A globally coherent fingerprint of climate
change impacts across natural systems
Camille Parmesan* & Gary Yohe†
* Integrative Biology, Patterson Laboratories 141, University of Texas, Austin, Texas 78712, USA
† John E. Andrus Professor of Economics, Wesleyan University, 238 Public Affairs Center, Middletown, Connecticut 06459, USA
...........................................................................................................................................................................................................................
Causal attribution of recent biological trends to climate change is complicated because non-climatic influences dominate local,
short-term biological changes. Any underlying signal from climate change is likely to be revealed by analyses that seek systematic
trends across diverse species and geographic regions; however, debates within the Intergovernmental Panel on Climate Change
(IPCC) reveal several definitions of a ‘systematic trend’. Here, we explore these differences, apply diverse analyses to more than
1,700 species, and show that recent biological trends match climate change predictions. Global meta-analyses documented
significant range shifts averaging 6.1 km per decade towards the poles (or metres per decade upward), and significant mean
advancement of spring events by 2.3 days per decade. We define a diagnostic fingerprint of temporal and spatial ‘sign-switching’
responses uniquely predicted by twentieth century climate trends. Among appropriate long-term/large-scale/multi-species data
sets, this diagnostic fingerprint was found for 279 species. This suite of analyses generates ‘very high confidence’ (as laid down by
the IPCC) that climate change is already affecting living systems.
The Intergovernmental Panel on Climate Change
1
(IPCC) assessed
the extent to which recent observed changes in natural biological
systems have been caused by climate change. This was a difficult task
despite documented statistical correlations between changes in
climate and biological changes
2–5
. With hindsight, the difficulties
encountered by the IPCC can be attributed to the differences in
approach between biologists and other disciplines, particularly
economists. Studies in this area are, of necessity, correlational rather
than experimental, and as a result, assignment of causation is
inferential. This inference often comes from experimental studies
of the effects of temperature and precipitation on the target species
or on a related species with similar habitats. Confidence in this
inferential process is subjective, and differs among disciplines, thus
resulting in the first divergence of opinion within the IPCC.
The second impasse came fromdifferences in perspective onwhat
constitutes an ‘important’ factor. Anyone would consider a cur-
rently strong driver to be important, but biologists also attach
importance to forces that are currently weak but are likely to persist.
In contrast, economic approaches tend to discount events that will
occur in the future, assigning little weight to weak but persistent
forces. Differences of opinion among disciplines can therefore stem
naturally from whether the principal motivation is to assess the
magnitude of immediate impacts or of long-term trajectories. Most
field biologists are convinced that they are already seeing important
biological impacts of climate change
1–4,6–9
; however, they have
encountered difficulty in convincing other academic disciplines,
policy-makers and the general public. Here, we seek to improve
communication, provide common ground for discussion, and give
a comprehensive summary of the evidence.
How should a ‘climate fingerprint’ be defined? A straightforward
view typical of an economist would be to conclude that climate
change was important if it were principally responsible for a high
proportion of current biotic changes. By this criterion a climate
fingerprint appears weak. Most short-term local changes are not
caused by climate change but by land-use change and by natural
fluctuations in the abundance and distribution of species. This fact
has been used by non-biologists to argue that climate change is of
little importance to wild systems
10
. This approach, however, effec-
tively ignores small, systematic trends that may become important
in the longer term. Such underlying trends would be confounded
(and often swamped) by strong forces such as habitat loss. Biologists
have tended to concentrate on studies that minimize confounding
factors, searching for trends in relatively undisturbed systems and
then testing for significant associations with climate change. Econ-
omists have viewed this as biased (nonrandom exclusion of data)
whereas biologists view this as reducing non-climatic noise. Thus,
economists focus on total direct evidence and apply heavy time
discounting; biologists apply a ‘quality control’ filter to available
data, accept indirect (inferential) evidence and don’t apply time
discounting.
The test for a globally coherent climate fingerprint does not
require that any single species show a climate change impact with
100% certitude. Rather, it seeks some defined level of confidence in a
climate change signal on a global scale. Adopting the IPCC ‘levels of
confidence’
11
and applying the economists’ view of a fingerprint, we
would have “very high confidence” in a fingerprint if we estimated
that more than 95% of observed changes were principally caused
by climate change, “high confidence” between 95% and 67%,
“medium confidence” between 33% and 67%, and “low confidence”
below 33%. In contrast, the biologists’ confidence level comes from
the statistical probability that global biotic trends would match
climate change predictions purely by chance, coupled with support-
ing experimental results showing causal relationships between
climate and particular biological traits.
Here, we present quantitative estimates of the global biological
impacts of climate change.We search for a climate fingerprint in the
overall patterns, rather than critiquing each study individually.
Using the biologists’ approach, we synthesize a suite of correlational
studies on diverse taxa over many regions to ask whether natural
systems, in general, have responded to recent climate change.
Furthermore, we attempt a cross-fertilization by applying an
economists’ measure—the estimated proportion of observed
changes for which climate trends are the principal drivers—to
data sets chosen using biologists’ criteria. We call this a ‘global
coherence’ approach to the detection of climate change impacts.
First, we explore a biologists’ confidence assessment with two
types of analyses of observed change: statistical meta-analyses of
effect size in restricted data sets and more comprehensive categori-
cal analyses of the full literature. Second, we present a probabilistic
model that considers three variables: proportion of observations
matching climate change predictions, numbers of competing expla-
nations for each of those observations, and confidence in causal
articles
NATURE |VOL 421 | 2 JANUARY 2003 | www.nature.com/nature 37© 2003 Nature Publishing Group

attribution of each observation to climate change. These three
variables feature equally in a model that explores an economists’
‘confidence’ assessment. Finally, we explore diagnostic ‘sign-switch-
ing’ patterns that are predicted uniquely by climate change.
The evidence
A few studies indicate evolutionary responses of particular species
to climate change
12–14
, but the generality of evolutionary response
remains unknown. Here, we focus on phenological (timing) shifts,
range boundary shifts, and community studies on species abun-
dances (Table 1).
Meta-analyses
We developed databases suitable for meta-analysis
15
on two
phenomena: range-boundary changes and phenological shifts. To
control for positive publishing bias, we used only multi-species
studies that reported neutral and negative results as well as positive
(see Methods).
For range boundaries, suitable data spanned 99 species of birds
16
,
butterflies
17
and alpine herbs
18,19
(see Methods). The meta-analysis
showed that the range limits of species have moved on average 6.1
(^2.4) km per decade northward or m per decade upward,
significantly in the direction predicted by climate change (boot-
strapped 95% confidence interval of the mean (CI
mean
) ¼ 1.3–
10.9 kmm
21
per decade; one-sample t-test, degrees of freedom
(d.f.) ¼ 98, t ¼ 2.52, P ¼ 0.013; Table 2).
For phenologies, suitable data were reported for herbs
20–23
,
shrubs
20–25
, trees
20,23–25
, birds
20,21
, butterflies
26
and amphibians
27,28
,
a total of 172 species (see Methods). There was a mean shift towards
earlier spring timing of 2.3 days per decade, with a bootstrapped
95% CI of 1.7–3.2 days advancement per decade (significant at
P , 0.05).
Categorical analyses
The remaining studies were not included in the meta-analyses,
either because they were on single species or because they did not
present data in the raw form of x unit change per y time units per
species. These less-detailed datawere simplified into four categories:
changed in accord with or opposite to climate change predictions,
changed in some other fashion or stable (see Methods).
As with previous studies
17
, analyses ignore species classified as
‘stable’. This category does not represent a single result, as apparent
stability could arise from a diversity of situations
17
such as: 1) the
phenology, abundance or distribution of the species is not driven by
climatic factors; 2) the species is actually changing, but poor data
resolution could not detect small changes; and 3) the phenology,
abundance or distribution of the species is driven by climatic
factors, but fails to respond to current climate change. Such failure
could stem from anthropogenic barriers to dispersal (habitat
fragmentation) or from a lag in response time. Lags are expected
when limited dispersal capabilities retard poleward/upward colo-
nization
29
, or when a necessary resource has slower response time
than the focal species
17
.
Phenological shifts. We quantitatively assessed 677 species
reported in the literature (Table 1). Over a time period range of
16–132 years (median 45 yrs), 27% showed no trends in phenolo-
gies, 9% showed trends towards delayed spring events, whereas the
remaining 62% showed trends towards spring advancement.
Observed trends include earlier frog breeding
27,28
, bird nesting
30–
32
, first flowering
20–25
, tree budburst
23–25
, and arrival of migrant birds
and butterflies
20,21,26,33
(Table 1). Shifts in phenologies that have
occurred are overwhelmingly (87%) in the direction expected from
climate change (P , 0.1 £ 10
212
; Table 2).
Distribution/abundance shifts. In a quantitative assessment cover-
ing.1,046 species, we were able to categorize 893 species, functional
Table 1 Summary of data studying phenological and distributional changes of wild species
Taxon Ref. number Total no. of species
(or species groups)
Spatial
scale Time scale
(range years)
Change in direction
predicted (n)
Change opposite
to prediction (n)
Stable
(n)
No prediction
(n)
LRC
...................................................................................................................................................................................................................................................................................................................................................................
Phenological changes
Woody plants 20,23,24*,25* n ¼ 38 sp 2 1 35–132 30 1 7 –
Herbaceous plants 20,21* n ¼ 38 sp 1 1 63–132 12 – 26 –
Mixed plants 22* n ¼ 385 sp 1 46 279 46 60 –
Birds 20,21*,30,31,32,33 n ¼ 168 sp 2 3 1 21–132 78 14 76 –
Insects 26 n ¼ 35 sp 13 – 2
Amphibians 27,28 n ¼ 12 sp 2 16–99 9 – 3 –
Fish 20 n ¼ 2sp 1 132 2 – – –
...................................................................................................................................................................................................................................................................................................................................................................
Distribution/abundance changes
Tree lines 54,55,56* n ¼ 4spþ 5 grps 2 1 70–1,000 3 sp þ 5grps – 1 –
Herbs and shrubs 18,19,41*,42* n . 66 sp, 15 detailed 3 28–80 13 2 – –
Lichens 36 4 biogeographic grps (n ¼ 329 sp) 1 22 43 9 113 164
Birds 8* n ¼ 3sp 50 – – –
16,57* N sp (n ¼ 46 sp) 2 20–36 13 15 18
Ssp(n ¼ 73 sp) 2 20–36 36 16 21 6
43* Low elevation (.91 sp) 1 20 71 11 9 –
High elevation (.96 sp) 1 20 37 27 32 –
Mammals 37 n ¼ 2sp 5 2 – –
Insects 17,49* n ¼ 36 sp 1 1 98–137 23 2 10 1
17 N boundaries (n ¼ 52 sp) 4 1 7 –
Sboundaries (n ¼ 40 sp) 198 10 228
Reptiles and amphibians 43* n ¼ 7sp 17 6 –1–
Fish 39 4 biogeographic grps (n ¼ 83 sp) 1 – 2grps – 1grp 1grp
40* N sp (n . 1 sp) 70 .1 –
Ssp(n . 1 sp) 1 . –
Marine invertebrates 34*,40* N sp (n . 21) 1 1 66–70 .19 2 – .1 sp not classified
Ssp(n . 21) 1 1 66–70 .20 1 –
Cosmopolitan sp (n ¼ 28 sp) – – 28
Marine zooplankton 40* Cold water (n . 10 sp) 170 .10 – – .8 sp not classified
Warm water (n . 14 sp) .14 – –
35 6 biogeographic grps (n $ 36 sp) 1 39 6 grps – – –
...................................................................................................................................................................................................................................................................................................................................................................
N, species with generally northerly distributions (boreal/arctic); S, species with generally southerly distributions (temperate); L, local; R, regional (a substantial part of a species distribution; usually along a
single range edge); C, continental (most or the whole of a species distribution). No prediction indicates that a changemay have been detected, but the changewas orthogonal to global warming predictions,
was confounded by non-climatic factors, or there is insufficient theoretical basis for predicting how species or system would change with climate change.
*Study partially controlled for non-climatic human influences (for example, land-use change). Studies that were highly confoundedwith non-climatic factors were excluded. (See Supplementary Information
for details of species classification.)
articles
NATURE |VOL 421 | 2 JANUARY 2003 | www.nature.com/nature38 © 2003 Nature Publishing Group