How much will it cost to save grassland diversity?

Abstract

Conservation initiatives are failing to arrest the global loss of biodiversity. From our mechanistic studies of ecology and economics, we suggest that for grazing lands the root cause of this failure is a powerful economic deterrent to measures designed to protect diversity. We identify an exponential relationship between monetary returns and intensification of farming methods over an extremely wide range of grassland productivities and farm systems. At intermediate to high levels of fertility, however, this exponential increase in financial benefit from intensification is associated with a decline in biodiversity and an acceleration of the ecological processes driving species losses from grassland ecosystems.

Full text

Page 1

How much will it cost to save grassland diversity?
J.G. Hodgsona,*, G. Montserrat-Martı´b, J. Tallowinc, K. Thompsond,
S. Dı ´aze, M. Cabidoe, J.P. Grimed, P.J. Wilsond, S.R. Bandd, A. Bogardf,
R. Cabidog, D. Ca ´ceresh, P. Castro-Dı´ezb, C. Ferreri, M. Maestro-Martı´nezb,
M.C. Pe ´rez-Rontome ´b, M. Charlesf, J.H.C. Cornelissenj, S. Dabbertk,
N. Pe ´rez-Harguindeguye, T. Krimlyk, F.J. Sijtsmal, D. Strijkerm,
F. Vendraminie, J. Guerrero-Campob, A. Hyndf, G. Jonesf,
A. Romo-Dı´ezn, L. de Torres Espunyb,
P. Villar-Salvadorb, M.R. Zake
aPeak Science and Environment, Station House, Leadmill Hathersage, Hope Valley S32 1BA, UK
bDepartamento Ecologı ´a Funcional y Biodiversidad, Instituto Pirenaico de Ecologı ´a (CSIC) Aptdo. 202, E-50080 Zaragoza, Spain
cInstitute of Grassland and Environmental Research, North Wyke Research Station, Okehampton, Devon EX20 2SB, UK
dUnit of Comparative Plant Ecology, Department of Animal and Plant Sciences, The University, Sheffield S10 2TN, UK
eInstituto Multidisciplinario de Biologia Vegetal and Ca ´tedra de Biogeografı´a, FCEyN (CONICET- Universidad Nacional de Co ´rdoba),
Casilla de Correo 495 (5000), Co ´rdoba, Argentina
fDepartment of Archaeology, The University, Sheffield S1 4ET, UK
gDepartamento de Producio ´n Animal, Facultad de Ciencias Agropecuarias, Universidad Nacional de Co ´rdoba, Casilla de Correo 509,
5000 Co ´rdoba, Argentina
hDepartamento de Desarrollo Rural, Facultad de Ciencias Agropecuarias, Universidad Nacional de Co ´rdoba, Casilla de Correo 509,
5000 Co ´rdoba, Argentina
iDepartamento de Agricultura y Economı ´ a Agraria, Universidad de Zaragoza, Miguel Servet 177, 50013 Zaragoza, Spain
jDepartment of Systems Ecology, Faculty of Earth and Life Sciences, Free University, De Boelelaan 1087, 1081 HV Amsterdam, The Netherlands
kInstitut fu ˜r Landwirtschaftliche Betriebslehre, Universita ¨t Hohenheim, Schloss Osthof-Su ˜d, 70599 Stuttgart, Germany
lFaculty of Economics, University of Groningen, Postbus 800, 9700 AV Groningen, Netherlands
mFaculty of Spatial Sciences, University of Groningen, Postbus 800, 9700 AV Groningen, Netherlands
nInstitut Bota `nic de Barcelona, Parc Montjuı ¨c, Av. dels Muntanyans s/n 08038, Barcelona, Spain
Received 24 March 2004; received in revised form 7 July 2004; accepted 7 July 2004
Abstract
Conservation initiatives are failing to arrest the global loss of biodiversity. From our mechanistic studies of ecology and econom-
ics, we suggest that for grazing lands the root cause of this failure is a powerful economic deterrent to measures designed to protect
diversity. We identify an exponential relationship between monetary returns and intensification of farming methods over an extre-
mely wide range of grassland productivities and farm systems. At intermediate to high levels of fertility, however, this exponential
increase in financial benefit from intensification is associated with a decline in biodiversity and an acceleration of the ecological proc-
esses driving species losses from grassland ecosystems.
? 2004 Elsevier Ltd. All rights reserved.
Keywords: Biodiversity; Site productivity; Economic factors; Effectiveness of conservation initiatives
0006-3207/$ - see front matter ? 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biocon.2004.07.016
*Corresponding author. Tel.: +44 1433650381.
E-mail address: j.hodgson@sheffield.ac.uk (J.G. Hodgson).
www.elsevier.com/locate/biocon
Biological Conservation 122 (2005) 263–273
BIOLOGICAL
CONSERVATION

Page 2

1. Introduction
Efforts to protect declining populations of animals
and plants include documentation of the circumstances
of species loss (Amar et al., 2003), studies of habitat
requirements (Lo ˜hmus, 2003), and the development of
management agreements that include financial compen-
sation for farmers (Ferraro and Kiss, 2002). Recent re-
views (Balmford et al., 2002; Vitousek et al., 1997)
indicate that these initiatives are failing to arrest the
processes of diversity loss; many species-rich ecosys-
tems have rapidly declined in both biodiversity and sur-
facearea, andmostfrom
intensification of land use. Here, therefore, we investi-
gate the apparent conflict between the maintenance of
high biodiversity and the maximising of economic prof-
it. We use grasslands as a model system. The cultural
and economic aspects of grassland management are rel-
atively well understood and grazing lands are a very
important habitat, including many biodiverse and con-
servationally important ecosystems and occupying 22
million km2, 17% of the world?s land surface, over an
exceptional range of climatic and edaphic conditions
(Steinfeld et al., 1997). Moreover, grasslands are beset
by severe conservation problems (Schellnhuber et al.,
2001).
The main purposes of agriculture are to feed local
communities and, through producing enough food for
non-local communities as well, to generate an income
comparable to that of non-agricultural workers. In con-
sequence, the most fundamental relationship driving
land use change and biodiversity losses in grazing lands
is that between site productivity and economic yield.
Unfortunately, broad studies of carrying capacity of
grasslands (McNaughton et al., 1989; Oesterheld et al.,
1992, 1998; Fritz and Duncan, 1993) have not estimated
economic yield. Moreover, arguably the most agronom-
ically important systems, intensively managed grass-
lands (heavily fertilised and reseeded) were excluded
from these studies. For this reason, the initial step in
our analysis is to compare estimates of habitat produc-
tivity with those for livestock carrying capacity and
farming income. In this analysis, we focus on three con-
trasted regions, subtropical Argentina, Mediterranean
Spain and temperate England. Subsequently, for one
area, England, we examine the relationship between eco-
nomic yield and biodiversity.
The economic role of biodiversity within ecosystems
remains very much an enigma. Some predict compata-
bility between the goals of achieving high economic
yields and high biodiversity (Hector et al., 1999; Tilman,
1999; Tilman et al., 2001), while the analyses of others
suggests incompatability (Balmford et al., 2002; Tilman
et al., 2002; Vitousek et al., 1997). Our comparisons of
biodiversity and economic yield are, therefore, expected
to inform policy on the conservation of biodiversity.
asingle cause,the
2. Theoretical and logistical background
The work was carried out in three stages, each involv-
ing a variety of interlinked procedures (Fig. 1). During
the first stage, we used ecological methodologies to esti-
mate leaf nitrogen, and by inference the productivity, of
our target vegetation types. The second stage was car-
ried out totally independently by agronomists and in-
volved an assessment of both carrying capacity and
economic profit. In the third stage, we considered the
relationship between biodiversity and economic yield.
Our progress depended critically on an ability to assess
agronomic, ecological and economic variables in a
standardised manner across continents and very differ-
ent vegetation types. It also required the integration of
ecological and economic approaches. This section ex-
plains the theoretical and logistical basis for the choice
of the variables used and describes some of the difficul-
ties in bringing together the disparate disciplines of ecol-
ogy and economics.
2.1. Site productivity
We have opted to assess leaf nitrogen concentration,
a character universally recognised as having an impor-
tant impact upon potential productivity and ecosystem
function (Field and Mooney, 1986; Reich et al., 1992).
Nitrogen concentration is one of a whole suite of chem-
ical and structural characteristics associated with species
capable of rapid growth (Dı´az et al., 2004; Field and
Mooney, 1986; Pearsall, 1950; Wright et al., 2004) and
is used here to assess the productivity of our studied
grassland communities. It correlates positively with con-
centrations of other inorganic plant macronutrients
(Garten, 1976; Thompson et al., 1997; Vendramini
et al., 2000) and with species attributes of fundamental
importance to nutrient use and cycling within ecosys-
tems [e.g., maximum relative growth rate, nutritional
value to herbivores, litter decomposition rate (Dı ´az
et al., 2004; Grime et al., 1997)].
Leaf nitrogen can be estimated quickly and subse-
quently used in combination with plant cover values
from releve ´s abstracted from the phytosociological liter-
ature to estimate site productivity. The method is rapid
but, because measurements of species abundance do not
relate to plant biomass, our values identify the quality
rather than necessarily the quantity of plant material
available to grazing livestock. The method is, therefore,
inappropriate for assessing productivity in highly eroded
areas and other sites where the cover of vegetation is
naturally low. We have, therefore, chosen to exclude
sites with incomplete vegetation cover (total cover val-
ues for all species <90%). Winter annuals, which live
for only half of the year but which are an important
component of some Spanish communities, were also a
problem. We gave them a weighting of 0.5.
264
J.G. Hodgson et al. / Biological Conservation 122 (2005) 263–273

Page 3

Previous studies (Fritz and Duncan, 1993; McNaugh-
ton et al., 1989; Oesterheld et al., 1992, 1998) have used
direct measurement of plant productivity or indirect sur-
rogates (e.g., from rainfall data or satellite imagery).
Each of these approaches also has its problems. Direct
measurements are time consuming and may require spa-
tial and temporal replication to ensure general rele-
vance. Rainfall data are often inappropriate. For
example, in England, because of differences in the inten-
sity of land use, areas with the same rainfall can, and of-
ten do, differ markedly in productivity. The third
alternative, satellite imagery, is beyond the technical re-
sources of many laboratories. Therefore, despite its lim-
itations, we favour the use of leaf nitrogen as a predictor
of ecosystem productivity. It is an approach that can
readily be extended to grazing systems in other parts
of the world.
2.2. Carrying capacity and economic returns
Our estimates of habitat productivity derive from
known interrelationships between nutrient dynamics
and ecosystem function. Ideally, therefore, our assess-
ments of carrying capacity and economic returns would
have also involved universally applicable protocols,
similarly underpinned by agroeconomic theory. In prac-
tice, this was not possible and as a result the data relate
primarily to published national and regional guidelines
1a Develop predictor
equation for estimating
leaf nitrogen from
simple leaf characters
Measure leaf
characters for >1000
species
1b Estimate leaf
nitrogen for >1000
species
Abstract floristic
descriptions of
vegetation (from
phytosociological
literature)
1c Estimate leaf
nitrogen for vegetation
types
2a Estimate
mathematical
relationship between
livestock carrying
capacity and leaf
nitrogen
Vegetation survey of
grasslands of C.
England
(>700 samples)
2b Estimate
mathematical
relationship between
economic profit and
leaf nitrogen
3 Estimate
mathematical
relationship between
economic profit and
biodiversity
Abstract data on
livestock carrying
capacity (from
agronomic literature)
Convert
data on sheep and
cattle into standard
livestock units
Calculate economic
marginal returns from
agronomic sources
Fig. 1. Flow diagram showing the interrelationships between the measurements, analyses and outputs. The main procedures are numbered and
central with additional ecological inputs to the left and additional agro-economic ones to the right. (1) Stages in the process of estimating soil fertility.
(2) Identifying stocking rate and economic yield and their relationship to fertility. (3) Interrelations between biodiversity and economic yield.
J.G. Hodgson et al. / Biological Conservation 122 (2005) 263–273
265

Page 4

for broad vegetation community types. We have as-
sumed that the different methods of assessing carrying
capacity used in each country are equivalent. Another
assumption has been that the protocol of Halle and
Soffe (1988) was sufficiently robust to be applied to
our international range of livestock. Their procedure
transforms data on sheep and cattle into standard live-
stock units, a necessary precursor before any compari-
sons of stocking rates in different vegetation types.
This conversion is not necessary when estimating eco-
nomic profit and marginal economic returns were de-
rived from data on the local market value of animals
and on direct and indirect agricultural costs. For sim-
plicity in these economic calculations, we have assumed
that maximising stocking rate maximises profit. In the
future, we will need to refine this approach to economic
matters and take more account of the complexities of
today?s world markets: for example, high product qual-
ity may produce a higher economic yield than maximiz-
ing product output (e.g., organic vs. non-organic
farming; dairy vs. beef cattle).
2.3. Differences in ecological and economic perspectives
Economists and ecologists have different back-
grounds and perspectives making consensus difficult.
In particular, most ecologists see the relationship be-
tween fertility, stocking rate and marginal returns as a
consequence of two separate but interrelated mechanis-
tic processes. They consider that the first, and funda-
mental, process relates to ecosystem dynamics. The
supply of nutrients (and water) controls rates of plant
growth and nutritional quality of herbage and hence
livestock carrying capacity and, less precisely, marginal
returns. The second process is that of agricultural inter-
vention with ecosystem processes. This involves an in-
crease in productivity for economic gain. The level to
which productivity will be raised relates to factors such
as the cost of agricultural improvement relative to the
economic rewards. The impacts of this process are, how-
ever, complicated by the fact that benefits may only be
short-term if, through ecological ignorance, modifica-
tions to the ecosystem cannot be sustained in the longer
term (see Balmford et al., 2002). By contrast, some econ-
omists tend to consider the relationship between fertility
and stocking rate simply in terms of economic causality:
high economic yields imply the use of additional inputs
and the level of the input is primarily determined by the
relative price of the input (Hayami and Ruttan, 1971).
As future work begins to address economic theories
more explicitly and directly, these differences in ?mecha-
nistic semantics?, which relate to whether the impacts of
ecological or economic processes are pre-eminent, could
potentially become a major obstacle to effective interdis-
ciplinary collaboration. Fortunately, such problems
were slight during the current investigation.
3. Materials and methods
3.1. Study areas
We studied three contrasted regions: central-western
Argentina (lowland and montane; subtropical with hot
summers, no rain in winter), north-eastern Spain (low-
land; mediterranean climate with hot dry summers and
wet winters) and England (lowland and montane; tem-
perate with both summer and winter cool and wet).
The study areas are described in more detail in Dı ´az
et al. (2004). In Argentina, we worked on five broad veg-
etation types using data from Cabido et al. (1992) and
Cantero et al. (2001). These included three lowland
types [semi-arid wood pasture (chaco) and two variants
of arid scrub pasture (monte): (i) with rainfall 150
mm y?1and (ii) rainfall <150 mmy?1] and two from
the mountains [wood pasture (mountain chaco) and
grassland]. Our six Spanish vegetation types were dry
shrubby pasture (Rosmarino-Ericion), dry pasture with
annuals (two types: Eremopyro-Lygeion and Thero-
Brachypodion), dry shrubby pasture on gypsum soils
(Gypsophilion), nitrophilous communities in arid areas
(Salsolo-Peganion)andmoist
Cynodontion) – Braun Blanquet and Bolo ´s (1957).
The 12 grasslands from England relate to Rodwell
(1991–2000) and included calcareous grassland (two
types: CG1/CG2 and CG3/CG4), fen meadow (two
types: M24 and M25), heathland (two types: H8/H12
and H7), improved pasture (MG 7), mesotrophic
pasture (MG5), rough pasture (MG9), semi-improved
pasture (MG6), semi-improved rush pasture (MG6/
MG10) and upland grassland (U1, U3 and U4). Most
of the vegetation types included in this study are natural
or semi-natural and are subjected to various degrees of
livestock grazing from very extensive to intensive. The
most productive grasslands in England, classified by
Rodwell (1991–2000) as MG7, are, however, sustained
by high inputs of fertilizer and by periodic ploughing
and reseeding.
pasture(Trifolieto-
3.2.
productivity
Estimatingleafnitrogenconcentrationand
Insufficient data were available to use direct measure-
ments of the productivity of plant communities or spe-
cies. We, therefore, estimated leaf nitrogen by multiple
regression using the following easily measured leaf char-
acters: maximum leaf area for individual leaves (mm2),
specific leaf area (mm2leaf area mg leaf mass?1), dry
matter content (100 · dry mass of leaf/saturated mass
of leaf), leaf toughness (= leaf tensile strength; Newtons
mm leaf width?1) and leaf thickness (mm). Details on
ecological interpretation and measurement of traits are
given in Hendry and Grime (1993), Grime et al.
(1997), Westoby (1998) and Cornelissen et al. (2003).
266
J.G. Hodgson et al. / Biological Conservation 122 (2005) 263–273

Page 5

Values are generally the average of at least six replicates
per species. For all measurements only material from
healthy, sexually mature plants growing in the field in
unshaded habitats was used and species were included
from a spectrum of fertile and infertile and droughted
and undroughted habitats. Leaf nitrogen values (mgg?1)
were abstracted from Thompson et al. (1997), Vendra-
mini et al. (2000) and unpublished data sources. All four
predictor variables were offered as independent variables
to each of four regressions (data for the three countries
separately and the combined dataset for all three coun-
tries). The predictor variables, and leaf nitrogen, were
log10-transformed and the dry matter content data were
square root transformed to normalize their distributions
prior to carrying out multiple regression analyses.
The multiple regression equation was subsequently
used to estimate leaf nitrogen concentration for more
than 1000 species. These estimates of leaf nitrogen con-
centration for herbaceous and woody species were then
combined with phytosociological data for our 23 vegeta-
tion types listed above (Braun Blanquet and Bolo ´s,
1957; Cabido et al., 1992; Cantero et al., 2001; Rodwell,
1991–2000) to assess the productivity. Percentage cover
values for individual species or in the case of Rodwell
(1991–2000), where data were only available in the form
of summary tables, average scores, were multiplied by
estimates of leaf nitrogen and an average value of leaf
nitrogen was calculated for each vegetation type.
3.3. Carrying capacity and economic returns
We estimated the carrying capacity of each vegeta-
tion type using published national or regional data/
guidelines for livestock numbers carried hectare?1and
type and weight of livestock for broad vegetation com-
munity types (Cabido, 1998; Ca ´ceres, 2001; Crofts and
Jefferson, 1999; Daget and Poissonet, 1972; Kirkham
and Wilkins, 1994; MAFF, 1995–1996; Peretti, 1998;
Tallowin et al., 2002), augmented for Argentina by
stocking level figures from unpublished reports from
governmental agencies and from consultations with lo-
cal agronomists and farmers. To allow comparison be-
tween and within regions, the livestock data were then
converted to livestock units (LU) hectare?1using a
standard system in which a non-lactating bovine weigh-
ing 550 kg was 1 LU. Halle and Soffe (1988) provide
conversion values for the full liveweight range of sheep
and cattle. However, for the purposes of this study the
livestock data from Spain related to sheep with an aver-
age weight of 45 kg, which represented 0.10 LU. The
data from Argentina related to cattle of 450 kg average
liveweight, which represented 0.82 LU. The livestock
data from England related to sheep with an average
weight of 80 kg (0.17 LU), and cattle, which were either
mature, with an average liveweight of 500 kg (0.91 LU)
or immature with an average weight of 350 kg (0.64
LU). In the Spanish grasslands, there is an almost two-
fold seasonal variation in estimated nitrogen and six-
fold variation in sustainable stocking density for some
plant communities and values represent an average
for all four seasons. In other areas, without a history
of transhumance, separate seasonal analyses were not
carried out.
Estimates of economic outputs for the UK were
based upon average gross margin data for the year
2000 (Meat and Livestock Commission, 2001) for beef
suckler herds in lowland, hill and upland grasslands.
The data take account of cow subsidy, calf sale value
at 250–300 kg live weight and assumed that 89% of
calves were reared. Data for Argentina also relate to
beef cattle with calves and were similarly calculated
(Peretti, 1998) after extensive discussions with farmers.
It is assumed that there is only 70% weaning year?1
and that the calves are sold when 125–140 kg. No subsi-
dies apply. Spanish values are for sheep with lambs and
include direct and indirect costs and subsidies. They re-
fer to average regional costings for Aragon in 1997 [eco-
nomic profit €16.3 sheep?1(direct and indirect costs
€75.2; economic returns €91.5, including a subsidy of
€20.2); Toma ´s, 1999]. Further details are available from
the authors.
3.4. The relationship between biodiversity and economic
returns
In this analysis, we used unpublished vegetation data
for the grazed grasslands of Central England consisting
of more than seven hundred 1 m2sampling points and
including both upland and lowland and calcareous,
acidic, wet and ?improved? pasture. For each quadrat,
the leaf nitrogen concentration of the vegetation was as-
sessed by the methodology described in Section 3.2. A
general equation defining economic yield from leaf
nitrogen concentration, described in Section 4 (Results),
was then used to convert the estimate of productivity to
one relating to marginal returns. The relationship be-
tween small-scale biodiversity (species m?2) and eco-
nomic yield was then investigated.
4. Results
Leaf nitrogen concentration was derived as a function
of specific leaf area and leaf toughness in each of the
Argentine, English and Spanish datasets (Table 1). No
term of the predictor equation differed significantly be-
tween the three countries and we were able to generate
a general predictor equation for use in subsequent anal-
yses by combining the three datasets (Table 1). The close
relationship between measured values and those pre-
dicted by the general regression equation for all three
countries is illustrated in Fig. 2. Maximum leaf size, leaf
J.G. Hodgson et al. / Biological Conservation 122 (2005) 263–273
267

Page 6

thickness and leaf dry matter did not contribute signifi-
cantly to any of the predictor equations.
Estimates of leaf nitrogen concentration, livestock
carrying capacity and marginal returns for our target
vegetation types are included in Table 2. Nitrogen con-
centration of the vegetation successfully predicted both
livestock carrying capacity and economic output for
all three regions (Fig. 3(a) and (b)). Moreover, a linear
increase in leaf nitrogen was associated with an expo-
nential increase in both carrying capacity and marginal
returns. Biodiversity did not show the same simple rela-
tionship. For the grasslands of Central England vegeta-
tion of low biodiversity was shown to occur throughout
the full range of economic yields and nitrogen concen-
trations, but high biodiversity of higher plants was re-
stricted to low and intermediate values (Fig. 4). This
same pattern was consistently observed in upland and
lowland regions and on calcareous and non-calcareous
geological strata.
5. Discussion
5.1. Conclusions
Within the range of productivity associated with our
study, increasing fertility causes a large, apparently
exponential, increase in livestock carrying capacity
and in marginal returns. High levels of biodiversity
(species m?2) are, however, confined to less productive
conditions, with an inherently low carrying capacity for
livestock and low marginal returns. Thus, over a wide
range of productivity, management of grasslands to
maintain high biodiversity is incompatible with man-
agement for maximum economic profit. Moreover,
the relationships observed were both statistically strong
and consistently expressed in very different environ-
ments despite the fact that: (a) the analyses were car-
riedoutusingestimates
surrogate for productivity, (b) the protocols for assess-
ing carrying capacity of livestock differed between
countries and (c) the approach to economic matters
was simplistic and ignored many of the complexities
of today?s world markets. The mathematical relation-
ships identified in Fig. 3 are, however, an oversimplifi-
cation.Thetheoretically
between productivity, economic yield and agricultural
intensification are described by a ?humped-backed?
curve (Engel, 2003). At low levels of soil fertility, there
is a positive relationship between productivity and
marginal returns, and an incentive for further agricul-
tural intensification and fertilizer input. At intermedi-
ate levels, the increase in marginal returns declines
and at high soil fertility the addition of fertilizers is
not cost-effective and marginal yield decreases. Agricul-
tural intensification will stop at an economic optimum
before this last stage. We suspect that the ?hump? and
downward curve is missing from our dataset, because
on more fertile soils arable crops are the economically
preferred option. This reservation about the true shape
of the curve does not, however, affect our conclusions.
Productivity, livestock carrying capacity and biodiver-
sity are all strongly interrelated and high biodiversity
and high economic yield are incompatible at higher lev-
els of productivity.
ofleafnitrogenasa
expectedrelationships
Table 1
Predictor equations for log10leaf N (mgg?1)
All data ± SEArgentina ± SEEngland ± SE Spain ± SE
Log10specific leaf area
Log10leaf toughness
Constant
R2
Number of species
P
+0.255 ± 0.040
?0.167 ± 0.022
+0.063 ± 0.048
0.57
174
1 · 10?20
+0.294 ± 0.081
?0.203 ± 0.045
+0.052 ± 0.081
0.57
47
1 · 10?10
+0.248 ± 0.080
?0.150 ± 0.029
+0.044 ± 0.110
0.45
83
4 · 10?12
+0.472 ± 0.093
?0.160 ± 0.042
?0.148 ± 0.106
0.63
44
1 · 10?10
Specific leaf area and leaf toughness (=leaf tensile strength) are measured in mm2leaf area mg leaf mass?1and Newtons mm leaf width?1,
respectively. The three predictor variables, maximum leaf size, leaf dry matter content and leaf thickness that did not contribute significantly to any
of the regression equations, are excluded from the table.
-0.2
0.0
0.2
0.4
0.6
0.8
-0.4-0.200.20.40.60.8
Measured log10 leaf N (% dry wt)
Predicted log10 leaf N (% dry wt)
Fig. 2. The correspondence between the predictor equation for leaf
nitrogen and measured leaf nitrogen. Argentina, England and Spain
are represented by black, grey and white squares, respectively. Each
data point represents a species.
268
J.G. Hodgson et al. / Biological Conservation 122 (2005) 263–273

Page 7

Table 2
Estimates of leaf nitrogen, carrying capacity and economic yield for Argentine, English and Spanish grazing lands
Log10leaf N (mgg?1)Carrying capacity
(livestock units, ha?1y?1)
Economic yield including
subsidies (£ha?1y?1)
Argentina (semi–tropical, arid, upland and lowland) – cattle-grazed and water limited
Lowland arid scrub pasture (Monte, rainfall <1500 mmy?1)
Lowland arid scrub pasture (Monte, rainfall 1500 mmy?1)
Lowland wood pasture (Chaco)
Upland wood pasture (Mountain chaco)
0.201
0.215
0.243
0.271
0.04
0.05
0.08
0.20
2.42
5.30
7.12
16.21
Spain (Mediterranean, lowland) – sheep-grazed and water limited (except moist pasture)
Dry shrubby pasture (Rosmarino-Ericion)
Dry pasture with annuals (Eremopyro-Lygeion)
Dry shrubby pasture on gypsum soils (Gypsophilion)
Dry pasture with annuals (Thero-Brachypodion)
Nitrophilous communities in arid areas (Salsolo-Peganion)
Moist pasture (Trifolieto-Cynodontion)
0.197
0.219
0.236
0.251
0.374
0.412
0.14
0.08
0.11
0.20
0.28
0.49
27.05
14.72
20.63
39.10
53.56
95.03
England (temperate, upland and lowland) – variously sheep and cattle grazed; not droughted
Healthland (H8/H12)
Healthland (H7)
Fen meadow (M25)
Upland grassland (U1, U3, U4)
Fen meadow (M24)
Calcareous grassland (C3/C4)
Calcareous grassland (C1/C2)
Rough pasture?(MG9)
Semi-improved rush pasture?(MG6/MG10)
Mesotrophic pasture?(MG5)
Semi-improved pasture?(MG6)
Improved pasture?(MG7)
0.342
0.349
0.367
0.375
0.386
0.393
0.394
0.404
0.419
0.419
0.421
0.443
0.17
0.15
0.52
0.38
0.43
0.88
0.40
1.00
1.45
0.54
1.31
1.75
45.33
37.88
124.30
100.03
147.49
299.70
137.20
343.00
498.49
171.47
447.86
600.25
Plant communities identified by abbreviations in parentheses in the English list are described in detail in Rodwell (1991–2000). Grasslands to which
high or low levels of inorganic fertilizer are normally added are identified by?and?, respectively.
0.01
0.10
1.00
10.00
1.01.52.02.53.0
Estimated leaf nitrogen (% dry wt)
Stocking rate (livestock units ha-1)
1
10
100
1000
1.0 1.52.02.53.0
Estimated leaf nitrogen (% dry wt)
Economic yield including subsidies (£ ha
-1y
-1)
(a)
(b)
Fig. 3. The relationship between estimated leaf nitrogen and (a) livestock carrying capacity and (b) annual economic yield. Each data point
represents a different vegetation type and data from Argentina, England and Spain are represented as in Fig. 1. (a) All data: y = 0.0016e2.382x
(R2= 0.82, n = 23, P = 4 · 10?9) [Argentina R2= 0.88, n = 5, P = 0.02; England R2= 0.79, n = 12, P = 1 · 10?6; Spain R2= 0.80, n = 6, P = 0.02]. (b)
All data: y = 0.0354e3.407x(R2= 0.83, n = 23, P = 1 · 10?9) [Argentina R2= 0.85, n = 5, P = 0.03; England R2= 0.83, n = 12, P = 4 · 10?5; Spain
R2= 0.80, n = 6, P = 0.02].
J.G. Hodgson et al. / Biological Conservation 122 (2005) 263–273
269

Page 8

5.2. Relationship to other work on stocking densities
Our study includes an extremely diverse range of hab-
itats from heathland and semi-desert to intensively man-
aged heavily fertilised grasslands. This is arguably the
widest range of habitats included within a single investi-
gation of this type and unlike similar work (Fritz and
Duncan, 1993; McNaughton et al., 1989; Oesterheld
et al., 1992, 1998), we have both included a direct esti-
mate productivity, rather than an indirect one, rainfall,
and have added an economic dimension. All these
studies are in broad agreement, indicating a strong
positive impact of productivity on carrying capacity.
Significantly, however, the best fit to our data in Fig. 3
is an exponential expression, the more extreme of those
previously suggested for the relationship between plant
productivity (estimated in relation to rainfall) and her-
bivore carrying capacity for semi-natural grasslands;
some previous authors have identified an exponential
(Oesterheld et al., 1992, 1998) and others a log/log
relationship (Fritz and Duncan, 1993; McNaughton
et al., 1989).
5.3. Relationship to other work on biodiversity
Higher economic return, the goal of agricultural
production, is associated with a decline in biodiversity
at intermediate to high productivity (Fig. 4). This find-
ing is consistent with those of Janssens et al. (1998) and
the ?hump-back? model of Grime (1973a,b) (see also
Connell, 1978). This model predicts that under condi-
tions of low productivity, biodiversity will be low sim-
ply because few species can survive such conditions. In
productive habitats, ?monopolists? will tend to predom-
inate and the biodiversity of the vegetation will again
be low. At intermediate productivity, neither of the
two extreme groups is advantaged and in consequence
biodiversity is potentially high. In our study (Fig. 4),
the hump is ?filled? because biodiversity is also affected
by, for example, differences in local or regional species
0
10
20
30
40
50
1 10 100100010000
Estimated economic yield (£ ha-1)
No. of species m-2
0
10
20
30
40
50
110 100100010000
Estimated economic yield (£ ha-1)
No. of species m-2
0
10
20
30
40
50
110100100010000
Estimated economic yield (£ ha-1)
No. of species m-2
(a)
(c)
(b)
Fig. 4. The relationship between biodiversity and annual economic yield for grazing lands in Central England [(a) upland limestone, (b) upland non-
calcareous strata, (c) lowland]. Communities represented include CG2-7, H9, H12, M25, MG4-11, MG13, U1-2, U4-5 and U20 (see Rodwell (1991–
2000) for community descriptions). Economic yields of £10, £100, £1000 and £10000 correspond, respectively, to values of 0.7, 1.4, 2.0 and 2.7 leaf
nitrogen (% dry weight). Lowland England was not subdivided because fewer samples are available from this predominantly arable landscape.
Comparable datasets are not available for Argentina and Spain.
270
J.G. Hodgson et al. / Biological Conservation 122 (2005) 263–273

Page 9

pools, soil type and the immaturity of many ?improved?
grasslands.
5.4. Implications for conservation
Productivity and economic output of grasslands ap-
pear functionally and arithmetically related to each
other. We have been able to explain agro-economic yield
for grazing lands from knowledge of the nutrient re-
source within ecosystems and it has been found possible
to predict how yield and diversity change in response to
increased productivity. Except where soil moisture be-
comes limiting, an increase in nutrient input allows
greater intensity of land use and has the potential to in-
crease marginal returns over a wide range of climatic
conditions. Over the upper part of this range, however,
biodiversity is potentially compromised. Through his-
tory, the apparently exponential nature of the relation-
ship between fertility and yield, illustrated in Fig. 3,
has been a powerful primary economic driver of agricul-
tural land-use change towards ever more productive
grassland and arable systems. Full recognition of the
nature of this relationship, in which small increases in
fertility can lead to large economic gains and small de-
creases in fertility result in a dramatic loss of productiv-
ity and economic yield, requires us to radically review
the theoretical basis of our conservation strategies. Even
where conserving biodiversity is subsidised in environ-
mentally sensitive areas, agreements between farmers
and statutory authorities have proved difficult to imple-
ment, a situation sometimes exacerbated by conflicts be-
tween agricultural and environmental policies (Ka ¨chele
and Dabbert, 2002). We suspect that many of the envi-
ronmental planners who are responsible for brokering
these agreements have underestimated the strength of
the relationship between fertility and economic yield
and we interpret the widespread use of flat-rate compen-
satory payments (e.g., Meat and Livestock Commission,
2001) as evidence of this. In future policy debates and
evaluations, we will need to combine, as we have done
here, ecological and economic indicators (Armsworth
and Roughgarden, 2001; Strijker et al., 2000; Toman,
1998).
The current loss of biodiversity through agricultural
intensification is a relatively new phenomenon. In previ-
ous centuries pastoral agriculture has operated primarily
on the ‘‘left’’, ascending slope of the hump-backed curve
of Fig. 5, a region where rising productivity and biodi-
versity may coincide. In many parts of the world, tech-
nology has now taken us to the ‘‘right’’, descending
slope of the biodiversity curve, where increasing produc-
tivity is resulting in a loss of biodiversity. This is partic-
ularly true in the case of UK, where the level of fertility
is higher than in Argentina or Spain (Table 2). Unfortu-
nately, this ‘‘right’’ slope corresponds to a range of soil
fertilities where marginal returns following intensifica-
tion appear to be still increasing exponentially (Fig.
3(b)), and where in consequence the incentives for
increasing agricultural intensity, and the disincentives
for maintaining existing high biodiversity, are great.
The immediate challenges are: (a) to use our new under-
standing of the relationships between fertility, biodiver-
sity and stocking rate (Figs. 3 and 4), to develop
sustainable, economically viable, biodiverse agriculture
(Balmford et al., 2002) and to improve the effectiveness
of agri-environment schemes [see Kleijn et al. (2001) for
a criticism of such schemes and Osterburg (2001) for a
refutation]; (b) to find viable ecological and economic
procedures to return agriculturally ‘‘improved’’ and de-
graded sites to the ‘‘high biodiversity corridor’’ of Figs.
3 and 5; and (c) to ensure that changes in the rules gov-
erning agricultural subsidies do not cause the abandon-
ment of species-rich grazing land, as seems likely in our
Spanish study site, where each sheep is produced at an
economic loss. The ultimate challenge is to extend our
approach to achieve mechanistic links between all key
ecosystem and economic processes. Without a much im-
proved level of interdisciplinary understanding, effective
global policies for the conservation of biodiversity and
for promoting sustainable agriculture are likely to re-
main extremely ineffectual activities.
Acknowledgements
This work was supported by the Darwin Initiative
(DEFRA-UK),CLIMB(EuropeanScienceFoundation),
the European Union, the British Council, NERC, FON-
CyT,Fundacio ´nAntorchas.WethankE.Orechia(INTA
CruzdelEje)andG.Ferrer,whoprovidedvaluableinfor-
mation and all past and present members of UCPE.
References
Amar, A., Redpath, S., Thirgood, S., 2003. Evidence for food
limitation in the declining hen harrier population on the Orkney
Islands, Scotland. Biological Conservation 111, 377–384.
Fig. 5. Theoretical relationships between grassland biodiversity, con-
servation interest and annual economic yield. The diagram is based
upon Fig. 4 and an assessment of the conservation interest of each
vegetation sample used in that analysis.
J.G. Hodgson et al. / Biological Conservation 122 (2005) 263–273
271

Page 10

Armsworth, P.R., Roughgarden, J.E., 2001. An invitation to
ecological economics. Trends in Ecology and Evolution 16, 229–
234.
Balmford, A., Bruner, A., Cooper, P., Costanza, R., Farber, S., Green,
R.E., Jenkins, M., Jefferiss, P., Jessamy, V., Madden, J., Munro,
K., Myers, N., Naeem, S., Paavola, J., Rayment, M., Rosendo, S.,
Roughgarden, J., Trumper, K., Turner, R.K., 2002. Ecology -
Economic reasons for conserving wild nature. Science 297, 950–
953.
Braun Blanquet, J., Bolo ´s, O. de, 1957. Las comunidades vegetales de
la depresio ´n del Ebro y su dinamismo. Anales de la Estacio ´n
Experimental de Aula Dei 5.
Cabido, M., 1998. Programa de manejo ganadero en establecimientos
del Parque Nacional Quebrada del Condorito. APN-GEF-BIRF,
Co ´rdoba, Argentina.
Cabido, M., Acosta, A., Carranza, M.L., Dı´az, S., 1992. La vegetacio ´n
del chaco arido en el W de la Provincia de Co ´rdoba, Argentina.
Documents Phytosociologiques XIV, 447–456.
Ca ´ceres, D., 2001. Diagno ´stico socio-productivo de pequen ˜os prod-
uctores de la Reserva Hı´drica Provincial Pampa de Achala. APN-
GEF-BIRF, Co ´rdoba, Argentina.
Cantero, J.J., Cabido, M., Nu ´n ˜ez, C., Petryna, L., Zac, M.R., Zobel,
M., 2001. Clasificacio ´n de los pastizales de suelos sobre rocas
metamo ´rficas de las sierras de Co ´rdoba, Argentina. Kurtziana 29,
27–77.
Connell, J.H., 1978. Diversity in tropical rain forests and coral reefs.
Science 199, 1302–1310.
Cornelissen, J.H.C., Lavorel, S., Garnier, E., Dı´az, S., Buchmann, N.,
Gurvich, D.E., Reich, P.B., ter Steege, H., Morgan, H.D., van der
Heijden, M.G.A., Pausas, J.G., Poorter, H., 2003. Handbook of
protocols for standardised and easy measurements of plant
functional traits worldwide. Australian Journal of Botany 51,
335–380.
Crofts, A., Jefferson, R.G., 1999. The Lowland Grassland Manage-
ment Handbook, second ed. English Nature/The Wildlife Trusts,
English Nature, Peterborough, England.
Daget, Ph., Poissonet, J., 1972. Un procedure d?estimation de la valeur
pastorale des paturages. Fourrages 49, 31–40.
Dı´az, S., Hodgson, J.G., Thompson, K., Cabido, M., Cornelissen,
J.H.C., Jalili, A., Montserrat-Martı´, G., Grime, J.P., Zarrinkamar,
F., Asri, Y., Band, S.R., Basconcelo, S., Castro-Dı´ez, P., Funes,
G., Hamzehee, B., Khoshnevis, M., Pe ´rez-Harguindeguy, N.,
Pe ´rez-Rontome ´, M.C., Shirvany, F.A., Vendramini, F., Yazdani,
S., Abbas-Azimi, R., Bogard, A., Boustani, S., Charles, M.,
Dehghan, M., de Torres, L., Guerrero-Campo, J., Hynd, A., Jones,
G., Kowsary, E., Kazemi-Saeed, F., Maestro-Martı´nez, M.,
Romo-Dı´ez, A., Shaw, S., Siavash, B., Villar-Salvador, P., Zak,
M.R.J., 2004. The plant traits that drive ecosystems: evidence from
three continents. Journal of Vegetation Science 15, 295–304.
Engel R., 2003. Winter wheat response to available nitrogen and water.
Available from: <http:landresources.montana.edu/FertilizerFacts/>.
Ferraro, P.J., Kiss, A., 2002. Ecology – direct payments to conserve
biodiversity. Science 298, 1718–1719.
Field, C., Mooney, H.A., 1986. The photosynthesis–nitrogen relation-
ship in world plants. In: Givnish, T.V. (Ed.), On the Economy of
Plant Form and Function. Cambridge University Press, Cam-
bridge, pp. 25–55.
Fritz, H., Duncan, P., 1993. Large herbivores in rangelands. Nature
364, 292–293.
Garten Jr., C.T., 1976. Correlations between concentrations of
elements in plants. Nature 261, 686–688.
Grime, J.P., 1973a. Competitive exclusion in herbaceous vegetation.
Nature 242, 344–347.
Grime, J.P., 1973b. Control of species density in herbaceous vegeta-
tion. Journal of Environmental Management 1, 151–167.
Grime, J.P., Thompson, K., Hunt, R., Hodgson, J.G., Cornelissen,
J.H.C., Rorison, I.H., Hendry, G.A.F., Ashenden, T.W., Askew,
A.P., Band, S.R., Booth, R.E., Bossard, C.C., Campbell, B.D.,
Cooper, J.E.L., Davison, A., Gupta, P.L., Hall, W., Hand, D.W.,
Hannah, M.A., Hillier, S.H., Hodkinson, D.J., Jalili, A., Liu, Z.,
Mackey, J.M.L., Matthews, N., Mowforth, M.A., Neal, A.M.,
Reader, R.J., Reiling, K., Ross-Fraser, W., Spencer, R.E., Sutton,
F., Tasker, D.E., Thorpe, P.C., Whitehouse, J., 1997. Integrated
screening validates primary axes of specialisation in plants. Oikos
79, 259–281.
Halle, R.J., Soffe, R.J., 1988. Primrose McConnell?s The Agricultural
Notebook, 18th ed. Butterworths, London.
Hayami, Y., Ruttan, V.W., 1971. Agricultural Development: An
International Perspective. John Hopkins Press, London.
Hector, A., Schmid, B., Beierkuhnlein, C., Caldeira, M.C., Diemer,
M., Dimitrakopoulos, P.G., Finn, J.A., Freitas, H., Giller, P.S.,
Good, J., Harris, R., Hogberg, P., Huss-Danell, K., Joshi, J.,
Jumpponen, A., Korner, C., Leadley, P.W., Loreau, M., Minns,
A., Mulder, C.P.H., O?Donovan, G., Otway, S.J., Pereira, J.S.,
Prinz, A., Read, D.J., Scherer-Lorenzen, M., Schulze, E.D.,
Siamantziouras, A.S.D., Spehn, E.M., Terry, A.C., Troumbis,
A.Y., Woodward, F.I., Yachi, S., Lawton, J.H., 1999. Plant
diversity and productivity experiments in European grasslands.
Science 286, 1123–1127.
Hendry, G.A.F., Grime, J.P., 1993. Methods in Comparative Plant
Ecology. Chapman and Hall, London.
Janssens, F., Peeters, A., Tallowin, J.R.B., Bakker, J.P., Bekker,
R.M., Fillat, F., Oomes, M.J.M., 1998. Relationship between soil
chemical factors and grassland diversity. Plant and Soil 202, 69–
78.
Ka ¨chele, H., Dabbert, S., 2002. An economic approach for a better
understanding of conflicts between farmers and nature conserva-
tionists – an application of the decision support system MODAM
to the Lower Odra Valley National Park. Agricultural Systems 74,
241–255.
Kirkham, F.W., Wilkins, R.J., 1994. The productivity and response to
inorganic fertilizers of species-rich wetland hay meadows on the
Somerset moors – nitrogen response under hay cutting and
aftermath grazing. Grass and Forage Science 49, 152–162.
Kleijn, D., Berendse, F., Smit, R., Gilissen, N., 2001. Agri-environ-
ment schemes do not effectively protect biodiversity in Dutch
agricultural landscapes. Nature 413, 723–725.
Lo ˜hmus, A., 2003. Do Ural owls (Strix uralensis) suffer from the lack
of nest sites in managed forests?. Biological Conservation 110, 1–9.
MAFF,1995–1996. Environmentally
London.
McNaughton, S.J., Oesterheld, M., Frank, D.A., Williams, K.J., 1989.
Ecosystem-level patterns of primary productivity and herbivory in
terrestrial habitats. Nature 341, 142–144.
Meat and Livestock Commission, 2001. Beef Yearbook 2001. Meat
and Livestock Commission, Milton Keynes.
Oesterheld, M., DiBella, C.M., Kerdiles, H., 1998. Relation between
NOAA-AVHRR satellite data and stocking rate of rangelands.
Ecological Applications 8, 207–212.
Oesterheld, M., Sala, O.E., McNaughton, S.J., 1992. Effect of animal
husbandry on herbivore-carrying capacity at a regional scale.
Nature 356, 234–236.
Osterburg, B., 2001. Stellungnahme fu ¨r das Ministerium fu ¨rverbrauc-
herschutz, erna ¨hrung und landwirtschaft zum artikel ‘‘Agri-envi-
ronment schemes do not effectively protect biodiversity in Dutch
agricultural landscapes’’. In: Kleijn, D., Berendse, F., Smit, R.,
Gilissen, N. (Eds.), Nature, vol. 413, 18, October 2001. pp. 723–
725. Availablefrom: <http://www.bal.fal.de/download/Oster-
burg_Nature_2001.pdf> (2001).
Pearsall, W.H., 1950. Mountains and Moorlands. Bloomsbury Books,
London.
Peretti, M., 1998. Monitoreo econo ´mico de los sistemas productivos
predominantes del sector agropecuario de Co ´rdoba – Resultados
Campan ˜a 97/98 y Proyeccio ´n Campan ˜a 98/99. Serie Economı ´ a
SensitiveAreas.MAFF,
272
J.G. Hodgson et al. / Biological Conservation 122 (2005) 263–273

Page 11

Agraria, vol. VI. Convenio INTA–UNRC-SAGyRR, Co ´rdoba,
Argentina.
Reich, P.B., Walters, M.B., Ellsworth, D.S., 1992. Leaf life-span in
relation to leaf, plant, and stand characteristics among diverse
ecosystems. Ecological Monographs 62, 365–392.
Rodwell, J.S. 1991–2000. British Plant Communities. Cambridge
University Press, Cambridge.
Schellnhuber, H.-J., Kokott, J., Beese, F.O., Fraedrich, K., Kruse-
Graumann, L., Neumann, C., Renn, O., Schulze, E.-D., Tilzer, M.,
Velsinger, P., Zimmermann, H., 2001. World in Transition: Con-
servationandSustainableUseoftheBiosphere.Earthscan,London.
Steinfeld, H., de Haan, C., Blackburn, H., 1997. Livestock–Environ-
ment Interactions. Issues and Options. WRENmedia, Suffolk.
Strijker, D., Sijtsma, F.J., Wiersma, D., 2000. Evaluation of nature
conservation – an application to the Dutch Ecological Network.
Environmental & Resource Economics 16, 363–378.
Tallowin, J.R.B., Smith, R.E.N., Goodyear, J., Bullock, J.M., Lehain,
P., 2002. Sustainable management of lowland purple moor-grass
and rush pastures; constraints and opportunities for livestock
farmers. In: Frame, J. (Ed.), Conservation Pays. British Grassland
Society, Reading, pp. 47–50.
Thompson, K., Parkinson, J.A., Band, S.R., Spencer, R.E., 1997. A
comparative study of leaf nutrient concentrations in a regional
herbaceous flora. New Phytologist 136, 679–689.
Tilman, D., Reich, P.B., Knops, J., Wedin, D., Mielke, T., Lehman,
C., 2001. Diversity and productivity in a long-term grassland
experiment. Science 294, 843–845.
Tilman, D., 1999. Ecology – diversity and production in European
grasslands. Science 286, 1099–1100.
Tilman, D., Cassman, K.G., Matson, P.A., Naylor, R., Polasky, S.,
2002. Agricultural sustainability and intensive production prac-
tices. Nature 418, 671–677.
Toman, M., 1998. Why not to calculate the value of the world?s
ecosystem services and natural capital. Ecological Economics 25,
57–60.
Toma ´s, O., 1999. El futuro del ovino aragone ´s. Surcos 64, 6–9.
Vendramini, F., Dı´az, S., Pe ´rez Harguindguy, N., Cabido, M.,
Llano-Sotelo,J.M.,Castellanos,
quı ´mica y caracteres foliares en plantas de distintos tipos
funcionales del centro-oeste de Argentina. Kurtziana 28, 181–
193.
Vitousek, P.M., Mooney, H.A., Lubchenko, J., Melillo, J.M.,
1997. Human domination of Earth?s ecosystems. Science 277,
494–499.
Westoby, M., 1998. A leaf-height-seed (LHS) plant ecology strategy
scheme. Plant and Soil 199, 213–227.
Wright, I.J., Reich, P.B., Westoby, M., Ackerly, D.D., Baruch, Z.,
Bongers, F., Cavender-Bares, J., Chapin, T., Cornelissen, J.H.C.,
Diemer, M., Flexas, J., Garnier, E., Groom, P.K., Gulias, J.,
Hikosaka, K., Lamont, B.B., Lee, T., Lee, W., Lusk, C., Midgley,
J.J., Navas, M.L., Niinemets, U., Oleksyn, J., Osada, N., Poorter,
H., Poot, P., Prior, L., Pyankov, V.I., Roumet, C., Thomas, S.C.,
Tjoelker, M.G., Veneklaas, E.J., Villar, R., 2004. The worldwide
leaf economics spectrum. Nature 428, 821–827.
A.,2000.Composicio ´n
J.G. Hodgson et al. / Biological Conservation 122 (2005) 263–273
273