Vegetation mapping and nature conservation: a case study in Terceira Island (Azores)

Page 1

Vegetation mapping and nature conservation: acase study in Terceira Island (Azores)

EDUARDO DIAS, RUI B. ELIAS* and VASCO NUNESDepartamento de Ciencias Agrarias, Universidade dos Acores, Terra Cha, 9701-851 Angra do

Heroısmo, Azores, Portugal; *Author for correspondence (e-mail: [email protected]; fax: þ351-

295402205)

Received 14 October 2002; accepted in revised form 2 June 2003

Key words: Azores Islands, Endemic vegetation, NATURA 2000 network, Nature conservation, Ve-

getation classification, Vegetation mapping

Abstract. In order to emphasize the importance of vegetation mapping for nature conservation purposes

a case study in Terceira island (Azores) is presented, in which the importance of the natural vegetation of

the eastern slope of Santa Barbara volcano (which is part of the Site of Community Importance of Santa

Barbara–Pico Alto) is evaluated through the elaboration of its vegetation map. Fourteen (14) different

natural vegetation types were identified: grasslands (1 type), peat bogs (2 types), scrubs (2), forests (5),

successional vegetation (3) and vegetation of rocky slopes (1). All communities are protected under the

Habitat and Species Directive (EC/92/43) and most of them are endemic to the Azores Islands. This fact,

together with the significant number of Azorean endemic taxa (18), Macaronesian endemic taxa (5) and

species protected under the Habitat and Species Directive (7), gives this area an important conservation

value that justifies future protection actions. Vegetation mapping is an important tool for the char-

acterization, evaluation and implementation of managing plans of natural areas of the Azores islands.

The use of a floristic-based classification, supported by multivariate analysis and structural data, is an

efficient methodology for the construction of these maps. The data collected comprise an important set of

information about the distribution and abundance of natural vegetation types and endemic and rare

species. This information was not available until now and is indispensable for the elaboration of man-

agement plans of Special Zones for Conservation that will be part of the NATURA 2000 network.

Nomenclature: Hansen and Sunding (1993), Schafer (2002).

Introduction

Less than a century ago vegetation maps were largely unknown; however, this has

changed over the years and today much energy is being spent on vegetation

mapping. Vegetation is so closely tied to its environment that an appreciation of its

character can reveal the qualities of the sites on which it occurs. The information

contained in these maps is applied to many fields, such as landscape planning,

agriculture, climatology, nature conservation, forestry, geography, plant and animal

ecology (Bredenkamp et al. 1998; Cross 1998; Dias 1988; Kuchler and Zonneveld

1988; Zerbe 1998).

In the past 5 years we have been studying the natural vegetation of the Azores

islands, especially the vegetation of Sites of Community Importance of the NAT-

URA 2000 network (Habitat and Species Directive EC/92/43). These studies aim to

# 2004 Kluwer Academic Publishers. Printed in the Netherlands.

Biodiversity and Conservation 1519–1539, 2004.13:

Page 2

identify the distribution of natural plant species and vegetation for nature con-

servation purposes. In this paper a case study of vegetation mapping for con-

servation purposes is presented, namely the natural vegetation map of the eastern

slope of Santa Barbara volcano, which is part of the Site of Community Importance

of Santa Barbara–Pico Alto.

Methods

Study area

The study area is located on the western part of Terceira island, namely on the

eastern slope of Santa Barbara volcano (Figure 1). Occupying 588 ha, between 520

and 900 m altitude, this area has important endemic plant species and vegetation.

The precipitation ranges between 4109 (at 600 m altitude) and 13.054 l/m2 (at

980 m altitude). The average annual temperature ranges between 15 and 11.3 8C (at

600 and 980 m, respectively). Wind velocities are much higher at 980 m, and fre-

quently the average daily velocities are above 40 km/h (Dias 1996). These values

reflect changes in the climatic characteristics with altitude: in the higher altitude

areas the conditions are more stressful due to wind exposure, very low temperatures

in winter, and soil water saturation. In the lower altitude areas the conditions are not

so extreme, but some periods of hidric stress could occur (Dias 1996).

Most of the soils in the study area are andosols with placic, which are modern soils

that developed from volcanic pyroclastic material, under a wet temperate atlantic

climate (Madruga 1986). According to Pinheiro (1990) these soils have an A00–A0–

AC profile, with high quantities of organic matter and a characteristic Bsm horizon

(called Placic), defined by an accumulation of iron. The presence of placic has an

important ecological meaning, since it limits water drainage (Dias 1996). Besides

andosols there are areas of incipient lithic soils developing on recent trachyte rocks.

According to Mendes (1989) these soils have poorly differentiated horizons with a

(A)–R or (A)–CR profile. The substrate of the study area is mainly composed of

trachyte material, both pyroclastic (pumice fall deposits) and lava flows (domes and

coulees). Several trachyte domes (lava domes) dominate the landscape; some of them

erupted in 1761 AD (Zbyszewski et al. 1971; Self 1976; Elias 2001).

Data collection

The methodology used for floristic-based vegetation descriptions (Mueller-Dom-

bois and Ellenberg 1974; Goldsmith et al. 1986), which has been used (by our

team) in the scope of the LIFE II Project (European Union) (Mendes 1998; Va-

gueiro 1999; Elias 2001), was followed. In the first phase the boundaries of the

main vegetation types were established, on a physiognomic basis, with the help of

aerial photography (flight realized in 1985 at 2286 m altitude; scale 1:17.000). This

was followed by an exploratory phase where information was gathered (on the

1520

Page 3

field) with the objective of confirming (also on a physiognomic basis) the bound-

aries established earlier and registering other types of vegetation not revealed by

aerial photography analysis.

During the third phase several plots were located, following the methodology of

Braun-Blanquet (Mueller-Dombois and Ellenberg 1974), on the several types of

vegetation previously recognized. The areas of the plots varied between 25 and

Figure 1. Location of the study area: (A) Azores Islands with the location of Terceira; (B) Terceira

Island with the location of the study site.

1521

Page 4

100 m2, according to Dias (1996). Afterwards the data were normalized for 25 m2

by simple arithmetic calculation. The plots were located at random inside the

boundaries of each vegetation type identified earlier. The number of plots im-

plemented in each vegetation type varied according to the respective area and

accessibility. No plots were located in the Juniperus Forested Peat Bog because it

lies in a mountainous area in rough terrain which is very difficult to access. To

characterize this vegetation the data of Dias (1996), from plots located in a con-

tiguous area that has the same physiognomy, was used. For the characterization of

the Vegetation of Rocky Slopes a qualitative approach was used, since this vege-

tation occurs on vertical slopes of crevasses where it is not viable to implement

plots. In all the plots the absolute percent coverage (0–100%, estimated visually) of

all vascular plant species and the bryophytes Sphagnum spp. and Polytrichum

commune was measured. Several other features related to the site, like soil type,

geology and geomorphology, presence of seedlings and saplings of the vascular

species, soil and vegetation profile, were also noted.

Since the trachyte domes (lava domes) vegetation presented a clear zonation, a

method adapted from the transect method described by Mueller-Dombois and El-

lenberg (1974) was used for its study. This vegetation has been the object of a more

detailed study by Elias (2001) and Elias and Dias (submitted) on the primary

succession on lava domes. In that work data of vegetation bioarea along transects

were collected. This different approach was used because of the zonation presented

by the vegetation of these geologic structures, which is unique in the study area and

could not be correctly accessed by the use of plots. On the other end, all the domes

present in the study area are in a successional stage which justified this different

approach. Since the type of information was different (plots-percent cover and

transects-bioarea), the data from the domes vegetation were not included in the

multivariate analysis. However, the type of vegetation is clearly different and its

structure is characterized according to Elias (2001) (Appendix 1).

Data analysis

Based on the collected data a plots-versus-species matrix (using the cover values of

each species) was created. A multivariate analysis was applied to this matrix in order to

separate different types of vegetation, using the program Syn-Tax 5.0 (Podani 1994). A

hierarchical agglomerative clustering was used, with the Euclidean Distance as the

similarity coefficient and the Complete Linkage as the method for group formation

(Podani 1994; Fernandez-Palacios and Santos 1996). The data were transformed

logarithmically (base 10) in order to accentuate the lognormal distribution properties

of the variables. On the initial matrix, data from six plots (100, 101, 103, 147, 153 and

164) established earlier by Dias (1996) in the study area were included. Individual

cluster analysis was also applied to the main vegetation types individualized in the first

dendrogram but removing the plots implemented in semi-natural areas.

The evaluation of the results was performed by the program Eval (Syn-Tax 5.0,

Podani 1994) through the explanatory variables, to identify the species that

1522

Page 5

contribute most to the objects’ individualization at a given dissimilarity level.

Principal Component Analysis (PCA) was used to evaluate the ecological ten-

dencies reflected in the dendrogram and to analyze the relationship between species

and plots. A Geographic Information System (GIS) was used to draw a digital map.

Ortophotomaps, on a scale of 1:25 000, were created from the aerial photography

adjusted to digital files of the study site. The digital mapping was performed using

the software Geomedia Professional V3 and printed with CAD quality from the

computer application Intergraph Smartsketch.

Results and discussion

The dendrogram depicted in Figure 2 resulted from the analysis of the species

versus plots matrix. Four major types of vegetation can be distinguished: grass-

lands, peat bogs, scrubs and forests. There is a clear grouping of the formations

dominated by herbaceous plants and by Sphagnum spp. (grasslands and peat bogs)

on the left (A). In the center we have the scrubs divided in two main groups: (B1)

wet formations and (B2) mountain formations, both with a high percentage of

Calluna vulgaris. On the right we have the secondary succession scrubs and forests

(C) (with the exception of plots 36 and 11) and the mature forest formations (D)

dominated by Juniperus brevifolia, Laurus azorica and Ilex perado ssp. azorica.

The ordination (PCA) (Figure 3) confirms the four major types of vegetation

(grasslands, peat bogs, scrubs and forests), as these groups appear in specific areas

of the scatter diagram. It is most likely that the positive sector of axis 1 is related to

more stressful conditions like wind exposure, past human interference or lack of

nutrients. In fact, 11 of the 12 plots implemented in the semi-natural areas (sec-

ondary succession formations) are in this sector. The negative sector of this axis

should correspond to more stable conditions (forests). Axis 2 probably expresses

higher (positive sector) or lower (negative sector) soil water accumulation. Thus, it

is natural that peat bogs appear in the positive sector.

A PCA biplot diagram (Figure 4) showed that the space occupied by forests is

ecologically defined by J. brevifolia, I. perado ssp. azorica, L. azorica, Dryopteris

azorica, Elaphoglossum semicylindricum, and Luzula purpureo-splendens. Scu-

tellaria minor, Agrostis castellana and Holcus rigidus define grasslands. Sphagnum

spp., Eleocharis multicaulis and Juncus bulbosus define peat bogs. The scrub

formations are defined by C. vulgaris, Blechnum spicant and Pteridium aquilinum.

The individual cluster analyses (Figure 5) without the plots located in semi-

natural areas (and including two plots located on the banks of mountain streams)

show nine vegetation types: transition peat bog (A); peat bog (B); Holcus–Agrostis

grassland (C); wet arborescent scrub (D); mountain scrub (E); Juniperus–Ilex forest

(F); gallery forest (G); Juniperus woodland (H, plot 36) and Juniperus–Erica forest

(H, plot 11). We included plots 7 and 8 in the analysis of peat bogs (in spite of the

fact that the initial cluster included them in the wet scrubs group) because they

present a different structure than that of the scrubs, which is supported by the fact

that the PCA grouped them together with peat bogs. This also happened with plot

1523

Page 6

147, which was grouped together with the Juniperus–Ilex forest (in the individual

cluster analysis) but has the structure of a Juniperus woodland. These are good

examples of the difficulty of classifying island vegetation using only floristic

parameters, as was noted earlier by Dias (1989, 1993, 1996) and Fernandez-Pala-

cios (1987) due to the poverty of the flora (the vegetation presented in this paper,

from grasslands to forests, is defined by only 46 species – Appendix 1), which leads

Figure 2. Dendrogram of the plots from the study area, obtained from the 39 plots and 41 species

matrix (species with a low contribution to group formation were removed). A – Peat bogs and grasslands;

B1 – Scrubs (wet formations); B2 – Scrubs (mountain formations); C – Secondary succession scrubs and

forests; D – Mature forests.

1524

Page 7

to several cases of formations that are floristically similar but very different in

structure. A structural approach should always be used at least on a qualitative level

like in the present case, since the use of floristic-structural parameters (biovolume

or bioarea) is a very time-consuming process that cannot always be used when we

want a faster assessment of the vegetation.

Vegetation map (Plate 1)

Of the 588 ha studied, 198 are occupied by pasture and Criptomeria japonica

plantations (cultivated vegetation). About 127 hectares suffered human interference

in the past (semi-natural vegetation). This interference was mainly due to cattle

grazing or wood collecting (Dias 1996). As a result this area is now occupied by

recolonizing scrubs, mixed scrubs (recolonizing vegetation mixed with remnants of

the original vegetation) and C. japonica mixed woods (this species was planted in a

more sparse way, resulting in the existence of patches with native species). The

other 263 ha of the study area are occupied by natural vegetation. It comprises

formations almost undisturbed by man. Table 1 shows the vegetation types iden-

tified (by the clusters, PCA and structural analysis) and the respective dominant

Figure 3. PCA ordination diagram overlaid with the classification obtained in the dendrogram (plots

located in semi-natural areas are underlined). P – Peat bogs; G – Grassland; S – Scrubs; and F – Forests.

1525

Page 8

species in each layer. In Appendix 1 the cover values of the species in the plots and

the percentage of bioarea of each species on the lava domes (successional vege-

tation) are presented.

Natural vegetation types

Juniperus–Ilex Forest

Endemic natural forest, dominated by the dense canopies of J. brevifolia and by the

higher and sparser canopies of I. perado ssp. azorica. Also present are L. azorica,

Culcita macrocarpa, Myrsine africana and Vaccinium cylindraceum. The structure

shows five poorly differentiated layers (emergent canopy, canopy, undifferentiated

shrub/high herbaceous, undifferentiated low herbaceous/bryophyte and epiphytic) and

the maximum height is 5.6 m. It occupies 40 ha of the study area, mainly on trachyte

lava flows.

Figure 4. PCA biplot diagram, showing the correlation between species and plots. Only the most

important species vectors are presented. The areas correspond to the results presented in Figure 3: P –

Peat bogs; G – Grassland; S – Scrubs; and F – Forests.

1526

Page 9

Juniperus woodland

Endemic natural woodland, dominated by the dense canopies of J. brevifolia. Also

dominant are M. africana and D. azorica. The structure shows four layers (canopy,

undifferentiated shrub/high herbaceous, low herbaceous and epiphytic) and the

maximum height is 4.5 m. It occupies 19 ha of the study area, mainly on trachyte

lava flows.

Juniperus forested peat bog

Rare and endemic Atlantic peat bog with sparse canopies of J. brevifolia. From the

floristic point of view this community is very similar to the Juniperus woodland.

However, the lower canopy cover allows light to reach the ground. The greater

Figure 5. Individual dendrograms of each vegetation group identified, without the plots located in

semi-natural areas (and including plots 2 and 5, located on the banks of mountain streams). A –

Transition peat bog; B – Peat bog; C – Holcus–Agrostis grassland; D – Wet arborescent scrub; E –

Mountain scrub; F – Juniperus–Ilex forest; G – Gallery forest; H – Juniperus wood and Juniperus–Erica

forest (plots 36 and 11, respectively).

1527

Page 10

amount of light, together with soil water accumulation, enables a high cover of

Sphagnum spp. (Dias 1996). The structure shows four layers (canopy, high her-

baceous, undifferentiated low herbaceous/bryophyte and epiphytic) and the vege-

tation is dominated by J. brevifolia and Sphagnum spp. It occupies 19 ha of the

study area, mainly on pumice fall deposits.

Juniperus–Erica forest

Another endemic formation, which occupies a small area of 3 ha surrounded by C.

japonica plantations and represents ‘islands’ of original vegetation. The vegetation

is dominated by J. brevifolia and E. azorica. The structure shows four layers

(canopy, undifferentiated shrub/high herbaceous, undifferentiated low herbaceous/

bryophyte and epiphytic) and the maximum height is 5 m.

Gallery forest

This is a transitional community between the terrestrial and aquatic ecosystems. It

develops on the slopes of mountain streams that run across the mountain scrubs of

the southwest part of the study area. Along the banks of these mountain streams is a

herbaceous vegetation type protected under Directive EC/92/43. This vegetation is

very different from the surrounding scrubs, having a dense canopy of J. brevifolia

(about 4 m high) that provides an under-canopy environment similar to that of the

forests. This environment enables a high cover of C. macrocarpa, M. africana and

L. purpureo-splendens.

Plate 1. Vegetation map of the eastern slope of Santa Barbara volcano (Terceira island, Azores).

1528

Page 11

Ta

ble

1.

Dom

inan

tsp

ecie

s(a

ccord

ing

toth

eper

cent

cover

)in

each

layer

of

the

nat

ura

lveg

etat

ion

types

iden

tifi

edin

the

stu

dy

area

.O

ther

imp

ort

ant

spec

ies

inea

ch

lay

erar

eal

sog

iven

(wit

hin

bra

cket

s).

H.

her

bac

eou

s–

Hig

hher

bac

eous;

L.

her

bac

eous

–L

ow

her

bac

eou

s;*

–E

pip

hyti

cla

yer

.

Nat

ura

lveg

etat

ion

typ

eS

pec

ies

Can

op

yla

yer

Sh

rub

/H.

her

bac

eou

sla

yer

L.

her

bac

eou

s/b

ryo

ph

yte

lay

er

Jun

iper

us–

Ilex

fore

stJu

nip

eru

sb

revi

foli

aM

yrsi

ne

afr

ica

na

Ble

chn

um

spic

an

t

Ilex

per

ad

oss

p.

azo

rica

Cu

lcit

am

acr

oca

rpa

Lu

zula

pu

rpu

reo

-sp

len

den

s

La

uru

sa

zori

ca(D

ryo

pte

ris

azo

rica

)(D

ryo

pte

ris

aem

ula

)

(Va

ccin

ium

cyli

nd

race

um

)(L

ysim

ach

iaa

zori

ca)

Sp

hag

nu

msp

p.

Hym

eno

phyl

lum

tun

bri

gen

se*

Jun

iper

us

wo

od

lan

dJu

nip

eru

sb

revi

foli

aM

yrsi

ne

afr

ica

na

Dry

op

teri

sa

emu

la

Dry

op

teri

sa

zori

caB

lech

nu

msp

ica

nt

Hym

eno

phyl

lum

tun

bri

gen

se*

Jun

iper

us

fore

sted

pea

tb

og

Jun

iper

us

bre

vifo

lia

Cu

lcit

am

acr

oca

rpa

Lu

zula

pu

rpu

reo

-sp

len

den

s

(Dry

op

teri

sa

emu

la)

Sp

hag

nu

msp

p.

Hym

eno

phyl

lum

tun

bri

gen

se*

Jun

iper

us–

Eri

cafo

rest

Jun

iper

us

bre

vifo

lia

Myr

sin

ea

fric

ana

Dry

op

teri

sa

emu

la

Eri

caa

zori

caC

ulc

ita

ma

cro

carp

aB

lech

nu

msp

ica

nt

(Ile

xp

era

do

ssp

.a

zori

ca)

(Pte

rid

ium

aq

uil

inum

)(S

ph

ag

num

spp

.)

Hym

eno

phyl

lum

tun

bri

gen

se*

Gal

lery

fore

stJu

nip

eru

sb

revi

foli

aM

yrsi

ne

afr

ica

na

Lu

zula

pu

rpu

reo

-sp

len

den

s

(Ile

xp

era

do

ssp

.a

zori

ca)

Cu

lcit

am

acr

oca

rpa

(Sp

ha

gn

um

spp

.)

Hym

eno

phyl

lum

tun

bri

gen

se*

1529

Page 12

Ta

ble

1.

(con

tin

ued

)

Nat

ura

lveg

etat

ion

typ

eS

pec

ies

Can

op

yla

yer

Sh

rub

/H.

her

bac

eou

sla

yer

L.

her

bac

eou

s/b

ryo

ph

yte

lay

er

Wet

arb

ore

scen

tsc

rub

Call

un

avu

lga

ris

(Ble

chn

um

spic

an

t)

Jun

iper

us

bre

vifo

lia

(Ele

och

ari

sm

ult

ica

uli

s)

(Myr

sin

ea

fric

ana)

Sp

hag

nu

msp

p.

(Va

ccin

ium

cyli

nd

race

um

)

(Eri

caa

zori

ca)

Pte

rid

ium

aq

uil

inu

m

(Cu

lcit

am

acr

oca

rpa)

Mo

un

tain

scru

bC

all

un

avu

lga

ris

Ble

chn

um

spic

an

t

Jun

iper

us

bre

vifo

lia

Ele

och

ari

sm

ult

icau

lis

(Myr

sin

ea

fric

ana)

(Des

cha

mp

sia

foli

osa

)

Pte

rid

ium

aq

uil

inu

m(H

olc

us

rig

idu

s)

(Cu

lcit

am

acr

oca

rpa)

(Fes

tuca

jub

ata

)

(Sp

ha

gn

um

spp

.)

Tra

nsi

tion

pea

tb

og

Call

un

avu

lga

ris

Ele

och

ari

sm

ult

icau

lis

Jun

iper

us

bre

vifo

lia

(Ble

chn

um

spic

an

t)

Jun

cus

effu

sus

(Ag

rost

issp

.)

(Pte

rid

ium

aq

uil

inu

m)

Sp

hag

nu

msp

p.

Pea

tb

og

(Ju

ncu

sef

fusu

s)(E

leo

cha

ris

mu

ltic

au

lis)

(Ju

ncu

sbu

lbo

sus)

Sp

hag

nu

msp

p.

1530

Page 13

Ta

ble

1.

(con

tin

ued

)

Nat

ura

lveg

etat

ion

typ

eS

pec

ies

Can

op

yla

yer

Sh

rub

/H.

her

bac

eou

sla

yer

L.

her

bac

eou

s/b

ryo

ph

yte

lay

er

Ho

lcu

s–A

gro

stis

gra

ssla

nd

Ag

rost

isca

stel

lan

a

Ho

lcu

sri

gid

us

(Lo

tus

uli

gin

osu

s)

(Lys

imach

iaa

zori

ca)

(Scu

tell

ari

am

inor)

Succ

essi

onal

veg

etat

ion

Jun

iper

us

scru

bJu

nip

eru

sb

revi

foli

aB

lech

nu

msp

ica

nt

Ca

llu

na

vulg

ari

s(C

ulc

ita

ma

cro

carp

a)

Jun

iper

us–

call

un

asc

rub

Jun

iper

us

bre

vifo

lia

Ble

chn

um

spic

an

t

Ca

llu

na

vulg

ari

sC

ulc

ita

ma

cro

carp

a

(Vacc

iniu

mcy

lindra

ceum

)(S

ph

ag

num

spp

.)

(Myr

sin

ea

fric

an

a)

Jun

iper

us

fore

stJu

nip

eru

sb

revi

foli

aM

yrsi

ne

afr

ica

na

Lu

zula

pu

rpu

reo

-sp

len

den

s

Ilex

per

ad

oss

p.

azo

rica

Cu

lcit

am

acr

oca

rpa

Sp

hag

nu

msp

p.

(La

uru

sa

zori

ca)

(Va

ccin

ium

cyli

nd

race

um

)

1531

Page 14

Wet arborescent scrub

According to Dias (1996) this endemic scrub is characteristic of wet, low-exposure,

mountainous slopes. In the study area they precede the mountain scrubs, in the

southwest zone, that occur at higher, more exposed, altitudes. Its similarity with

mountain scrubs is evident by the dominance of C. vulgaris and J. brevifolia.

However, the less exposed conditions enable a higher coverage of C. macrocarpa,

V. cylindraceum and L. azorica. It occupies 28 ha of the study area, mainly on

pumice fall deposits.

Mountain scrub

The vegetation is dominated by C. vulgaris and J. brevifolia. It occurs in high

mountainous areas on pumice fall deposits. There are three types of patches: shrub

patches (dominated by C. vulgaris and J. brevifolia); herbaceous patches (domi-

nated by Deschampsia foliosa, Festuca jubata or E. multicaulis) and bryophyte

patches (dominated by Sphagnum spp.). With a maximum height of 1.3 m these

scrubs occupy 54 ha of the study area.

Transition peat bogs

These occupy, together with peat bogs, 14 ha of the study area. They are the

most common Azorean peat bogs. They have a soligenic flux of minerotrophic

water with a higher nutrient availability. Its irregularity is expressed through a

hummock-hollow structure that allows an active mineralization on the hummocks

and the consequent advance of woody species (Dias 1996). Polytrichum com-

mune and Juncus effusus occur on the hummocks as happens with, in a more

stable situation, C. vulgaris and J. brevifolia. The hollows are dominated by

Sphagnum spp.

Peat bog

These are homogeneous, flat, floristically poor, peat bogs. However, the highest

diversity of Sphagnum species occurs here (Mendes 1998). In the vegetation map

(Plate 1) peat bogs and transition peat bogs are represented together under the

designation ‘Peat Bog’.

Holcus–Agrostis grassland

This community, limited to a 0.4 ha area in the south of the study area, is dominated

by H. rigidus and A. castellana.

Successional vegetation

This vegetation type was the object of a more detailed study by Elias (2001)

(Appendix 1). It presents a clear zonation with different communities in the base,

middle and top areas of the domes. This zonation reflects the habitat characteristics

and the colonization and successional histories of these geological formations. In

1532

Page 15

the study area we have several trachyte domes with ages ranging from 240 years to

about 2000 years.

Juniperus scrub

Develops in young domes that represent 20 ha of the study area. The vegetation

cover values are very low mainly in the middle (with only 20% of the surface

covered) and top (50% cover) areas. A Calluna–Juniperus Pioneer Scrub and a

Juniperus Pioneer Scrub, respectively, develop in these areas. These communities

present a clear patchy horizontal structure with ‘spots’ of vegetation in depressions

and cracks in the rock surface and with the more exposed surfaces almost clear of

vegetation.

Juniperus–Calluna scrub

Develops in more ancient domes [370 years in the case study of Elias (2001)] in a

27 ha area. The vegetation has a greater degree of coverage and higher height but

the zonation and patchy horizontal structure, in the middle and top areas, is still

clear. The vegetation at the base of the domes is already well developed and more

similar to the surroundings. The middle and top areas are characterized by a scrub

vegetation that is very similar in composition to the pioneer vegetation, but with

higher cover values.

Juniperus forest

Develops in 2000 year old domes on a 40 ha area. In these domes both the zonation

and horizontal structure, though present, are less evident. The base of the domes is

characterized by a Juniperus–Ilex forest (similar to the surrounding vegetation).

The lateral area presents a Juniperus wet scrub with a high coverage of Sphagnum

spp. The top is occupied by a Juniperus–Ilex–Laurus forest. In spite of these

differences these domes have a physiognomy dominated by J. brevifolia and I.

perado ssp. azorica.

Vegetation of rocky slopes

Due to extreme difficulties of the terrain no plots were implemented here. This

is a community with an important botanical and scientific value due to the

presence of rare species and floristic diversity. Some of the species observed are

Lactuca watsoniana, Euphorbia stygiana, Smilax canariensis or Diphasium

madeirense.

Conclusions

The study area has an important set of plant species and vegetation types. Eighteen

Azorean endemic taxa, 5 Macaronesian endemic taxa, 10 species protected by the

1533

Page 16

Bern Convention and 7 species protected under the Habitat and Species Directive

(Table 2) were identified. All communities are protected under Appendix 1 of the

Habitat and Species Directive (Table 3). This is an important area of natural ve-

getation almost undisturbed by man, mainly because of the rough terrain and poor

soil use capabilities. It is surrounded by an area that suffered extensive use in the

past (semi-natural vegetation) but that is now in a process of recovering its natural

vegetation. With the help of management plans it should be possible to return its

natural status. The number of protected habitats and endemic species gives this area

an important conservation value that justifies future protection actions. Further-

more, most of the species are endemic and living fossils of the Tertiary subtropical

laurel forests of Europe (64–25 million years ago). Most of the taxa are paleoen-

demic and relics of ancient floras that occupied the southern part of Europe and part

of the North American continent (Love and Love 1967; Bramwell 1972, 1976;

Raven and Axelrod 1974; Sergio 1984).

Table 2. List of taxa identified in the study area that are endemic to the Azores (AzE) or Macaronesia

(McE), protected under Appendix 1 of the Bern Convention (Bern) or Appendix 2 of the Habitat and

Species Directive (Ap2) (EC/92/43).

Taxa AzE McE Bern Ap2

Agrostis gracililaxa Franco var. gracililaxa x x

Arceuthobium azoricum Hawksworth et Wiens x x

Culcita macrocarpa C. Presl x x

Deschampsia foliosa Hack. x

Diphasiastrum madeirense (Wilce) Holub x x

Dryopteris azorica (Christ) Alston x

Elaphoglossum semicylindricum (Bowd.) Benl x

Erica azorica Hochst. ex Seub. x x x

Euphorbia stygiana Wats. x x x

Holcus rigidus Hochst. ex Seub. x

Huperzia dentata (Herter) Holub x

Hypericum foliosum Ait. x

Ilex perado Ait. ssp. azorica (Loes.) Tutin x

Juniperus brevifolia (Seub.) Antoine x x

Lactuca watsoniana Trel. x x x

Laurus azorica (Seub.) Franco x

Luzula purpureo-splendens Seub. x

Lysimachia azorica Hornem. ex Hook. x

Platanthera micrantha (Hochst. ex Seub.) Schlecht x

Rubia agostinhoi Dans. & P. Silva x

Smilax canariensis Willd. x x

Tolpis azorica (Nutt.) P. Silva x

Trichomanes speciosum Willd. x x

Woodwardia radicans (L.) J. E. Sm. x x

Vaccinium cylindraceum J. E. Sm. x

Viburnum tinus L. ssp. subcordatum (Trel.) P. Silva x

Total 18 5 10 7

1534

Page 17

Vegetation mapping is an important tool for the characterization, evaluation and

implementation of managing plans of natural areas of the Azores islands. The use

of a floristic-based classification, supported by multivariate analysis and structural

data, is an efficient methodology for the construction of these maps. The data

collected during our studies in this and other Sites of Community Importance

provide us with an important set of information about the distribution and abun-

dance of natural vegetation types and endemic and rare species in the Azores

islands. This information was not available until now and is indispensable for the

elaboration of management plans for Special Zones of Conservation that will be

part of the NATURA 2000 network.

Acknowledgements

This study was supported by the Project LIFE96NAT-P-3022 ‘Study and con-

servation of Azorean natural patrimony’ of the Departamento de Ciencias Agrarias

da Universidade dos Acores/Direcao Regional dos Recursos Florestais.

Table 3. Natural vegetation types of the study area that are protected under Appendix 1 of the Habitat

and Species Directive (EC/92/43).

Natural vegetation types of the

study area

Appendix 1 habitat types NATURA 2000

codes

Mediterranean mountainous

coniferous forests

Juniperus woodland Macaronesian Juniperus woods 9560

Mediterranean sclerophyllous forests

Juniperus–Ilex forest Macaronesian laurel forests 9360

Juniperus–Erica forest

Juniperus forest

Forests of temperate Europe

Juniperus forested peat pog Bog woodland 91D0

Temperate heaths and scrubs

Wet arborescent scrub Endemic macaronesian heaths 4050

Juniperus–Calluna scrub

Mountain scrub Alpine and boreal heaths 4060

Sphagnum acid bogs

Peat bog Raised bogs 7110

Transition peat bog Blanket bogs 7130

Natural grasslands

Holcus–Agrostis grassland Macaronesian mountain grasslands 6180

Chasmophytic vegetation on rocky slopes

Rocky slopes vegetation Chasmophytic vegetation on siliceous

rocky slopes

8220

Chasmophytic vegetation on rocky slopes

Juniperus scrub Pioneer vegetation on rock surfaces 8230

Other rocky habitats

Fields of lava and natural excavations 8320

1535

Page 18

Ap

pen

dix

1.

Cover

val

ues

(A–

J,%

)(l

ow

eran

dh

igh

erli

mit

sar

eg

iven

)o

fth

esp

ecie

sid

enti

fied

inth

eplo

tsim

ple

men

ted

inth

edif

fere

nt

types

of

veg

etat

ion

of

the

stu

dy

area

.A

–Ju

nip

eru

s–Il

exF

ore

st;

B–

Jun

iper

us

wo

od

lan

d;

C–

Jun

iper

us

fore

sted

pea

tb

og

(Dia

s1

99

6);

D–

Jun

iper

us–

Eri

cafo

rest

;E

–G

alle

ryfo

rest

;F

–W

et

arb

ore

scen

tsc

rub

;G

–M

oun

tain

scru

b;

H–

Tra

nsi

tio

np

eat

bo

g;

I–

Pea

tb

og

;J

–H

olc

us–

Agro

stis

gra

ssla

nd

.i–

iii:

Per

cen

tage

(%)

of

bio

area

of

the

spec

ies

iden

tifi

edin

the

succ

essi

onal

veg

etat

ion.

i–

Jun

iper

us

scru

b;

ii–

Jun

iper

us–

Ca

llu

na

scru

b;

iii

–Ju

nip

eru

sfo

rest

(Eli

as2

00

1).

Sp

ecie

sV

eget

atio

nty

pes

AB

CD

EF

GH

IJ

iii

iii

Jun

iper

us

bre

vifo

lia

50

–60

75

–9

08

06

09

02

5–

50

25

–60

18

–50

0–

64

9.0

32

.93

1.3

Ilex

per

ad

oss

p.

azo

rica

20

–60

0.1

–5

25

10–

20

0–

11

–3

0–

12

.01

.71

2.2

La

uru

sa

zori

ca2

5–

70

0.2

–5

50

–6

0–

60

–1

0.1

8.3

Eri

caa

zori

ca0

–9

5–

25

40

3–

12

0–

60–

22

0–

80

.31

.43

.6

Va

ccin

ium

cyli

nd

race

um

20

–40

30

1–

510

–15

4–

15

0–

30–

65

8.3

11

.83

.8

Myr

sin

ea

fric

ana

18

–40

25

–5

02

–1

51

52

0–

50

1–

15

2–

70–

38

.09

.74

.8

Ca

llu

na

vulg

ari

s1

0–

25

0–

60

–1

15

75

–90

25

–70

35

–80

0–

35

19

.63

3.4

0.9

Cu

lcit

am

acr

oca

rpa

30

–80

15

–20

30

30

–80

9–

20

5–

25

0–

55

.61

.81

0.9

Dry

op

teri

sa

zori

ca1

5–

20

2–

40

0.2

–2

02

.60

.2

Pte

rid

ium

aq

uil

inum

5–

21

1–

620

0–

130

–60

14

–25

0–

45

0–

30

.43

.9

Jun

cus

effu

ses

1–

60

–1

32

–1

26

–5

00

–2

0

Wo

od

wa

rdia

rad

ica

ns

3

Dry

op

teri

sa

emu

la3

–1

62

–2

51

–5

20

0–

60

–5

0.3

<0

.1<

0.1

Ble

chn

um

spic

an

t2

1–

54

3–

18

2–

318

15

6–

12

6–

40

1–

25

0–

85

1.7

1.6

0.8

Ele

och

ari

sm

ult

ica

uli

s1

01

2–

27

15–

40

0–

70

Lys

imach

iaa

zori

ca6

–1

00

.5–1

3–

60

–6

0–

135

<0

.10

.22

.4

Lu

zula

pu

rpu

reo

–sp

len

den

s9

–6

01

–3

05

–3

02

00

–3

1–

40

1–

30

.20

.57

.3

Ag

rost

issp

.3

–4

00

–5

Ag

rost

isca

stel

lan

a1

–3

00

–2

04

8

Ho

lcu

sri

gid

us

0–

11

–3

0–

10

8–

15

1–

30–

15

48

0.8

Des

cha

mp

sia

foli

osa

3–

18

0–

1

Fes

tuca

jub

ata

2–

9

Lo

tus

uli

gin

osu

s3

5

Sel

ag

inel

lakr

au

ssia

na

0–

90

–1

01

–1

0<

0.1

Scu

tell

ari

am

inor

0–

10

30

1536

Page 19

Ap

pen

dix

1.

(co

nti

nued

)

Sp

ecie

sV

eget

atio

nty

pes

AB

CD

EF

GH

IJ

iii

iii

Sib

tho

rpia

euro

pa

ea0

–1

0–

0.1

9<

0.1

Ru

bia

ag

ost

inh

oi

0–

30

–3

0–

10

–1

0–

11

.00

.1

To

lpis

azo

rica

0–

10

–1

2–

10

0–

15

13

<0

.1

Hu

per

zia

den

tata

0–

10

–1

0–

3<

0.1

0.3

<0

.1

Jun

cus

bu

lbo

sus

0–

10

–3

00

–3

0

Ca

rex

ech

ina

ta0

–1

00

–1

Ca

rex

per

egri

na

0–

1

Hyd

roco

tyle

vulg

ari

s0

–2

00

–1

0

Pla

tan

ther

am

icra

nth

a0

–1

Po

ten

till

aa

ng

lica

0–

10

–1

1–

10

0–

10

–1

20

Leo

nto

do

nta

raxa

coid

es0

–1

0.1

Hym

enophyl

lum

tunbri

gen

se1

2–

16

10

–30

5–

81

22

0–

30

1–

12

0–

30

–1

0.5

0.6

0.6

Ela

ph

og

loss

um

sem

icyl

indri

cum

5–

90

.2–1

0–

60

–1

0.1

0.1

0.4

Po

lytr

ichu

mco

mm

un

e3

–5

01

15

–25

0–

60

0–

25

20

<0

.12

.2

Sp

ha

gn

um

spp

.6

0–

70

70

–10

01

02

02

0–

90

0–

80

70

–80

80

–90

31

.6<

0.1

4.5

Arc

euth

ob

ium

azo

ricu

m0

–1

0–

25

0–

15

0.4

Sm

ila

xca

na

rien

sis

0.5

Vib

urn

um

tin

um

ssp

.su

bco

rda

tum

0.9

Ag

rost

isg

raci

lila

xavar

.g

raci

lila

xa0

.10

.3

Osm

un

da

reg

ali

s0

.1

Cen

tauri

um

scil

loid

es<

0.1

0.1

Nu

mb

ero

fp

lots

62

21

24

55

31

––

–

1537

Page 20

References

Bramwell D. 1972. Flora of Macaronesia Project. Taxon 21: 730–732.

Bramwell D. 1976. The endemic flora of the Canary Islands. In: Kunkel G. (ed) Biogeography and Ecology

of the Canary Islands. Kluwer Academic Publishers, Dordrecht, The Netherlands, pp. 207–240.

Bredenkamp G., Chytry M., Fischer H.S., Neuhauslova Z. and Van Der Maarel E. 1998. Vegetation

mapping: theory, methods and case studies: Introduction. Journal of Vegetation Science 1: 162–164.

Cross J.R. 1998. An outline map of the potential natural vegetation of Ireland. Journal of Vegetation

Science 1: 241–252.

Dias E. 1988. Carta da vegetacao da ilha Terceira. Actas das 1a Jornadas Atlanticas de Proteccao do Meio

Ambiente: 169–185. Angra do Heroısmo, Azores, 25 de Junho a 1 de Julho. Camara Municipal de

Angra do Heroısmo.

Dias E. 1989. Flora e vegetacao endemica na ilha Terceira, Acores. M.Sc. Dissertation. Departamento de

Ciencias Agrarias, Universidade dos Acores, Angra do Heroısmo, Azores.

Dias E. 1993. Methodological problems on syntaxonomy of islands: vegetation biovolume as a richer

abundance-structural parameter. Proceedings of the 36th International Association of Vegetation

Science Symposium, Tenerife, 12–16 April.

Dias E. 1996. Vegetacao natural dos Acores. Ecologia e sintaxonomia das florestas naturais. Ph.D. Dis-

sertation, Departamento de Ciencias Agrarias, Universidade dos Acores, Angra do Heroısmo, Azores.

Elias R.B. 2001. Sucessao primaria em domas traquıticos. M.Sc. Dissertation in Insular Ecology and

Evolution. Departamento de Ciencias Agrarias, Universidade dos Acores, Angra do Heroısmo, Azores.

Fernandez-Palacios J.M. 1987. Analisis de la variacion espacial de la vegetacion de Tenerife e inter-

pretacion en terminos ambientales. Ph.D. Dissertation, Universidade de La Laguna.

Fernandez-Palacios J.M. and Santos A. 1996. Ecologia de las islas Canarias – Muestreo y analisis de

poblaciones y comunidades. Sociedad La Cosmologica, Santa Cruz de La Palma.

Goldsmith F.B., Harrison C.M. and Morton A.J. 1986. Description and analysis of vegetation. In: Moore

P.D. and Chapman S.B. (eds) Methods In Plant Ecology. Blackwell Science, Oxford, UK, pp. 437–524.

Hansen A. and Sunding P. 1993. Flora of Macaronesia. Checklist of Vascular Plants, Vol. 4, Revised ed.

Sommerfeltia 17, pp. 1–295.

Kuchler A.W. and Zonneveld I.S. (eds) 1988. Vegetation Mapping. Kluwer Academic Publishers,

Dordrecht, The Netherlands.

Legislation Council Directive EC/92/43 of 21 May 1992. On the conservation of natural habitats and of

wild fauna and flora, O.J. L206, 22.07.92.

Love A. and Love D. 1967. The origin of the North Atlantic flora. Aquilo, ser. Botanica 6: 52–66.

Madruga J.S. 1986. Andossolos dos Acores – Contributo para o seu estudo. M.Sc. Dissertation, De-

partamento de Ciencias Agrarias, Universidade dos Acores, Angra do Heroısmo, Azores.

Mendes C.M.F. 1998. Contributo para a caracterizacao de turfeiras de Sphagnum spp. na ilha Terceira.

Relatorio de Estagio. Departamento de Ciencias Agrarias, Universidade dos Acores, Angra do Her-

oısmo, Azores.

Mendes J.P. 1989. Caracterizacao dos solos do Perımetro Florestal da ilha Terceira. Relatorio de Estagio.

Departamento de Ciencias Agrarias, Universidade dos Acores, Angra do Heroısmo, Azores.

Mueller-Dombois D. and Ellenberg H. 1974. Aims and Methods of Vegetation Ecology. Wiley, London.

Pinheiro J.V.F. 1990. Estudo dos principais tipos de solos da ilha Terceira (Acores). M.Sc. Dissertation,

Departamento de Ciencias Agrarias, Universidade dos Acores, Angra do Heroısmo, Azores.

Podani J. 1994. Multivariate Data Analysis in Ecology and Systematics – A Methodological Guide to the

SYN-TAX 5.0 Package. SPB Academic Publishing, The Hague, The Netherlands.

Raven P. and Axelrod D. 1974. Angiosperm biogeography and past continental movements. Annals of

the Missouri Botanical Garden 61: 539–673.

Schafer H. 2002. Flora of the Azores. Margraf Verlag, Weikersheim, Germany.

Self S. 1976. The recent volcanology of Terceira, Azores. Journal of the Geological Society London 132:

645–666.

Sergio C. 1984. The distribution and origin of Macaronesian bryophyte flora. Journal of Hattori

Botanical Laboratory 56: 7–13.

1538

Page 21

Vagueiro P.C.S. 1999. Contributo para a caracterizacao ecologica e ordenamento da Reserva Florestal

Natural do Biscoito da Ferraria. Relatorio de Estagio. Departamento de Ciencias Agrarias, Uni-

versidade dos Acores, Angra do Heroısmo, Azores.

Zerbe S. 1998. Potential natural vegetation: validity and applicability in landscape planning and nature

conservation. Journal of Vegetation Science 1: 165–172.

Zbyszewski G., Medeiros C., Ferreira O. da Veiga and Torre de Assuncao C. 1971. Carta geologica de

Portugal, Notıcia explicativa da folha ilha Terceira. Servicos Geologicos de Portugal, Lisboa, Portugal.

1539