; PiNto et al. 2001; Muoz-Schicket al. 2001; PiNto 2007; PiNtoand kirberG2009). An overwhelming part of these high-grow-ing columnar cacti, which reach impressive sizesof up to seven meters, are in extremely poor condi-tion along their northern distribution zone betweenBlanco Encalada (2421S) and Arica (18S). Mostof them are desiccated and collapsed, and retainonly the skeleton of the inner tissue of stems andlarger branches (Fig. 7). The occurrence of still vi-tal individuals is mostly constrained to higher el-evations that are strongly inuenced by fogs. Thereproduction rate in affected populations is low ornone, at least in the northern part of this region, andthe vegetative activity is restricted to the few years

with abundant rainfall (PiNto 2007).

Fig. 7: Collapsing individuals ofEulychnia iquiquensisin the lower belt of columnar cacti at around 500 m a.s.l. on the south-ern escarpment of Cerro Moreno (left) and extensive dieback of the same species on the escarpment above Cobija (right).


11/19


A considerable number of further cacti appar-ently suffer in a similar way, such as various speciesof the genera Copiapoa (C. humilisssp. tocopillana, C.solaris) and Eriosyce, (E. iquiquensis, E. islayensis, E.

paucostatassp. echinus) as well as Cylindropuntia tuni-

cata, Haageocereus australisand Cumulopuntia sphaerica(belMoNte et al. 1998; PiNtoand kirberG 2005;personal observations). The loss of vitality of thesetaxa seems to be a recent process, since at leastsome of them showed noticeably better conditionsaround the middle of the last century. This is truefor example, for populations ofEriosyce islayensisatPoconchile (hinterland of Arica), which were stillcomposed of vigorous, prospering individuals inthe 1950s, but meanwhile all of them are desic-cated or dead (compare ritter 1980 versus PiNtoand kirberG 2005). ritter (1980) noted with me-ticulous precision in his detailed book on cacti ofnorthern Chile (eld work in the 1950s and 60s)each remarkable peculiarity of the studied species(e.g., endangerment by diseases, drought damages) without, however, mentioning the conspicuouspoor condition of the same species affected nowa-days. Recent dieback processes are also conrmedby various older inhabitants of the coastal zone.

Many shrubs and perennial herbs show cor -responding signs of decline along the region be -tween Iquique and Antofagasta. Plants with ac-tive vegetative organs within affected populationsoccur only sporadically and are mostly restricted

to higher elevations, dry valleys and gullies. Thissituation stands in a striking contrast to vegetationdescriptions by researchers and travellers up tothe mid-20th century. For example, reiche (1907),WerderMaNN (1931) andJohNStoN (1929) ment ionrather rich plant formations consisting of annualand perennial herbs and cacti above Iquique dur-ing the rst decades of the previous century, whileit is nearly bare of vegetation today (PiNto andluebert 2009).

Furthermore, JaFFuel (1936), reiche (1907),JohNStoN (1929) and barroS (1941) registered arelatively species-rich shrub and herb vegetation in

the vicinity of Tocopilla (2205S) apparently ap-pearing in an almost annual rhythm. According to various historical travel reports from the 18th and19th century, an abundant herb and shrub vegeta-tion also prospered in Cobija (2236S) during the winter and spring periods and provided sufcientfodder for limited animal husbandry (Frezier1713; Feuille 1714; caete 1787; ocoNNor 1825all cited in bittMaNN 1977). Today, an overwhelm-ing part of shrubs and cacti around Tocopilla and

Cobija is desiccated or dead. Despite intensive bo-tanical eld research, many shrub and herb speciescollected at Tocopilla and Cobija until the mid-20 th century were not registered during the pastfew decades (see also luebert et al. 2007). Among

them Adiantum chilense, Alstroemeria violacea, Bahiaambrosioides, Calceolaria paposana, Centaurea cachinalen-sis, Cheilanthes mollis, Chuquiraga ulicina, Cylindropuntiatunicata, Nolanacf. deexa, and Ophryosporus anomalusare to be mentioned for Tocopilla. Further south atCobija, Conanthera campanulata, Copiapoa humilisssp.tocopillana, Ephedra chilensis, Frankenia chilensis, Nolanadiffusa, N. cf. deexa, N. sedifolia, Senna brongniartii,

Alstroemeria violacea, Ophryosporus anomalus, Polyachyrusfuscus, and Puya boliviensis seem to have been lost,although the occurrence of many of these species was registered as frequent in collections of W.bieSe in 1949 (herbarium SGO). Even if the recentabsence of the aforementioned and further speciesdoes not automatically prove their denite disap-pearance from the area, differences in the formerand recent oristic composition are obvious.

The local lack of formerly abundant taxa goesalong with a recent reduction of the distributionarea of some species. PhiliPPi (1860), JohNStoN(1929) and FollMaNN (1967) registered for exam-ple the occurrence ofOxalis giganteafor the north-ern areas from Miguel Diaz (2433S) up to CerroMoreno (2329S). According to heibl (2005) itspresent distribution limit is located far further

south around Quebrada El Medano (2450S).PhiliPPi (1860) and ritter (1980) mentionedthe cactus Cylindropuntia tunicata for the zone be-tween El Cobre (2413S) and Arica (1820S).Interestingly, the second author noted that manyof its individuals around Tocopilla and Arica didnot survive the period of a severe drought aroundthe mid-20th century (ritter 1980). Currently, thenorthern distribution limit of this coastal speciesseems to be reduced to Miguel Diaz (2432), about700 km further south.

According to some researchers, lichen vegeta-tion has also impoverished along the coastal region.

Between 1965 and the early 1990s, repeated oristicsamplings byFollMaNN (1995) document a drasticreduction of species richness of up to 39-46% atselected study sites. However, this phenomenon isnot only restricted to the north, but was also not-ed in the central coastal region of Chile (3230S).Likewise, a decline of epiphytic lichens as well asof the vascular epiphyte Tillandsia geissei has beenreported for the past few decades (ruNdel et al.1991; ruNdel and dilloN 1998).


12/19

180 Vol. 65 No. 2

Changes in vegetation are apparently associ-ated with losses in the regional fauna. arious au-thors mentioned the occurrence of Guanaco popu-lations (Lama guanicoe ) for the coastal mountainsbetween Iquique and Antofagasta up to the recent

past (bauver 1707 and Feuillee 1714 in bittMaNN1977; MaNN et al. 1953; Nuez and varela 1967).Well preserved traces of their past presence can stillbe detected along the entire northern coastal zone(richter 1995; larraiNet al. 2001). In the south- wards adjacent area, at places where a denser andmore vigorous vegetation provides sufcient feedand the decimation by man is minimal, Guanacossustain their existence until today (e.g., north ofPaposo, Pan de Azucar etc.). While this suggests thatat least until the middle of the past century sufcientresources in form of lichens, cactus fruits and moreabundant herbs and shrubs ensured the long-termsurvival of former populations around and north of Antofagasta, these preconditions are not given anylonger.

6 Recent climate trends

Regarding the spatial dimension of the vegeta-tion decline and the number of affected taxa, large-scale changes of environmental conditions aresupposed to be a likely cause of the deteriorationprocess. In particular, changes in precipitation, fog

and cloudiness as principal factors inuencing theLoma ecosystems have to be considered as poten-tial triggers for dieback effects and thus are stud-ied in detail. Owing to difculties cited in chap-Owing to difculties cited in chap-ter 2, no trend analyses were performed for fog.Nevertheless, it is to be expected that tendencies inlow-level cloudiness reect to some degree possiblegeneral tendencies in fogs, since the predominantadvection fog is formed by coastward moving stra-tocumulus clouds.

6.2 Precipitation

In spite of very low amounts, long-term rainfallrecords reveal pronounced low-frequency changeson the interdecadal time scale over the past century(see Fig. 8), closely linked to the Interdecadal PacicOscillation (IPO, i.e., large-scale multidecadal cli-mate variability in the Pacic region, e.g., FollaNdet al. 1999). Two relatively humid and two drierperiods are distinguishable; a humid phase fromthe 1920s to the mid-1940s and from the mid-1970s

until the end of the century, associated with positive(warm) IPO-phases, as well as a dry phase fromthe mid-1940s to the early 1970s and a short onearound 1910, related to negative (cold) IPO-phases.A quite distinctive feature of the latter dry phase is

the occurrence of a prolonged extremely dry period

a

nnualprecipitation

[m

m

]

20

0

0

40

60

80

100

120

140Copiap

50

100

150

200

250

300

350

400

450La Serena

10

20

30

40 57mmAntofagasta

0

5

10

15

20

25

1 90 0 1 91 0 1 92 0 1 93 0 1 94 0 1 95 0 1 96 0 1 97 0 1 98 0 1 99 0 2 00 0-3

-2

-1

0

1

2

3

4Arica

Iquique

5

10

15

20

25

1 90 0 1 91 0 1 92 0 1 93 0 1 94 0 1 95 0 1 96 0 1 97 0 1 98 0 1 99 0 2 00 0-3

-2

-1

0

1

2

3

4

1 90 0 1 91 0 1 92 0 1 93 0 1 94 0 1 95 0 1 96 0 1 97 0 1 98 0 1 99 0 2 00 0

-3

-2

-1

0

1

2

3

4

1 90 0 1 91 0 1 92 0 1 93 0 1 94 0 1 95 0 1 96 0 1 97 0 1 98 0 1 99 0 2 00 0

-3

-2

-1

0

1

2

3

4

1 90 0 1 91 0 1 92 0 1 93 0 1 94 0 1 95 0 1 96 0 1 97 0 1 98 0 1 99 0 2 00 0

-3

-2

-1

0

1

2

3

4

0,8 1,4 1,5

3,1 1,2 1,0

9,3 1,2 3,2

35 10 20

118 88 80

IPO-Index

Fig. 8: Annual rainfal l at ve northern cl imate stations for19002008 (bars) and IPO-index (line) for the same period,smoothed by an 11-yr low pass HadSST2 Chebyshev lter(http://www.iges.org/c20c/IPO _v2.doc). Rainfall aver-ages derived from annual precipitation for three periodscorresponding to distinct IPO-phases (192045, 194675,19762001) are indicated at the top of each graph. Greybands indicate periods of missing data. Note the differentscaling for precipitation on the y-axis


13/19


around 1950, when absolutely no rain was registeredalong the coastal desert north of Antofagasta formore than a decade (Fig. 8).

During the previous century, the interdec-adal precipitation variability, however, was super-

imposed by further changes of longer time scale,manifested in differences of mean annual rainfallamounts between the two more humid phas-es (Fig. 8), the latter being notably drier than theprevious one (exception: Arica). This variation inmultiannual rainfa ll means reveals a downward pre-cipitation trend over the whole period 1900-2008prevailing in most parts of the study area. Analysesof daily rainfall events (1 mm) indicate that thedifference resulted from a reduction of the meanrainfall intensity and less frequent rainy days dur-ing the latter humid period (Tab.1). The reduc-tion of rain frequency is particularly pronouncedat Iquique, where the mean annual number of dayswith rainfall 1 mm decreased from 1.2 during theperiod 19201945, to just 0.3 during 19762001. Atleast for the drier northern part of the desert, thisfact implies that there has been a sustained decreasein the frequency of rainfall events since the mid-century, probably as a result of a declining cyclonicactivity at lower latitudes (see Schulz et al. 2011).

Due to the lack of long rainfall records, it isdifcult to assess when this negative trend set in.However, longer rainfall series of Copiap and LaSerena in the central-southern portion of the desert

reveal a rather extended desiccation trend since atleast the late 19th century, though strongly modu-lated by signicant variability at the interdecadaltime scale (Fig. 9). The prevalence of more humidconditions prior to 1900 seems to be also supportedby palaeoclimatological studies indicating wetterconditions in central(-northern) Chile during thepast three to four centuries, especially in the 19thcentury (valero-GarceS et al. 2003; le queSNe etal. 2008).

6.3 Cloudiness

Analyses of the total cloud cover reveal a ratherstrong and signicant decline of cloudiness sincethe mid-1970s, particularly at Arica (Fig. 10). The

intensity of this trend decreases towards the higherlatitudes alongside the coastal desert. In Antofagasta,after a phase of increasing total cloud cover from themid-1960s to the mid-1970s and an abrupt decreasetowards the end of the 1970s, a rather stable regimeseemed to establish itself. Notwithstanding, studieson the stratocumulus deck show a slight but persist-ent decrease in stratocumulus cloud cover since themid-1970s in Antofagasta and to a lesser degree inLa Serena since at least 1980 (quiNtaNa and berrioS2007; berrioS 2008). In Arica and Iquique, in con-trast, no coherent tendencies were found in strato-cumulus cover during the period of 1980-2005.Furthermore, for the period 1960-2005, a notable,statistically signicant decline in the frequency of lowclouds occurrence was detected in Arica and a sta-tistically signicant increase in Antofagasta (data notshown here). The frequency tendency in low cloudslikely implies a downward trend in advection fogfrequency in Arica, while the opposite tendencies instratocumulus cover and frequency in Antofagasta donot permit any coherent conclusions about changes inhigh fog in this part of the desert. As fog character-istics are determined by a complex coaction of differ-ent factors, further studies are required for the analy-

sis of fog evolution during the recent past decades.The remarkable decrease of the total cloud coverat Arica is consistent with changes in the mean sea-sonal and annual frequency of cloudy and cloudlessdays (all types of clouds), which experienced a notablereduction of 45 days and a strong increase of 41 days,respectively, between the periods 19571976 and19772008 (Tab. 2). Changes of the same direction,although of lower magnitude, were registered also inAntofagasta.

Period Iquique Antofagasta Copiap La Serena

Mean number of days of rainfall

1.0mm

19201945 1.2 1.3 3.2 10.7

19461975 0.3 0.5 1.6 8.6

19762001 0.3 0.9 2.0 7.0

Median of daily rainfall 1.0mm

(in mm)

19201945 1.8 3.9 7.5 5.5

19762001 1.6 1.7 4.4 4.1

Tab.1: Frequency of wet events (mean number of days with rainfall amounts 1.0 mm) and median of daily rainfall atselected stations calculated for various periods.


14/19

182 Vol. 65 No. 2

7 Vegetation change driven by climate change?

Taking the desiccation symptoms of the plantcover and the rainfall as well as cloudiness declineinto account, changes in climate during the pastdecades must be considered as the decisive force forthe creeping dieback processes affecting the waterbalance and reducing the regeneration capacity ofplants. Although an extreme scarcity of rainfallsand the scantiness of their amounts are a specicfeature of the coastal desert in northern Chile, ex-

traordinary prolongations of dry periods such asaround the 1950s may complicate the establishmentof seedlings and a periodic revitalisation of the per-ennial vegetation. Taking Tocopilla as example, ta-ble 3 manifests the connection between the recent vegetation decline and decreasing rainfall events.Although signicant rain episodes occurred nearlyannually in the rst half of the past century, rain-

fall was registered only once (2002: 16 mm) over thecourse of the recent period 1994-2005. Likewise,rainfal l (> 1mm) occurred in Iquique in only one ofthe 16 years during the period 1993-2008 (Fig. 8). Inthe surroundings of both sites, rare rainfall eventsprovoke an exuberant sprout and bloom of annualherbs as well as an increased vitality of the perennialvegetation (Muoz-Schicket al. 2001; hoxey2004;H. larraiN, Universidad Bolivariana, Iquique, pers.com.), still noticeable during the following years(PiNto et al. 2001). Whereas in the surroundings of

Tocopilla, researchers found excellent conditions tocompile extensive plant collections within few hoursduring the wetter period at the beginning of the pastcentury, nowadays an analogous procedure requiresseveral days to get a relatively decent plant compila-tion, pointing to a relation between the recent rain-fall decline and the poor preservation status of thepresent vegetation. In addition to the drought-pro-

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

1875 1885 1895 1905 1915 1925 1935 1945 1955 1965 1975 1985 1995 2005

La Serena

Copiap

Fig. 9: Annual rainfall in La Serena (grey line) and Copiap (black line), normalized and smoothed by an 11-yr movingaverage.

3

3.5

4

4.5

5

5.5

6

1960 1970 1980 1990 2000

AricaAntofagasta

Fig. 10: Annual total cloud cover in Arica (black line) and Antofagasta (grey line).


15/19


voked effects, the reduced number of cloudy daysin the northern part of the coastal desert impliesa longer and more frequent exposure of the Loma vegetation to intense solar radiation, which mayenhance and prolong hydric stress by elevated tran-spiration rates. Trends in stratocumulus clouds alsosuggest a decrease in fog frequency in the northernarea (Arica) that is expected to have similar effectson the water balance of plants. Despite little being

known about the southern area, possible changesin fog characteristics not necessarily directly linkedto stratocumulus cover and frequency are not ex-cluded, and could be a further factor contributing tochanges in vegetation in the study area.

Apart from climate change, in few coastal sec-tions the decline of plant cover and species richnessmay suffer from enhanced vulnerability to furtherstressors such as air pollution. Investigations of theenvironmental conditions around copper depositsat Paposo conrm negative effects of mining onplants in this area since especially the emission ofdust particles occludes leaf stomata (ruelle 1995).

Heavy metals also might affect plant growth, eventhough some species indicate extraordinary toler-ance against elevated concentrations (e.g., Nolanadivaricata, ruelle 1995). One more negative effecton vegetation must be expected from coal dustemissions of a thermoelectric plant in Tocopilla, which strongly contaminates air in the surround-ings (DICTUC 2006). The effects of pollution are,notwithstanding, limited to the proximity of pollu-tion sources and do not explain the poor conserva-

tion status of plants in areas with little or no miningactivity such as e.g., the entire coastal area north ofIquique.

8 Conclusions and outlook

The results of the study indicate a strong declineof the Loma vegetation in the area north of 2430S

in part icular during the past ve decades. It is mani-fested by decreasing plant vitality, the disappearanceof various taxa from their northern distribution areaand a complete or partial dieback of individual plantpopulations. Particularly cacti and epiphytes are af-fected, and probably some perennial shrub and herbspecies. These vegetation changes are further like-ly responsible for the loss of Guanaco populationsfrom the arid coastal area between 20S and 2330S.

Since different plant species are affected, large-scale pathogenic effects as well as natural popula-tion oscillations (e.g., cohort dieback) can be widelyexcluded as crucial agents. Based on the current state

of knowledge, plant cover and species richness aremost likely affected by the recent changes in climate,i.e., by reduced precipitation inputs and frequenciesas well as by prolonged insolation phases due to di-minished cloudiness in the northernmost area. Theseclimatic effects are expected to have a negative im-pact on the water balance, reproduction and preser-vation of plants. At the local scale, other factors suchas environmental pollution might have additionallycontributed to a degradation of the vegetation.

Number of days with total cloud cover >6/8 Number of days with total cloud cover


16/19

184 Vol. 65 No. 2

In the context of prospective projections, thequestion about the resilience capacity of the affect-ed vegetation arises. According to current regionalclimate projections, the rainfall decline is expected

to continue at the arid coast of Chile towards theend of the century (FueNzalida et al. 2006). Apartfrom that, the IPO has shown predominantly nega-tive values since the end of the 20th century, sug-gesting that a shift towards a negative phase is un-derway and drier conditions will predominate at thearid coast of northern Chile during the next dec-ades. Consequently, chances for an improvement ofgrowing conditions of the vegetation are limited inthe near future and a tendency towards further im-poverishment of the northern Loma formations isto be expected, which would endanger the subsist-ence of this oristically diverse and highly sensitive

ecosystem.In the light of these projections, there is an in-

creasing need nding a way to preserve these endan-gered endemic species. Irrigation at selected sitesusing fog collectors is a potential in-situ techniqueand has shown effective results (PiNto and kirberG2009), but requires considerable manpower and -nancial input for a continuous maintenance of theinstallations. Another possibility is offered by ex-situcultivation or the establishment of seedbanks. The

latter one is already practiced by the Royal BotanicGardens in collaboration with INIA (Chile) in partsof the desert. Aside from focusing on northernLomas, also the southernmost parts of the desertecosystem need a more systematic incorporation

into these conservation strategies, since these areasstill host a variety of further endemic species thatare increasingly endangered by emerging land useactivities.

However, these strategies must be consideredonly a drop in the bucket since climate change is (andalways has been) a worldwide phenomenon that canhardly be countervailed by local actions.

Acknowledgements

The authors gratefully thank Gina Arancio,Melica Muoz-Schick, and Clodomiro Marticorenafrom the herbaria CONC, HULS, and the NationalMuseum of Natural HistorySGO for providing ex-tensive data on plant collections from the northernChilean coast. Gina Arancio and Helmut Walterare thanked for identication of the collected ma-terial. The Direccin Meteorolgica de Chile andJuan Quintana were generous supporters of the dataon precipitation. Special thanks go to the GermanNational Merit Foundation (Studienstiftung desDeutschen olkes) for a full-time grant for N.Schulz and to the Department of Geophysics of the

Universidad de Chile for the support of this work.

References

alexaNderSSoN, h. (1986): A homogeneity test applied toprecipitation data. In: Journal of Climatology 6, 661675. DOI: 10.1002/joc.3370060607

alMeyda, E. (1948): Pluvometra de las zonas del desierto y lasestepas clidas de Chile. Editoral Universitaria. Santiago.

araveNa, r.; Suzuki, o. and PollaStri, a. (1989): Coastalfog and its relation to groundwater in the I regionof northern Chile. In: Chemical Geology (Isotope

Geoscience Section) 79, 8391. DOI: 10.1016/0168-9622(89)90008-0

baharoNa, M. and GalleGoS, r. (2000): Surgencias en lacosta norte de Chile durante las temporadas Nia 19961997 y Nio 19971998. In: Revista de Geografa NorteGrande 27, 5360.

barroS, e. (1941): Excursion botanica a la provincia de An-tofagasta. In: Revista Universitaria 26 (2), 8388.

belMoNte, e.; FauNdez, l.; FloreS, J.; hoFFMaNN, a.; Mu-Noz, M. and teillier, S. (1998): Categorias de conser-

Year Rainfall

(mm)

Year Rainfall

(mm)

1925

19261927

1928

1929

1930

1931

193239

1940

1941

1942

1943

1944

1945

1946

1947

1948

1949

1950

1951

1952

9.1

0.55.8

6.0

0.0

15.0

0.0

no data

11.0

5.0

9.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

1.0

195263

19641965

1966

1967

1968

1969

1970

197193

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

no data

0.014.0

1.2

6.0

0.0

4.0

0.0

no data

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

16.0

0.0

0.0

0.0

Tab.3: Annual rainfall in Tocopilla.
http://dx.doi.org/10.1002/joc.3370060607http://dx.doi.org/10.1016/0168-9622(89)90008-0http://dx.doi.org/10.1016/0168-9622(89)90008-0http://dx.doi.org/10.1016/0168-9622(89)90008-0http://dx.doi.org/10.1016/0168-9622(89)90008-0http://dx.doi.org/10.1002/joc.3370060607


17/19


vacion de cactaceas nativas de Chile. In: Boletin MuseoNacional Historia Natural 47, 6989.

berrioS, P. (2008): Estratocmulos en la costa norte deChile: variabilidad y tendencia. Tesis Universidad de al-paraiso. alparaiso.

bittMaNN, b. (1977): Notas sobre la poblacion de la costadel Norte Grande chileno. Universidad del Norte. An-tofagasta.

cereceda, P.;oSSeS, P.; larraiN, h.; FariaS, M.; PiNto, r.and ScheMeNauer, r. S. (2002): Advective, orographicand radiation fog in the Tarapac region, Chile. In: At-mospheric Research 64, 261271. DOI: 10.1016/S0169-8095(02)00097-2

cereceda, P.; larraiN, h.; oSSeS, P.; FariaS, M. and eGaa,i. (2008): The spatial and temporal variability of fog andits relation to fog oases in the Atacama Desert, Chile.In: Atmospheric Research 87, 312323. DOI:10.1016/j.atmosres.2007.11.012

correa, H. F. (1990): Caracterizacin y evaluacin delfenomeno de la camanchaca en la III Regin de Ata-cama. Memoria, Universidad de Chile, Facultad de Cien-cias agrarias y forestales. Santiago.

DICTUC (2006): Anlisis de la calidad del aire para MP-10 en Tocopilla. Informe Final. Ponticia UniversidadCatlica. Santiago.

dilloN, M. o. (2005): LOMAFLOR, Searchable Database. Andean Botanical Information System. http://www.sacha.org

dilloN, M. o. and hoFFMaNN, a. e. (1997): Lomas forma-tions of the Atacama Desert, Northern Chile. In: daviS,

S. d.; heyWood, v. h.; herrera-Macbryde, o.; villa-loboS, J. and haMiltoN, a. (eds.): Centres of plant di-versity: a guide and strategy for their conservation. ol.3: The Americas. Cambridge, 528535.

dilloN, M. o. and ruNdel, P. W. (1990): The botanical re-sponse of the Atacama and Peruvian Desert ora to the198283 El Nio Event. In: GlyNN, P. W. (ed.): Glo-bal ecological consequences of the 198283 El Nio-Southern Oscillation. Elsevier Oceanography Series.Amsterdam, 487504.

elleNberG, h. (1959): ber den Wasserhaushalt tropischerNebeloasen an der stenwste Perus. In: BerichteGeobotanische Forschung des Institut Rbel fr 1958.

Zrich, 4774.eSPeJo, r.; deMerGaSSo, c.; GalleGuilloS, P.; PiaNtelli,

e. and eScudero, l. (2001): Climatolgcal and microbio-logical characteristics of the camanchaca phenomenonat Cerro Moreno, Antofagasta, Chile. In: Proceedings ofthe Conference on Fog and Fog Collection, St. Johns,Canada, 463466.

FollaNd, c. k.; Parker, d. e.; colMaN, a. and WaShiNG-toN, r. (1999): Large scale modes of ocean surface tem-perature since the late nineteenth century. In: Navarra,

a. (ed.): Beyond El Nio: Decadal and Interdecadal Cli-mate ariability. Berlin, 73102.

FollMaNN, G. (1967 ): Die Flechenora der nordchile-nischen Nebeloase Cerro Moreno. In: Nowa Hedwigia14, 215281.

(1995): On the impoverishment of the lichen ora and theretrogression of the lichen vegetation in coastal centraland northern Chile during the last decades. In: Crypto-In: Crypto-gamic Botany 5, 224231.

FueNzalida, h.; SaNchez, r. and Garreaud, r. (2005):A climatology of cut off lows in the Southern Hemi-sphere. In: Journal of Geophyical Research 110, D1801,DOI: 10.1029/2005JD005934

FueNzalida, h.; Falvey, M.; roJaS, M.; aceituNo, P. andGarreaud, r. (2006): Estudio de la variabilidad climti-ca en Chile para el siglo XXI. Informe nal. DGF Uni-Informe nal. DGF Uni-versidad de Chile.

GaJardo, r. (1994): La vegetacin natural de Chile. Clasi-cacin y distribucin geogrca. Editoral Universitaria.Santiago.

Garreaud, r. (2009): The Andes climate and weather. In:Advances in Geoscience 22, 311. DOI: 10.5194/adg-eo-22-3-2009

Garreaud, r. and rutllaNt, J. (1996): Anlisis meteorolg-ico del los aluviones de Antofagasta y Santiago de Chileen el periodo 19911993. In: Atmsfera 9, 251271.

Garreaud, r.; barichivich, J.; chriStie, d. a. and Maldo-Nado, a. (2008): Interannual variability of the coastalfog at Fray Jorge relict forests in semiarid Chile. In: Journal of Geophysical Research 113, G04011. DOI:

10.1029/2008JG000709Grau, J. (2000): El Nio Leben fr die untergehendePanzenwelt der Atacama. In: Biologie in unserer Zeit30, 413.

heibl, c. (2005): Studies on the systematics, evolution, andbiogeography of Oxalissections Caesiae, Carnosae, and Gi-ganteae, endemic to the Atacama Desert of northern Chile.Diploma thesis. Ludwig-Maximilian-Universitt Mnchen.

hoFFMaNN, a. andWalter, h. (2004): Cactaceas en la orasilvestre de Chile. Fundacin Claudio Gay. Santiago.

hoxey, P. (2004): Some notes on Copiapoa humilisand thedescription of a new subspecies. In: British Cactus andSucculent Journal 22, 2942.

JaFFuel, P. F. (1936): Excursiones botnicas a los alrededoresde Tocopilla. In: Revista Chilena de Historia Natural 40,265274.

JohNStoN, i. M. (1929): Papers on the ora of north-ern Chile. 1. The coastal ora of the departments ofChaaral and Taltal; 2. The ora of the nitrate coast. In:Contributions Gray Herbarium 85.

kohler, a. (1970): Geobotanische Untersuchungen in s-Geobotanische Untersuchungen in s-tendnen Chiles zwischen 27 und 42 Grad sdl. Breite.In: Botanische Jahrbcher 90, 55200.
http://dx.doi.org/10.1016/S0169-8095(02)00097-2http://dx.doi.org/10.1016/S0169-8095(02)00097-2http://dx.doi.org/10.1016/j.atmosres.2007.11.012http://dx.doi.org/10.1016/j.atmosres.2007.11.012http://dx.doi.org/10.1016/j.atmosres.2007.11.012http://dx.doi.org/10.1029/2005JD005934http://dx.doi.org/10.5194/adgeo-22-3-2009http://dx.doi.org/10.5194/adgeo-22-3-2009http://dx.doi.org/10.5194/adgeo-22-3-2009http://dx.doi.org/10.1029/2008JG000709http://dx.doi.org/10.1029/2008JG000709http://dx.doi.org/10.5194/adgeo-22-3-2009http://dx.doi.org/10.5194/adgeo-22-3-2009http://dx.doi.org/10.1029/2005JD005934http://dx.doi.org/10.1016/j.atmosres.2007.11.012http://dx.doi.org/10.1016/j.atmosres.2007.11.012http://dx.doi.org/10.1016/S0169-8095(02)00097-2http://dx.doi.org/10.1016/S0169-8095(02)00097-2


18/19

186 Vol. 65 No. 2

krauS, r. (1994): Die Bedeutung der stennebel in Chilefr die akteenpopulation. In: akteen und andere Su-kkulenten 45 (11), 242247.

lailhacar, S. (1986): Las grandes formaciones vegetalesde la zona desrtica y mediterrnea prerida y rida de

Chile: con nfasis en sus aptitudes forrajeras. In: BoletinSociedad Chilena de la Ciencia del Suelo 5, 145231.

larraiN, h.; cereceda, P.; PiNto, r.; lazaro, P.; oSSeS, P.and ScheMeNauer, r. S. (2001): Archaeological obser-vations at a coastal fog-site in Alto Patache, South ofIquique, Northern Chile. In: Proceedings of the Con-In: Proceedings of the Con-ference on Fog and Fog Collection, St. Johns, Canada,289292.

larraiN, h.; velSquez, P.; cereceda, r.; eSPeJo, r.;PiNto, r.; oSSeS, P. and ScheMNauer, r. S. (2002):Fog measurements at the site Falda erde north ofChaaral compared with other fog stations of Chile. DS.In: Atmospheric Research 64, 273284. DOI: 10.1016/S0169-8095(02)00098-4

le queSNe, c.; acua, c.; boNiNSeGNa, J. a.; rivera, a.and barichivich, J. (2008): Long-term glacier variationsin the Central Andes of Argentina and Chile, inferredfrom historical records and tree-ring reconstructedprecipitation. In: Palaeogeography, Palaeoclimatol-ogy, Palaeoecology 281, 334344. DOI: 10.1016/j.pal-aeo.2008.01.039

luebert, F.; Garcia, N. and Schulz, N. (2007): Observa-ciones sobre la ora y vegetacin de los alrededores deTocopilla (22S, Chile). In: Boletin Museo Nacional His-toria Natural 56, 2752.

MaNN, G.; zaPFe, h.; MartiNez, r . and Melcher, G.(1953): Colonias de guanacos (Lama guanicoe) en el desi-erto septentrional de Chile. In: Investigaciones Zoologi-cas de Chile 10, 1113.

MarticoreNa, c.; Matthei, o.; rodrGuez, r.; arroyo,M. k.; Muoz, M.; Squeo, F. and araNcio, G. (1998):Catlogo de la ora vascular de la Segunda Regin(Regin de Antofagasta), Chile. In: Gayana Botnica(Chile) 55, 2383.

MarticoreNa, M.; Squeo, F.; araNcio, G. and Muoz, M.(2001): Catlogo de la ora vascular de la I Regin deCoquimbo. In: Squeo, F.; araNcio, G. and Gutierrez,J. r. (eds.): Libro rojo de la ora nativa y de los sitios

prioritarios para su conservacin: Regin de Coquimbo.La Serena.

Muoz-Schick, M.; PiNto, r.; MeSa, a. and Moreira-Muoz, a. (2001): Oasis de neblina en los cerros cos-teros del sur de Iquique, regin de Tarapac, Chile, du-rante el evento El Nio 19971998. In: Revista Chilenade Historia Natural 74, 323339.

Nuez, l. and varela, J. (1967): Sobre los recursos de aguay el poblamento prehispnico de la costa del NorteGrande de Chile. In: Estudios Arqueologicos 3-4, 747.

PhiliPPi, r. a. (1860): Reise durch die Wste Atacama aufBefehl der chilenischen Regierung im Sommer 185354.Halle.

PiNto, r. (2007): Estado de conservacin de Eulychnia iq-uiquensis(Schumann) Britton et Rose (Cactaceae) en el

extremo norte de Chile. In: Gayana Botanica 64 (1),98109.

PiNto, r. and kirberG, a. (2005): Conservation status ofErosyce(Cactaceae) in northernmost Chile. In: Bradleya23, 716.

(2009): Cactus del extremo norte de Chile. Santiago.PiNto, r. and luebert, F. (2009): Datos sobre la ora vascu-

lar del desierto costero de Arica y Tarapaca, Chile, y susrelaciones togeogracas con el sur de Per. In: GayanaBotanica 66 (1), 2849.

PiNto, r.; barra, i. and Marquet, P. a. (2006): Geographi-cal distribution ofTillandsialomas in the Atacama Desert,northern Chile. In: Journal of Arid Environments 65,543552. DOI: 10.1016/j.jaridenv.2005.08.015

PiNto, r.; larraiN, h.; cereceda, P.; lazaro, P.; oSSeS, P. andScheMeNauer, r. S. (2001): Monitoring fog-vegetationcommunities at a fog-site in Alto Patache, south of Iq-uique, Northern Chile, during El Nio and La Niaevents (19972000). In: Proceedings of the Conferenceon Fog and Fog Collection. St Johns, Canada, 297300.

quiNtaNa, J. and berrioS, P. (2007): Study of the coastallow cloud in the northern coast of Chile: variability andtendency. 4th Int. Con. on Fog, Fog Collection and Dew.La Serena, 189192.

reiche, k. (1907): Grundzge der Panzenverbreitung in

Chile. In: eNGler, a. and drude, o. (eds.): Die egeta-tion der Erde. Leipzig.ricardi, M. (1957): Fitogeograa de la costa del departami-

ento de Taltal. In: Boletin Sociedad Biologica de Con-cepcion 32, 39.

richter, M. (1981): limagegenstze in Sdperu und ihre Auswirkungen auf die egetation. In: Erdkunde 35,1230. DOI: 10.3112/erdkunde.1981.01.02

(1995): limakologische Merkmale der stenkordillerein der Region Antofagasta (Nordchile). In: Geokody-namik 16 (3), 283332.

richter, M.; SchMidt, d. and Wilke, H. G. (1993): DasUnwetter von Antofagasta. In: Praxis Geographie 23,

4447ritter, F. (1980): akteen in Sdamerika. Band 3. Chile.ruelle, S. (1995): Etude dimpact de la pollution mtallique

sur le site de Paposo (II Rgion, Chili). Travail de ndtudes. Facult Universitaire des Sciences Agrono-miques de Cembloux.

ruNdel, P. W. and dilloN, M. o. (1998): Ecological pat-terns in the Bromeliaceae of the lomas formation ofCoastal Chile and Peru. In: Plant Systematics and Evolu-tion 212, 261278. DOI: 10.1007/BF01089742
http://dx.doi.org/10.1016/S0169-8095(02)00098-4http://dx.doi.org/10.1016/S0169-8095(02)00098-4http://dx.doi.org/10.1016/j.palaeo.2008.01.039http://dx.doi.org/10.1016/j.palaeo.2008.01.039http://dx.doi.org/10.1016/j.jaridenv.2005.08.015http://dx.doi.org/10.3112/erdkunde.1981.01.02http://dx.doi.org/10.1007/BF01089742http://dx.doi.org/10.1007/BF01089742http://dx.doi.org/10.3112/erdkunde.1981.01.02http://dx.doi.org/10.1016/j.jaridenv.2005.08.015http://dx.doi.org/10.1016/j.palaeo.2008.01.039http://dx.doi.org/10.1016/j.palaeo.2008.01.039http://dx.doi.org/10.1016/S0169-8095(02)00098-4http://dx.doi.org/10.1016/S0169-8095(02)00098-4


19/19


ruNdel, P. W. and Mahu, M. (1976): Community structureand diversity in coastal fog desert in northern Chile. In:Flora 165, 493505.

ruNdel, P. W.; dilloN, M. o.; PalMa, b.; MooNey, h. a.;GulMoN, S. l. and ehleriNGer, J. r. (1991): The phy-

togeography and ecology of the Coastal Atacama andPeruvian Deserts. In: Aliso 13 (1), 149.

ruNdel, P. W.; PalMa, b.; dilloN, M. o.; ShariFi, M. r.and booNPraGob, k. (1997): Tillandsia landbeckii in thecoastal Atacama Desert of northern Chile. In: RevistaChilena de Historia Natural 70, 341349.

ruNdel, P. W.; villaGra, P. e.; dilloN, M. o.; roiG-JueNt, S. and debaNdi, G. (2007): Arid and semi-aridecosystems. In:vebleN, t. t.; youNG, k. r. and orMe,a. r(eds): The physical geography of South America.Oxford, 158183.

Schulz, N. (2009): Loma-Formationen der sten-At-acama/Nordchile unter besonderer Bercksichti-gung rezenter egetations- und limavernderun-gen. http://www.opus.ub.uni-erlangen.de/opus/voll-texte/2009/1289/

Authors

Dr. Natalie SchulzPonticia Universidad Catlica del Per

LimaPeru

[email protected]

Prof. Dr. Patricio AceitunoDepartment of Geophysics

FCFMUniversity of Chile

SantiagoChile

[email protected]

Prof. Dr. Michael RichterDepartment of Geography

University of Erlangen-Nrnbergochstr. 4/4

91054 [email protected]

Schulz, N.; boiSier J. P. and aceituNo, P. (2011): Climatechange along the arid coast of northern Chile. Interna-tional Journal of Climatology (accepted).

SielFeld, W.; MiraNda, e. and torreS, J. (1995): Informacinpreliminar sobre los oasis de niebla de la costa de la Primera

Regin de Tarapac. Programa de Recursos Hidricos y Na-turales Renovables. Universidad de Tarapac, Iquique, Chile.

Squeo, F.; arroyo, M.; MarticoreNa, a; araNcio, G; Muoz-Schick, M.; NeGritto, M.; roJaS, G.; roSaS, M.; rodri-Guez, r.; huMaa, a.; barrera, e. and MarticoreNa, c.(2008): Catlogo de la ora vascular de la Regin de Ataca-ma. In: Squeo, F.; araNcio, G. and Gutierrez, J. r. (eds.):Libro rojo de la ora nativa y de los sitios prioritarios parasu conservacin: Regin de Atacama. La Serena.

valero-GarceS, b. l.; delGado-huertaS, a.; NavaS, a.; ed-WardS, l.; SchWalb, a. and ratto, N. (2003): Patterns ofregional hydrological variability in central-southern Alti-plano (18-26S) lakes during the last 500 years. In: Paleo-geography, Paleoclimatology, Paleoecology 194, 319338.

WerderMaNN, e. (1931): Die Panzenwelt Nord- und Mit-telchiles. In: egetationsbilder 21 (6/7), 3142.
http://www.opus.ub.uni-erlangen.de/opus/volltexte/2009/1289/http://www.opus.ub.uni-erlangen.de/opus/volltexte/2009/1289/http://www.opus.ub.uni-erlangen.de/opus/volltexte/2009/1289/http://www.opus.ub.uni-erlangen.de/opus/volltexte/2009/1289/http://www.opus.ub.uni-erlangen.de/opus/volltexte/2009/1289/

Documents

Schulz Et Al. 2011 Phytogeographic Divisions, Climate Change and Plant Dieback Coastal Desert Northern Chile