Impacts of altered precipitation, nitrogen deposition and plant competition on a Mediterranean seed bank

Page 1

Nitrogen fertilization and water supply affect germinationand plant establishment of the soil seed bank presentin a semi-arid Mediterranean scrubland

Raul Ochoa-Hueso • Esteban Manrique

Received: 17 July 2009 / Accepted: 4 March 2010 / Published online: 19 March 2010

� Springer Science+Business Media B.V. 2010

Abstract Anthropogenic nitrogen (N) inputs in ter-

restrial ecosystems are higher than those that occur

naturally and have been related to global biodiversity

loss and altered ecosystem functioning. However, its

effects on Mediterranean-type ecosystems, where

production is water-limited and N regulated, remain

unclear. We conducted a green-house experiment

where we evaluated the effects of four simulated

scenarios of N pollution (0, 10, 20 and 50 kg

N ha-1 year-1) and two differential water supply

regimes on the germination (experiment 1) and early

plant establishment (experiment 2) of a seed bank from

a semi-arid Mediterranean ecosystem of central Spain.

Seed bank density was estimated as 62,374 ±

3,279 seeds m-2. Approximately 99.5% of emerged

seeds corresponded to only 14 species of a total of 52,

the majority of which were the annual forb Sagina

apetala. The responses for N treatments were species-

specific, mainly positive or unimodal, with watering

treatments having some interactive effects. N and

water supply also affected total and specific produc-

tivity; the responses found for N treatments were

mainly humpback-shaped and an increased water

supply had additive effects on community establish-

ment in terms of total plant biomass. This response was

linked to forb responsiveness. Contrary to predictions,

grass biomass did not change with N supply; however,

grass to forb ratio was affected because of changes in

the latter. Overall, these experiments suggest a critical

load for plant biomass production and conclude that N

and water availability and supply can modify germi-

nation and plant establishment. This should be taken

into account when analysing the effects of global

change on the dynamics of plant communities where

annuals are dominant or vegetation must establish from

seed following a natural or anthropogenic disturbance

regime.

Keywords Additive effects � Critical load �Mediterranean ecosystems � N fertilization �Rainfall � Seed bank

Introduction

Global change is related to human activities such as

energy use and food production (Clark et al. 2007). It

comprises some ecological drivers such as climate

change (both temperature and rainfall) or N pollution;

both factors are predicted to impact Mediterranean

ecosystems in future scenarios (Sala et al. 2000;

Phoenix et al. 2006). These drivers are also expected

to have interactive and modulating effects in arid and

semi-arid ecosystems (van der Waal et al. 2009) as

net primary production (NPP) is usually co-limited by

them (Austin et al. 2004). A high percentage of plant

species from natural and semi-natural ecosystems are

R. Ochoa-Hueso (&) � E. Manrique

Department of Plant Physiology and Ecology, Centro de

Ciencias Medioambientales, CCMA-CSIC, C/Serrano

115-bis, 28006 Madrid, Spain

e-mail: [email protected]

123

Plant Ecol (2010) 210:263–273

DOI 10.1007/s11258-010-9755-4

Page 2

adapted to soils with low N availability. The accu-

mulation of this nutrient promotes plant susceptibility

to stresses such as pathogens, frost and drought

(Bobbink et al. 1998). Nitrogen fertilization also

produces changes at the ecosystem level such as

reduced biodiversity and substitution of characteristic

animal and plant species (Murray et al. 2006), which

are out-competed by more nitrophilous ones (Zabaleta

et al. 2003; Stevens et al. 2004). Changes in plant

composition will be related to changes in soil seed

banks (Kirkham and Kent 1997).

The soil seed bank is the set of viable seeds present

in the soil and mixed with litter at the surface

(Maranon 2001) and exerts a great effect on the

structure, dynamics and space–time distribution of

Mediterranean plant assemblages (Figueroa et al.

2004). Seed banks may be transient or persistent, two

main regeneration strategies of plants in response to

short-term or unpredictable disturbances, respectively

(Maranon 2001). Persistent seed banks are conse-

quently related to an evolutionary bet-hedging strat-

egy (Venable 2007). The interdependence of plant

assemblages and seed banks (Cox and Allen 2008)

points out the importance of studying seed banks for

a complete understanding of ecosystem response to

environmental change and disturbances.

Plant recruitment in Mediterranean ecosystems is

often more limited by the availability of microsites

for establishment than by the availability of seeds

(Mendez et al. 2008). In this context, spatial

variability of N (Davis 2007), phosphorus (Johnson

2004) and water availability play a key role in

regulating seed germination. For N, fertilization

effects on the herbaceous seed bank seem to be

species-specific (Davis 2007). An increase in soil

nitrate concentration, usually related to disturbances,

can stimulate germination (Maranon 2001; Luna and

Moreno 2009) or increase mortality rate through

enhanced microbial activity (Davis 2007). Post-fire

germination is triggered by nitrous oxide produced in

smoke (Keeley and Fotheringham 1998) and seeds

can also respond to a fire-related increase in light and

soil nitrate (Luna and Moreno 2009). For water,

optimal germination should be related to a cue that

indicates the favourability of the upcoming season, as

is the case for increased rainfall (Petru and Tielborger

2008).

In the context of global change, it is necessary to

know the interactions between the main components

of the soil system, the feedbacks between above-

ground and belowground processes, and the resilience

of the ecosystem to disturbance in order to improve

restoration efforts (Heneghan et al. 2008). In this

sense, understanding the soil seed bank dynamics of

an ecosystem plays a key role in restoration practices.

Studies dealing with N pollution in Mediterranean

ecosystems are scarce and most of the data comes

from experiments conducted in California Mediter-

ranean-type ecosystems (Vourlitis and Pasquini 2009;

Allen et al. 2009). Specifically in the Iberian Penin-

sula, we only know one published study related to the

effects of N pollution on plant community and

ecosystem functioning (Calvo et al. 2002), although

the climate of the study site was not Mediterranean.

Despite the recognised importance of water avail-

ability and temperature in ecosystem response to

anthropogenic N inputs (Xia and Wan 2008), we are

not aware of any study on the potentially interactive

effects of these drivers on the natural vegetation, and

particularly on its related soil seed bank, of central

Spain.

The main aims of this study were to investigate by

two different experiments the effects of N fertiliza-

tion (simulating scenarios of N deposition rate) and

water supply on (i) the germination of the seed bank

from a kermes oak thicket (Quercus coccifera L.)

(experiment 1) and (ii) on the establishment and

productivity of plants (experiment 2). Particularly, we

hypothesised that germination responses would be

species-specific, as has been demonstrated for a large

number of species. In this sense, nitrophilous species

would be favoured whilst species adapted to low-N

environments would present neutral or inhibitory

effects. For the plant establishment experiment, we

hypothesised a linear increase in productivity with N

fertilization, and also a shift of dominance from

annual forbs to annual grasses. Supporting this, it has

been demonstrated for California Mediterranean-type

ecosystems an increased dominance of invasive

grasses and competitive exclusion of native annual

forbs related to atmospheric N pollution (Allen et al.

2009; Fenn et al. 2003). Non-native grass invasion

has also been linked to increased fire risk in

Californian (Allen et al. 2009) and Australian (Rah-

lao et al. 2009) arid ecosystems and to the existence

of critical N thresholds for both germination and

growth. We experimentally evaluated such responses

using a simulated gradient.

264 Plant Ecol (2010) 210:263–273

123

Page 3

Methods

Study site

The soil seed bank used in this study was collected

from a kermes oak thicket located in the Nature

Reserve El Regajal-Mar de Ontıgola (Central Spain,

40�90N, 3�290W). The 570 ha of the Reserve are

mainly occupied by a well preserved kermes oak

thicket and agricultural areas (olive groves and

vineyards). Kermes oak thicket is located at the top

and middle parts of the hill slopes, and separated by

gullies. The altitude is about 500 m a. s. l. The

climate is semi-arid Mediterranean with cold winters

and hot summers. Total rainfall is about 425 mm year-1

and it occurs mostly between October and May, with

a prolonged summer drought period (Rivas-Martınez

1987). The extant vegetation is dominated by ever-

green sclerophyllous species like kermes oak or

rosemary (Rossmarinus officinalis L.) and small

therophytes during the brief growing season. These

therophytes include short-lived species such as

Limonium echioides (L.) Mill. (Plumbaginaceae)

and Asterolinon linum-stellatum (L.) Duby (Primul-

aceae). The former occurs exclusively in the inter-

spaces between rosemaries whilst the latter occurs

mainly, but not only, under the cover of shrubs. They

are wide-spread all over the study area (personal

observation), which makes them good species to

focus on; they are also responsive to N addition in the

field (unpublished data).

Seed bank sampling, processing and analysis

The seed bank experiments were conducted at the

greenhouses of the Centre for Environmental Sci-

ences, Institute for Natural Resources (Spanish

Research Council, Madrid, Spain).

Six open areas approximately 7–10 m wide and

with apparently homogeneous vegetation were cho-

sen in the interspaces between very dense kermes oak

scrubs. These interspaces were characterised by the

presence of rosemary as the main shrub species, a

well developed soil crust composed of cyanobacteria,

mosses and lichens, and the absence of an herbaceous

layer for most of the year. Soil samples were

collected on 31 August 2007, following the spring/

summer seed rain and prior to the onset of equinoctial

rains. Consequently, we collected both transient and

permanent seed banks without distinction. First,

twenty 10 9 10 cm soil sub-samples to a depth of

3 cm were collected in bare ground using a small

garden shovel to a final volume of 6 l per open area.

Once in the lab, samples were stored in a dark room

at 4�C (cold stratification) until processed. After

3 months, soil samples were air-dried for 2 days,

sieved (4-mm mesh), bulked and thoroughly mixed.

Sieving and mixing also helped to scarify the seeds

favouring germination. Aliquots of 400 ml of this soil

were spread in 64 plastic trays (26 9 26 9 11 cm)

on the top of a 5-cm deep layer composed of a

2:1 mixture of vermiculite and river sand (provided

by Eduardo Torroja Institute, Spanish Research

Council).

For soil nutrient analysis, soil nitrate (NO3-) was

extracted by shaking 10 g of sieved soil in 50 ml of

deionised water for 30 min and assessed colorimet-

rically; total N was evaluated by using Kjeldahl acid

digestion with SeSO4–K2SO4 as catalyst in a Tecator

20 digestion system. Analyses were done in a

Kjeltec-auto 1030 analyser (Tecator, Sweden). Soil

aqueous pH was assessed by stirring 10 g of sieved

soil in 25 ml of deionised water and measuring pH

after particle settlement and ionic equilibrium was

reached. Soil NO3- was relatively high in our soils,

23.67 mg kg-1; soil sampling after summer dry N

deposition conditioned the inorganic N availability,

which typically goes down as the rainy season goes

through because of losses via leaching, plant uptake

and denitrification (Allen et al. 2009). Total N values

were more moderate, 0.20%, which is typical for this

type of semi-arid ecosystems. Soil pH was 8.4 and

this conditioned the relative importance of inorganic

N source. Although we supplied both oxidised and

reduced forms, the former is assumed to be the

dominant form of inorganic N in the studied soils and

the chemical state that dominates all inorganic N

cycling processes (Chapin et al. 2002; Ochoa-Hueso

and Manrique, unpublished data).

Germination experiment

Thirty-two replicates were distributed amongst four

N treatments by weekly applying 0, 1.3, 2.6 or

6.5 mg of NH4NO3 in water solution. Fertilization

treatments corresponded to simulated N deposition

rates of 0, 10, 20 and 50 kg N ha-1 year-1. These

rates were selected because all but the highest fall

Plant Ecol (2010) 210:263–273 265

123

Page 4

within the predicted scenarios of N deposition rates

for the Mediterranean Basin in 2050 (Phoenix et al.

2006). These rates are also the same we are using in a

parallel field fertilization experiment conducted at the

same site. Half of the replicates (four per N

treatment) were irrigated with 350 ml of deionised

water twice per week (high-water supply) and the

other half only once per week (low-water supply).

These differential water supplies were chosen to

simulate a constant supply of 270–540 mm of rainfall

year-1, respectively. Overall, we had four replicates

per treatment. We used a watering can to simulate the

rainfall events. As seeds germinated, emerging seed-

lings were identified, registered, and then removed to

avoid competition. If direct identification was not

possible, seedlings were transplanted to individual

pots for further identification. We also placed control

trays all over the greenhouse to detect external seed

contamination. The experiment was terminated after

6 months, when no new seedlings emerged over a

period of 15 days.

Microcosm experiment

Similarly to the germination experiment, another set

of 32 trays were distributed amongst the same four N

treatments. Four trays of each N treatment were

irrigated with 350 ml of deionised once per week

(high-water supply) and another parallel set of trays

were irrigated every 2 weeks (low-water supply).

These differential water supplies were chosen to

simulate a constant rate of 135–270 mm of rainfall

year-1. Watering and fertilization treatments were

done exactly in the same way as the germination

experiment. In this experiment, plants were left to

establish freely until the beginning of withering,

when the vegetation was harvested (4 months later),

oven-dried at 65�C to constant weight, and then

weighted. For each tray, we measured soil pH, and

total above-ground net primary production (A-NPP);

to study the relative abundance of grasses and forbs

we separated aerial biomass between these two

functional groups. We also selected two species to

focus on, L. echioides and A. linum-stellatum,

recording their number of individuals and their

biomass. For Asterolinon in the high-water supply

treatment, we additionally measured the stem length

and fruit productivity, the latter as a measure of

fitness for this species. We wrongly thinned two

replicates from the no N addition and low-water

treatment at the beginning of the study and these trays

were then automatically transferred to the germina-

tion experiment; this resulted in 34 and 30 final

replicates per experiment.

Statistical analysis

For both germination and microcosm experiments we

used two-way ANOVAs to evaluate the effects and

interaction of N and water supply treatments. Tukey’s

post-hoc test was used for multiple comparisons. In

the germination experiment, individual species with a

density above 0.5% of total germinated seeds,

functional groups (woody plants, forbs and grasses)

and diversity were analysed. Diversity of germinated

seeds per tray was assessed using the Shannon–

Wiener equation (H0 = piRlnpi). Only when the

interaction term was significant, multiple compari-

sons were made for N treatments within the watering

treatments. One-way ANOVAs were used for the

effects of N supply on Asterolinon shoot length and

fruit yield in the microcosm experiment. Statistical

significance (P) was established at 0.05 level and

analyses were performed using SPSS 17.0.

Results

Germination experiment

During the course of this experiment a total of 27,240

seedlings emerged. Fifty-two species were recorded

and the response of those that emerged with abundance

higher than 0.5% of total germinated seeds (14 spp.)

was evaluated (Table 1). Sagina apetala Ard. (Caryo-

phyllaceae) was the most abundant species and ranged

from 41.05 to 52.03% of the total seeds emerged in the

different treatments. The lowest seed germination

occurred in low-water supply and with no N addition

(48,745 ± 5,962 seeds m-2) and the maximum was

associated with the high-water supply and 20 kg

N ha-1 year-1 (62,374 ± 3,279 seeds m-2).

Total grasses (df = 3,26; F = 3.17; P = 0.04), S.

apetala (df = 3,26; F = 5.73; P \ 0.01), and Sedum

gypsicola Boiss. & Reut. (Crassulaceae; df = 3,26;

F = 3.00; P = 0.05) were affected by N treatments.

Significant interactions were found between water

supply and N fertilization treatments for Arenaria

266 Plant Ecol (2010) 210:263–273

123

Page 5

Ta

ble

1N

um

ber

of

ger

min

ated

seed

str

ay-

1fo

r1

4sp

ecie

s(a

llan

nu

alfo

rbs

exce

pt

the

per

enn

ial

forb

S.

gyp

sico

la)

pre

sen

tin

the

soil

seed

ban

ko

fth

ek

erm

eso

akth

ick

etfr

om

Ara

nju

ez(c

entr

alS

pai

n)

Wat

erin

gtr

eatm

ent

Hig

hw

ater

Low

wat

erN

(sig

.)

Ntr

eatm

ent

(kg

Nha-

1yea

r-1)

010

20

50

010

20

50

Are

nari

ale

pto

clados

Car

yophyll

acea

e47.0

±5.9

a28.8

±5.8

a29.8

±6.0

a30.5

±4.2

a25.5

±2.3

a34.5

±5.6

ab45.8

±2.7

b39.8

±5.5

aba,

a,a,

a

Ast

eroli

non

linum

-ste

llatu

mP

rim

ula

ceae

43.0

±4.9

30.5

±5.6

41.2

±3.0

34.8

±2.9

31.5

±5.6

37.7

±3.1

36.8

±4.9

39.0

±2.1

a,a,

a,a

Cam

panula

erin

us

Cam

pan

ula

ceae

25.0

±3.3

23.3

±6.5

26.3

±4.3

27.3

±3.3

28.3

±1.9

32.8

±2.8

31.8

±4.5

28.8

±2.3

a,a,

a,a

Cra

ssula

till

aea

Cra

ssula

ceae

26.0

±3.7

22.5

±3.1

26.3

±3.8

25.0

±3.7

18.8

±2.2

17.5

±2.5

14.5

±1.9

22.7

±2.2

a,a,

a,a

Ero

phil

ave

rna

Bra

ssic

acea

e12.3

±1.8

a10.5

±0.5

a11.8

±0.6

a9.8

±0.6

a10.8

±2.3

a8.3

±2.0

a6.8

±2.9

a16.3

±1.6

aa,

a,a,

a

Fil

ago

pyr

am

idata

Ast

erac

eae

6.0

±1.1

7.3

±2.7

10.3

±1.6

10.3

±1.8

8.5

±1.8

12.0

±6.4

7.5

±2.2

7.3

±1.1

a,a,

a,a

Her

nia

ria

hir

suta

Car

yophyll

acea

e16.0

±1.4

12.5

±1.9

19.0

±3.4

12.3

±2.5

11.2

±2.0

18.0

±1.9

16.0

±2.7

15.5

±2.5

a,a,

a,a

Horn

ungia

pet

raea

Bra

ssic

acea

e8.3

±1.7

8.8

±1.6

6.5

±1.9

10.0

±2.3

5.0

±1.4

10.8

±0.6

11.0

±2.9

10.8

±1.3

a,a,

a,a

Iber

iscr

enata

Bra

ssic

acea

e5.3

±0.6

5.5

±1.7

5.5

±0.7

1.3

±0.5

4.7

±1.1

4.3

±1.1

5.3

±0.8

5.3

±1.1

a,a,

a,a

Lim

oniu

mec

hio

ides

Plu

mbag

inac

eae

21.0

±2.2

23.3

±4.8

33.3

±2.7

21.5

±4.0

19.5

±3.9

24.3

±0.3

22.0

±3.5

24.3

±3.3

a,a,

a,a

Pis

tori

nia

his

panic

aC

rass

ula

ceae

5.3

±2.2

7.8

±1.9

14.3

±5.2

8.0

±2.5

11.2

±2.1

9.8

±2.1

9.0

±3.2

9.5

±1.5

a,a,

a,a

Sagin

aapet

ala

Car

yophyll

acea

e338.5

±10.1

353.0

±22.2

409.5

±9.0

403.7

5±

13.3

351.0

±13.6

361.5

±9.9

360.0

0±

16.4

421.2

5±

33.8

a,a,

ab,

b

Sed

um

gyp

sico

laC

rass

ula

ceae

10.2

5±

3.3

7.2

5±

2.4

2.2

5±

1.0

5.0

0±

2.6

6.3

3±

2.1

3.5

0±

0.5

2.2

5±

0.5

9.5

0±

2.8

a,a,

a,a

Sher

ard

iaarv

ensi

sR

ubia

ceae

33.0

±5.5

a26.0

±4.2

a32.0

±3.4

a20.3

±2.6

a21.8

±3.1

a28.3

±2.3

a22.8

±3.2

a27.0

±2.7

aa,

a,a,

a

Tota

l826.3

±28.1

a733.5

±74.5

a907.8

±27.5

a773.5

±24.6

a713.7

±38.7

a801.5

±28.8

ab812.5

±21.3

ab884.5

±20.9

ba,

a,a,

a

H0

forb

s2.4

±0.1

2.3

±0.1

2.3

±0.0

2.1

±0.0

2.2

±0.1

2.3

±0.1

2.2

±0.1

2.2

±0.1

a,a,

a,a

Ric

hnes

s23.8

±1.1

23.3

±0.6

25.0

±0.7

21.5

±1.7

20.5

±0.9

24.3

±0.6

22.0

±2.1

22.7

5±

1.5

a,a,

a,a

Woody

2.8

±1.1

3.3

±1.1

3.2

±0.8

3.8

±1.3

1.7

±0.6

1.8

±0.6

3.0

±2.0

1.8

±0.9

a,a,

a,a

Gra

sses

201.3

±16.0

ab150.0

±27.0

ab217.8

±13.8

a141.3

±12b

144.7

±21.5

a178.5

±10.1

a204.0

±12.8

a184.5

±6.4

aa,

a,a,

a

Forb

s622.3

±16.1

a580.3

±96.2

a686.8

±17.3

a628.5

±20.7

a567.3

±20.3

a621.3

±18.3

ab605.5

±13.4

a698.3

±22.6

ba,

a,a,

a

Forb

sdiv

ersi

ty(H0 )

and

ger

min

ated

seed

sso

rted

by

funct

ional

gro

ups

dat

aar

eal

sopre

sente

d.

Mult

iple

com

par

isons

afte

rT

ukey

’spost

-hoc

test

sfo

rth

eN

trea

tmen

tsar

esh

ow

n.M

ult

iple

com

par

isons

for

the

N

trea

tmen

tsw

ithin

wat

erin

gtr

eatm

ents

are

also

show

nonly

wher

eth

ein

tera

ctio

nte

rmis

signifi

cant

(see

‘‘R

esult

s’’

sect

ion).

Dif

fere

nt

low

er-c

ase

lett

ers

indic

ate

signifi

cant

(P\

0.0

5)

dif

fere

nce

sbet

wee

n

gro

ups.

Sta

ndar

der

ror

isgiv

enfo

rea

chtr

eatm

ent.

N=

4(e

xce

pt

for

the

no

Nad

dit

ion

and

low

-wat

ertr

eatm

ent

wher

eN

=6)

Plant Ecol (2010) 210:263–273 267

123

Page 6

leptoclados (Reichenb.) Guss. (Caryophyllaceae;

df = 3,26; F = 6.54; P \ 0.01), Erophila verna

(L.) Chevall. (Cruciferae; df = 3,26; F = 3.34; P =

0.04), Sherardia arvensis L. (Rubiaceae; df = 3,26;

F = 3.07; P = 0.05), total forbs (df = 3,26; F =

4.46; P = 0.01), total grasses (df = 3,26; F = 3.47;

P = 0.03) and total emerged seeds (F = 4.45;

P = 0.01). Watering treatments did not affect seed

bank germination (P[ 0.05). All effects of N and

watering treatments on germination are shown in

Table 1.

Microcosm experiment

Microcosm A-NPP was significantly affected by N

(df = 3,22; F = 6.50; P \ 0.01) and water supply

(df = 1,22; F = 23.14; P \ 0.01; Fig. 1a), but the

interaction was not significant (df = 3,22; F = 1.05;

P = 0.39). Forbs A-NPP was affected by N

(df = 3,22; F = 9.87; P \ 0.01) and water treat-

ments (df = 1,22; F = 17.47; P \ 0.01) (Fig. 1b)

but the interaction term was also not significant

(df = 3,22; F = 0.14; P = 0.94). Grass A-NPP was

affected by water supply (df = 1,22; F = 7.41;

P = 0.01) but not by N treatments (df = 3,22,

FN = 0.73, P = 0.55; df = 3, 22, FNxW = 1.17,

P = 0.34) (Fig. 1c). Grass to forb ratio (Fig. 1d)

was altered by N fertilization (df = 3,22; F = 3.37;

P = 0.04) but water (df = 1,22; F = 0.31;

P = 0.56) and the interaction (df = 3,22; F = 0.42;

P = 0.74) effects were non-significant. Soil pH was

significantly modified by N treatments (df = 3,22;

F = 92.81; P \ 0.01, Fig. 1b) but not by water

supply (df = 1,22; F = 0.43; P = 0.52) (Fig. 1b),

whilst the interaction was marginally significant

(df = 3,22; F = 2.71; P = 0.07).

All effects for N and watering treatments on the

analysed species are shown in Table 2. Limonium total

individuals (df = 3,22, FN = 9.79, P \ 0. 01; df =

1,22, FW = 18.34, P \ 0.01), A-NPP (df = 3,22,

FN = 7.91, P \ 0.01; df = 1,22, FW = 16.88; P \0.01) and biomass per individual (df = 3,22, FN =

6.95, P \ 0.01; df = 1,22, FW = 8.80, P \ 0.01)

were independently affected by N and water supply;

interactive effects were also found for total individuals

(df = 3,22; F = 5.60; P \ 0.01) and biomass per

individual (df = 3,22; F = 5.34, P \ 0.01).

Asterolinon number of individuals was interac-

tively affected by N and watering treatments

(df = 3,22; F = 3.27; P = 0.04), but not individu-

ally (df = 3,22, FN = 1.56, P = 0.23; df = 1,22,

FW = 4.11, P = 0.06). A-NPP was affected by N

(df = 3,22; F = 4.26; P = 0.02) and water

(df = 1,22; F = 67.35; P \ 0.01), and the interac-

tion was also significant (df = 3,22; F = 4.12;

P = 0.02). Shoot length (df = 3,12; F = 4.32;

P = 0.03) and fruit yield (df = 3,12; F = 4.20;

P = 0.03) were affected by N treatments. Biomass

per individual was only significantly affected by

water supply (df = 1,22; F = 39.98; P \ 0.01).

Discussion

The density of seeds (*60,000 seeds m-2) present in

the soil of the kermes oak thicket at the end of the

summer was unusually high in comparison to other

studies including both transient and permanent seed

banks from Iberian (*16,000 seeds m-2; Caballero

et al. 2003), Chilean (*5,000 seeds m-2; Figueroa et al.

2004), Australian (*10,000 seeds m-2; Fisher et al.

2009) and Californian (*13,000 seeds m-2; Cox and

Allen 2008) Mediterranean ecosystems. Consequently,

plant recruitment was not limited in the field by seed

availability, as the establishment limitation paradigm

poses (Mendez et al. 2008). The high density recorded

here could be related to a well developed persistent soil

seed bank consisting of dormant seeds accumulated in

the soil profile, combined with a previous growing

season favourable for seed production and viability

(transient seed bank) (Caballero et al. 2008).

Seed emergence can be influenced both positively

and negatively by soil nitrate or ammonium concen-

tration (Rashid et al. 2007). Since restoration of

disturbed arid or semi-arid environments is often

hindered by low seedling establishment, information

on seed biology and germination is a valuable tool for

restoration practices (Commander et al. 2009). Soil

seed bank germination was significantly and interac-

tively affected by N and water supply, although the

response observed for each species or functional

group (woody, forbs and grasses) varied dramatically.

The main species present in the seed bank (S. apetala)

was found to have enhanced germination after

inorganic N addition, whilst increased germination

with fertilization for total and forb seeds were

apparent only in the low-water treatment (Table 1).

Ephemeral species germinate after the opening of

268 Plant Ecol (2010) 210:263–273

123

Page 7

windows of opportunity and must be able to detect

these suitable conditions (Hunt et al. 2009). Increases

in soil N availability related to disturbances usually

act as a signal for pioneer species to germinate,

establish and complete their life cycle (Luna and

Moreno 2009), and this seems to be the case for the

short-lived and nitrophilous S. apetala. Non-linear

responses of plant species to N fertilization are also

possible as suggested by the responses of A. leptoc-

lados and total grasses. Keeley and Fotheringham

(1998) found such humpback-shaped responses for

one species in the Caryophyllaceae family when

supplying with increased levels of nitrate. In contrast,

germination of I. crenata at high-water level showed

a trend to be reduced by N fertilization, suggesting a

toxic effect of high inorganic N concentrations. Forbs

diversity also seemed to be reduced by N in the high-

water supply treatment; this was most likely associ-

ated with the enhanced germination of S. apetala.

Therefore, increased N deposition in terrestrial eco-

systems may favour the nitrophilous species fraction

present in the seed bank, which would eventually

displace species of formerly oligotrophic environ-

ments (Bobbink et al. 1998).

Seeds present in the soil can also serve as a

resource for microorganisms such as fungi and

bacteria. Low inorganic N availability coupled with

high C:N ratios of the coats prevent rapid deteriora-

tion of the seeds. Soil N enrichment may therefore

favour the degradation of these seeds by micro-

organisms (Leishman et al. 2000; Chee-Sanford et al.

2006). Inorganic N in soils may also reduce seed

mortality by promoting germination and reducing

their exposure to microorganisms (Maranon 2001;

Davis 2007). Other factors that are also known to

affect germination, such as soil moisture and oxygen

concentration, are influenced by water supply and N

availability. These factors may, in turn, influence the

presence, survival and function of micro-organisms

and the responses found in our experiment could be

related to both direct and indirect treatment effects.

Fig. 1 Total above-ground net primary production (A-NPP)

(a), forbs A-NPP (b), grasses A-NPP (c), grass to forb ratio (d),

and soil pH (e) by N and watering treatments. (?W): high-

water supply; (-W): low-water supply. Different lower-case

letters below bars indicate significant (P \ 0.05) differences

after multiple comparisons. Standard error bars are shown.

N = 4 (except for the no N addition and low-water treatment

where N = 2)

b

0

0,5

1

1,5

2

2,5

3

3,5

4

4,5

5

Bio

mas

s (g

)

+W

-W

0,0

0,2

0,4

0,6

0,8

1,0

1,2

1,4

1,6

1,8

2,0

Gra

sses

: fo

rbs

+W

-W

0

0,5

1

1,5

2

2,5

3

3,5

Forb

s bi

omas

s (g) +W

-W

0

0,5

1

1,5

2

2,5

3

Gra

sses

bio

mas

s (g)

+W

-W

7,5

7,6

7,7

7,8

7,9

8,0

8,1

8,2

8,3

8,4

8,5

0 10 20 50

Soil

pH

Kg N ha-1 yr-1

+W

-W

b aaba

b abaa

a aaa

b ababa

b caa

a

b

c

d

e

Plant Ecol (2010) 210:263–273 269

123

Page 8

The overall effect of N addition on the seed bank of a

certain species or functional group would depend on

which is the dominant soil process influencing

germination (signal effect, toxicity, microbial activ-

ity, etc.).

Although large alterations of natural environments

are expected in association with N pollution, other

factors like fire or animal grazing and trampling (Yu

et al. 2008) are also responsible for profound changes

in vegetation composition and structure of the

community. Fire is very important in the vegetation

dynamics of Mediterranean ecosystems, producing

sudden changes in the composition of the seed bank

(Cox and Allen 2008) and affecting its germination

(Reyes and Trabaud 2009). It has been noted the

importance of evaluating the impact of such cata-

strophic events in conjunction with climate change

and atmospheric chemistry (Jentsch et al. 2007).

Seed production is also usually limited by N

availability (Miyagi et al. 2007), generating potential

long-term impacts on seed bank composition beyond

those associated with germination. Changes in the

quantity and quality of seeds produced per individual

are therefore plausible mechanisms that may further

generate changes in community structure (Breen and

Richards 2008; Manning et al. 2009). We see the

potential for this in Asterolinon; this species did not

alter its germination with the fertilization but its fruit

yield was increased in the 20 kg N ha-1year-1

treatment.

The soil seed bank and extant vegetation of

undisturbed ecosystems usually differ (Hopfensper-

ger 2007). This is not typically the case for Mediter-

ranean annual plant communities, where high

similarities are described (Maranon 2001), especially

in the early stages of succession (Luzuriaga et al.

2005). From this, it is logical to think that changes in

the rates of germination of species present in a

Mediterranean seed bank will directly affect the

extant vegetation. If favoured species produce a

greater amount of seeds, significant changes in

vegetation are possible. In our study, the species

with significant responses mostly belong to the

annual forbs group, the largest group within the seed

bank. The low number of shrub seeds in the soil seed

bank also indicates that they are not as persistent as

the annuals (Cox and Allen 2008). The long viability

of annual plant seeds in soils may in turn prevent

sudden changes in the seed bank composition ofTa

ble

2A

ster

oli

no

nli

nu

m-s

tell

atu

man

dL

imo

niu

mec

hio

ides

nu

mb

ero

fin

div

idu

als

tray

-1,

bio

mas

str

ay-

1,

and

bio

mas

sin

div

idu

al-

1

NS

ig.

Ind

ivid

ual

sS

ig.

Bio

mas

s(g

)S

ig.

Bio

mas

s(g

)/in

dS

ig.

Sh

oo

tle

ng

th(c

m)

Sig

.F

ruit

yie

ld(g

)

Hig

hw

ater

Lo

ww

ater

Hig

hw

ater

Lo

ww

ater

Hig

hw

ater

Lo

ww

ater

Hig

hw

ater

Hig

hw

ater

A.

lin

um

-ste

lla

tum

0a

35

.3±

3.2

a1

2.5

±7

.5a

a0

.32

±0

.03

ab0

.02

±0

.01

aa

0.0

09

±0

.00

10

.00

3±

0.0

00

ab4

.0±

0.2

ab0

.11

±0

.01

10

a3

2.5

±2

.9ab

25

.3±

7.0

aab

0.4

0±

0.0

7ab

0.1

2±

0.0

2b

a0

.01

2±

0.0

02

0.0

06

±0

.00

3ab

4.3

±0

.1ab

0.1

2±

0.0

3

20

a3

7.8

±3

.3a

27

.3±

2.6

ab

0.5

6±

0.0

6a

0.0

7±

0.0

1ab

a0

.01

5±

0.0

02

0.0

03

±0

.00

1a

4.9

±0

.1a

0.1

8±

0.0

2

50

a1

7.5

±4

.8b

28

.3±

8.0

aa

0.2

2±

0.0

8b

0.0

8±

0.0

2ab

a0

.01

1±

0.0

02

0.0

03

±0

.00

0b

3.9

±0

.3b

0.0

7±

0.0

3

L.

ech

ioid

es

0a

24

.5±

2.7

a1

1.5

±1

.5a

a0

.32

±0

.06

0.1

2±

0.0

6a

0.0

13

±0

.00

1a

0.0

11

±0

.00

5a

10

a2

7.8

±3

.6a

15

.5±

2.4

aab

0.6

6±

0.0

60

.35

±0

.11

ab0

.02

5±

0.0

02

a0

.01

5±

0.0

05

a

20

a2

9.3

±2

.7a

19

.3±

2.4

ab

0.7

9±

0.0

70

.56

±0

.12

bc

0.0

27

±0

.00

3a

0.0

29

±0

.00

5a

50

b9

.8±

1.7

1b

14

.3±

1.4

aa

0.5

1±

0.1

30

.25

±0

.01

c0

.05

1±

0.0

09

b0

.01

8±

0.0

01

a

Sh

oo

tle

ng

than

dfr

uit

yie

ldd

ata

isal

sop

rese

nte

dfo

rA

ster

oli

no

n.

Dif

fere

nt

low

er-c

ase

lett

ers

ind

icat

esi

gn

ifica

nt

(P\

0.0

5)

dif

fere

nce

sb

etw

een

Ntr

eatm

ents

(sig

.).

Mu

ltip

le

com

par

iso

ns

for

the

Ntr

eatm

ents

wit

hin

wat

erin

gtr

eatm

ents

are

also

sho

wn

on

lyw

her

eth

ein

tera

ctio

nte

rmis

sig

nifi

can

t(s

ee‘‘

Res

ult

s’’

sect

ion

).S

tan

dar

der

ror

isg

iven

for

each

trea

tmen

t.N

=4

(ex

cep

tfo

rth

en

oN

add

itio

nan

dlo

w-w

ater

trea

tmen

tw

her

eN

=2

)

270 Plant Ecol (2010) 210:263–273

123

Page 9

Mediterranean ecosystems. This potential storage

capacity of soils could be related to the low

percentage of germinated seeds observed in the same

field site in spring 2008 (personal observation) when

comparing with the seed bank density reported here.

NPP is commonly N-limited in natural ecosystems

throughout the globe (LeBauer and Treseder 2008). In

arid and semi-arid ecosystems, water stress, drought

and related changes in water potential are also known

to limit biomass production (Austin et al. 2004).

Therefore, and in agreement with our hypothesis, N

addition and increased water enhanced A-NPP in our

microcosm experiment. We found parallel responses

for A-NPP to N fertilization between watering treat-

ments suggesting additive effects for these drivers,

which is in agreement with the existing literature

(Zabaleta et al. 2003; Matesanz et al. 2008). In this

context, future N pollution scenarios could have the

same effect on productivity independent of rainfall

changes. However, pulse water events, key in semi-

arid ecosystems functioning (Austin et al. 2004), could

affect plant establishment and competition in a

different way (Jankju-Borzelabad and Griffiths 2006;

van der Waal et al. 2009).

Plant growth and fitness appeared to be related to a

critical load between our 20 and 50 kg N ha-1 year-1

treatments. This critical load is higher than others

proposed for Mediterranean ecosystems (Roda et al.

2002), but the greenhouse approach used here was far

from realistic. Consequently, further research will be

needed to detect real thresholds for the effects of this

driver. Although humpback-shaped responses are

wide-spread in response to N enrichment (Salemaa

et al. 2008), we acknowledge additional limitations of

our greenhouse approach for disentangling the under-

lying reasons for this response. Based on previous

studies, we can tentatively relate the reduction in

plant growth at high N fertilization to increased

evapo-transpiration rates and faster soil water deple-

tion (van der Waal et al. 2009) or to direct toxic

effects (van den Berg et al. 2005).

Grass to forb ratio significantly changed as a

consequence of forb responses; the experiment ren-

dered the highest grass to forb ratio with lowest N

supply and also a minima when forb mass peaked.

Consequently, grasses should have higher ability to

compete for low N than forbs, but not for water. The

existence of a critical load suggests that atmospheric N

deposition beyond a threshold would also result in the

dominance of annual grasses at expense of annual forbs

in natural and species-rich communities. In contrast to

N pollution effects on California Mediterranean-type

ecosystems, the predominant grasses here were all

native and never completely out-competed the native

forbs group. Phosphorous limitation (extractable P

within the thicket is usually below 1 ppm; unpublished

data) could be an underlying reason for this and has

been proposed as an explanation for the success of

invasive Mediterranean-Basin grasses in California

Mediterranean-type ecosystems (Edith Allen, personal

communication).

Soil acidification is usually linked to eutrophica-

tion in highly polluted areas and a dose related

decrease in soil pH is well documented in natural and

semi-natural ecosystems (Horswill et al. 2008).

Acidification is more problematic in soils with low

cation exchange capacity but this is not the case of

our study site. Low soil pH can reduce germination

(Roem et al. 2002) and plant performance (van den

Berg et al. 2005). Although soil pH was basic in all

treatments, the slight acidification (a decrease of

*0.5) after N fertilization in our microcosm exper-

iment could also be responsible for some of the

responses found. This is because soil pH determines

not only nutrient availability (Chapin 1980) but also

the distribution of microbial soil communities and

their processes (Porter et al. 1987; Reth et al. 2005).

Scaling up the results from this experiment, we can

conclude that modified rainfall patterns and anthro-

pogenic N inputs could alter species germination

ability and subsequently affect seedling survival and

establishment. Ecosystem productivity and net plant–

plant interaction outcome could also be modified,

which would generate a shift in plant communities in

natural and semi-natural ecosystems. This altered

ecosystem structure may negatively impact ecosys-

tem function. For instance, if vascular plant, and

specifically woody plant, productivity is enhanced

after N fertilization (Vourlitis et al. 2009), the

resulting asymmetric competition and shading would

displace lichen soil crusts (Cornelissen et al. 2001), a

key component of arid and semi-arid ecosystems

(Maestre et al. 2005). Global change drivers such as

atmospheric N deposition rate and rainfall amount are

therefore predicted to alter semi-arid Mediterranean

plant communities. This study highlights the impor-

tance of considering both factors in ecosystem

management and ecological restoration. This will be

Plant Ecol (2010) 210:263–273 271

123

Page 10

especially important where annuals are dominant or

where vegetation recovery depends on recruitment

from the seed bank.

Acknowledgements Spanish Ministerio de Educacion y

Ciencia financially supported this research by projects (CGL-

2009-11015 and AGL-2006-13848-C02-01/AGR) and a pre-

doctoral FPU fellowship to ROH (AP2006-04638). Comunidad

de Madrid also contributed to partially fund this work by

REMEDINAL (S-0505/AMB/0335) and S-0505/AMB/00321

projects. We are indebted to Raquel Santamarıa for helping

during the green-house work, Dr. M. Esther Perez-Corona for

critical reading and comments of an earlier version of the

manuscript, Dr. Edith B. Allen and Heather Schneider for

revision of the English and valuable scientific comments, and

Dr. Scott Meiners and two anonymous referees for improving the

manuscript. We are also indebted to Dr. Jose Javier Pueyo and

Dr. Begona Peco for their unselfish involvement in the project.

References

Allen EB, Rao LE, Steers RJ, Bytnerowicz A, Fenn ME (2009)

Impacts of atmospheric nitrogen deposition on vegetation

and soils at Joshua Tree National Park. In: Webb RH,

Fenstermaker JS, Heaton JS, Hughson DL, McDonald EV,

Miller DM (eds) The Mojave Desert: ecosystem processes

and sustainability. University of Nevada Press, Las Vegas,

pp 78–100

Austin AT, Yahdjian L, Stark JM, Belnap J, Porporato A,

Norton U, Ravetta DA, Schaeffer SM (2004) Water pulses

and biogeochemical cycles in arid and semiarid ecosys-

tems. Oecologia 141:221–235

Bobbink R, Hornung M, Roelofs JGM (1998) The effects of

air-borne nitrogen pollutants on species diversity in nat-

ural and semi-natural European vegetation. J Ecol

86:717–738

Breen AN, Richards JH (2008) Irrigation and fertilization

effects on seed number size germination and seedling

growth: implications for desert shrub establishment.

Oecologia 157:13–19

Caballero I, Olano JM, Loidi J, Escudero A (2003) Seed bank

structure along semi-arid gypsum gradient in Central

Spain. J Arid Environ 55:287–299

Caballero I, Olano JM, Escudero A, Loidi J (2008) Seed bank

spatial structure in semi-arid environments: beyond the

patch-bare area dichotomy. Plant Ecol 195:223

Calvo L, Alonso I, Marcos E, De Luis E (2002) Effects of

cutting and nitrogen deposition on biodiversity in Canta-

brian heathlands. Appl Veg Sci 10:43–52

Chapin FS (1980) The mineral nutrition of wild plants. Annu

Rev Ecol Syst 11:233–260

Chapin FS, Matson PA, Mooney HA (2002) Principles of

terrestrial ecosystem ecology. Springer, New York, p 436

pp

Chee-Sanford JC, Williams MM II, Davis A, Sims GK (2006)

Do microorganisms influence seed-bank dynamics? Weed

Sci 54:575–587

Clark C, Cleland EE, Collins SL, Fargione JE, Gough L, Gross

KL, Pennings SC, Suding KN, Grace JB (2007) Envi-

ronmental and plant community determinants of species

loss following nitrogen enrichment. Ecol Lett 10:596–607

Commander L, Merritt DJ, Rokich DP, Dixon KW (2009) Seed

biology of Australian arid zone species: germination of 18

species used for rehabilitation. J Arid Environ 73:617–625

Cornelissen J, Callaghan TV, Alatalo JM, Michelsen A, Gra-

glia E, Hartley AE, Hik DS, Hobbie SE, Press MC,

Robinson CH, Henry GHR, Shaver GR, Phoenix GK,

Gwynn Jones D, Jonasson S, Chapin FS, Molau U, Neill

C, Lee JA, Melillo JM, Sveinbjornsson B, Aerts R (2001)

Global change and arctic ecosystems: is lichen decline a

function of increases in vascular plant biomass? J Ecol

89:984–994

Cox RD, Allen EB (2008) Composition of soil seed banks in

southern California coastal sage scrub and adjacent exotic

grassland. Plant Ecol 198:37–46

Davis AS (2007) Nitrogen fertilizer and crop residue effects on

seed mortality and germination of eight annual weed

species. Weed Sci 55:123–128

Fenn ME, Baron JS, Allen EB, Rueth HM, Nydick KR, Geiser

L, Bowman WD, Sickman JO, Meixner T, Johnson DW,

Neitlich P (2003) Ecological effects of nitrogen deposition

in the Western United States. BioScience 53:404–420

Figueroa JA, Teillier S, Jaksic FM (2004) Composition size

and dynamics of the seed bank in a Mediterranean

shrubland of Chile. Austral Ecol 29:574–584

Fisher J, Lorenagan WA, Dixon K, Veneklaas EJ (2009) Soil

seed bank compositional change constrains biodiversity in

an invaded species-rich woodland. Biol Conserv 142:

29–256

Heneghan L, Miller SP, Baer S, Callaham MA Jr, Montgomery

J, Pavao-Zuckerman M, Rhoades CC, Richardson S

(2008) Integrating soil ecological knowledge into resto-

ration management. Restor Ecol 16:608–617

Hopfensperger KN (2007) A review of similarity between seed

bank and standing vegetation across ecosystems. Oikos

116:1438–1448

Horswill P, O’Sullivan O, Phoenix GK, Lee JA, Leake JR

(2008) Base cation depletion eutrophication and acidifi-

cation of species-rich grasslands in response to long-term

simulated nitrogen deposition. Environ Pollut 155:

336–349

Hunt JR, Cousens RD, Knights SE (2009) Heliotropium eu-

ropaeum only germinates following sufficient rainfall to

allow reproduction. J Arid Environ 73:602–610

Jankju-Borzelabad M, Griffiths H (2006) Competition for

pulsed resources: an experimental study of establishment

and coexistence for an arid-land grass. Oecologia

148:555–563

Jentsch A, Kreyling J, Beierkuhnlein C (2007) A new gener-

ation of climate-change experiments: events not trends.

Front Ecol Environ 5:365–374

Johnson S (2004) Effects of water level and phosphorus

enrichment on seedling emergence from marsh seed banks

collected from northern Belize. Aquat Bot 79:311–323

Keeley JE, Fotheringham CJ (1998) Smoke-induced seed

germination in California chaparral. Ecology 79:2320–

2336

272 Plant Ecol (2010) 210:263–273

123

Page 11

Kirkham FW, Kent M (1997) Soil seed bank composition in

relation to the above-ground vegetation in fertilized and

unfertilized hay meadows on a Somerset peat moor.

J Appl Ecol 34:889–902

LeBauer DS, Treseder KK (2008) Nitrogen limitation of net

primary productivity in terrestrial ecosystems is globally

distributed. Ecology 89:371–379

Leishman MR, Masters GJ, Clarke IP, Brown VK (2000) Seed

bank dynamics: the role of fungal pathogens and climate

change. Funct Ecol 14:293–299

Luna B, Moreno JM (2009) Light and nitrate effects on seed

germination of Mediterranean plant species of several

functional groups. Plant Ecol 203:123–135

Luzuriaga LL, Escudero A, Olano JM, Loidi J (2005) Regen-

erative role of seed banks following an intense soil dis-

turbance. Acta Oecol 27:57–66

Maestre FT, Escudero A, Martınez I, Guerrero C, Rubio A

(2005) Does spatial pattern matter to ecosystem func-

tioning? Insights from biological soil crusts. Funct Ecol

19:566–573

Manning P, Houston K, Evans T (2009) Shifts in seed size

across experimental nitrogen enrichment and plant density

gradients. Basic Appl Ecol 10:300–308

Maranon T (2001) Ecologıa del banco de semillas y dinamica

de comunidades mediterraneas. In: Zamora R, Pugnaire FI

(eds) Ecosistemas mediterraneos. Analisis funcional.

CSIC, AEET, Madrid, pp 153–182

Matesanz S, Escudero A, Valladares F (2008) Additive effects

of a potentially invasive grass and water stress on the

performance of seedlings of gypsum specialists. Appl Veg

Sci 11:287–296

Mendez M, Garcıa D, Maestre FT, Escudero A (2008) More

ecology is needed to restore Mediterranean ecosystems: a

reply to Valladares and Gianoli. Restor Ecol 16:210–216

Miyagi KM, Kinugasa T, Hikosaka K, Hirose T (2007) Ele-

vated CO2 concentration, nitrogen use and seed produc-

tion in annual plants. Glob Chang Biol 13:2161–2170

Murray PJ, Cook R, Currie AF, Dawson LA, Gange AC,

Grayston SJ, Treonis AM (2006) Interactions between

fertilizer addition plants and the soil environment: impli-

cations for soil and faunal structure and diversity. Appl

Soil Ecol 33:199–207

Petru M, Tielborger K (2008) Germination behaviour of annual

plants under changing climatic conditions: separating

local and regional environmental effects. Oecologia

155:717–728

Phoenix GK, Hicks WK, Cinderby S, Kuylenstierna JCI, Stock

WD, Dentener FJ, Giller KE, Austin AT, Lefroy RDB,

Gimeno BS, Ashmore MS, Ineson P (2006) Atmospheric

nitrogen deposition in world biodiversity hotspots: the

need for a greater global perspective in assessing N

deposition impacts. Glob Chang Biol 12:470–476

Porter WM, Robson AD, Abbott LK (1987) Factors controlling

the distribution of vesicular-arbuscular mycorrhizal fungi

in relation to soil pH. J Appl Ecol 24:663–672

Rahlao SJ, Milton SJ, Esler KJ, Van Wilgen BW, Barnard P

(2009) Effects of invasion of fire-free arid shrublands by a

fire-promoting invasive alien grass (Penisetum setaceum)

in South Africa. Austral Ecol 34:920–928

Rashid I, Reshi Z, Allaie RR, Wafai BA (2007) Germination

ecology of invasive alien Anthemis cotula helps it syn-

chronise its successful recruitment with favourable habitat

conditions. Ann Appl Biol 150:361–369

Reth S, Reichstein M, Falge E (2005) The effect of soil water

content, soil temperature, soil pH-value and the root mass

on soil CO2 efflux—a modified model. Plant Soil 268:

21–33

Reyes O, Trabaud L (2009) Germination behaviour of 14

Mediterranean species in relation to fire factors: smoke

and heat. Plant Ecol 202:113–221

Rivas-Martınez S (1987) Memoria del mapa de series de

vegetacion de Espana. ICONA, Madrid

Roda F, Avila A, Rodrigo A (2002) Nitrogen deposition in

Mediterranean forests. Environ Pollut 118:205–213

Roem W, Klees H, Berendse F (2002) Effects of nutrient

addition and acidification on plant species diversity and

seed germination in heathland. J Appl Ecol 39:937–948

Sala OE, Chapin FS, Armesto JJ, Berlow E, Bloomfield J,

Dirzo R, Huber-Sanwald E, Huenneke LF, Jackson RB,

Kinzig A, Leemans R, Lodge DM, Mooney HA, Oester-

held M, Poff NL, Sykes MT, Walker BH, Wall DH (2000)

Global Biodiversity Scenarios for the year 2100. Science

287:1770–1774

Salemaa M, Makipaa R, Oksanen J (2008) Differences in the

growth response of three bryophyte species to nitrogen.

Environ Pollut 152:82–91

Stevens C, Dise NB, Mountford JO, Gowing DJ (2004) Impact

of nitrogen deposition on the species richness of grass-

lands. Science 303:1876–1879

van den Berg L, Dorland E, Vergeer P, Hart MAC, Bobbink R,

Roelofs JGM (2005) Decline of acid-sensitive plant spe-

cies in heathland can be attributed to ammonium toxicity

in combination with low pH. New Phytol 166:551–564

Van der Waal C, de Kroon H, de Boer WF, Heitkonig IMA,

Skidmore AK, de Knegt HJ, van Langeveldem F, van

Wieren SE, Grant RC, Page BR, Slotow R, Kohi EM,

Mwakiwa E, Prins HHT (2009) Water and nutrients alter

herbaceous competitive effects on tree seedlings in a

semi-arid savannah. J Ecol 97:430–439

Venable DL (2007) Bet hedging in a guild of desert annuals.

Ecology 88:1086–1090

Vourlitis GL, Pasquini SC (2009) Experimental dry-season N

deposition alters species composition in southern Cali-

fornian mediterranean-type shrublands. Ecology 90:2183–

2189

Vourlitis GL, Pasquini SC, Mustard R (2009) Effects of dry-

season N input on the productivity and N storage of

Mediterranean-type shrublands. Ecosystems 12:473–488

Xia J, Wan S (2008) Global response patterns of terrestrial plant

species to nitrogen addition. New Phytol 179:428–439

Yu S, Bell D, Sternberg M, Kutiel P (2008) The effect of

microhabitats on vegetation and its relationship with

seedlings and soil seed bank in a Mediterranean coastal

sand dune community. J Arid Environ 72:2040–2053

Zabaleta ES, Shaw MR, Chiariello NR, Mooney HA, Field CB

(2003) Additive effects of simulated climate changes

elevated CO2 and nitrogen deposition on grassland

diversity. Proc Natl Acad Sci 100:7650–7654

Plant Ecol (2010) 210:263–273 273

123