\makeatletter
\@ifundefined{HCode}
{\documentclass[screen,CRGEOS,Unicode,biblatex,published]{cedram}
\addbibresource{crgeos20240844.bib}
\newenvironment{noXML}{}{}
\let\citep\parencite
\let\citet\textcite
\def\xcitealp#1#2{\citeauthor{#1}, \citelink{#1}{#2}}
\def\citealt#1#2{\citeauthor{#1}, \citelink{#1}{#2}}
\newcommand*{\citelink}[2]{\hyperlink{cite.\therefsection @#1}{#2}}
\def\defcitealias#1#2{}
\let\citetalias\textcite
\let\citepalias\parencite
\def\thead{\noalign{\relax}\hline}
\def\endthead{\noalign{\relax}\hline}
\def\tbody{\noalign{\relax}}
\def\tabnote#1{\vskip4pt\parbox{.95\linewidth}{#1}}
\def\tsup#1{$^{{#1}}$}
\def\tsub#1{$_{{#1}}$}
\def\ndash{\text{--}}
\def\hyphen{\text{-}}
\def\og{\guillemotleft}
\def\fg{\guillemotright}
\def\0{\phantom{0}}
\makeatletter
\g@addto@macro{\UrlBreaks}{\UrlOrds}
\gappto{\UrlBreaks}{\UrlOrds}
\RequirePackage{etoolbox}
\usepackage[most]{tcolorbox}
\usepackage{caption}
\usepackage{upgreek}
\def\inlinefig#1{\includegraphics{#1}}
\def\jobid{crgeos20240844}
%\graphicspath{{/tmp/\jobid_figs/web/}}
\graphicspath{{./figures/}}
\skip\footins=13pt
\def\xmorerows#1#2{#2}
\newcounter{runlevel}
\usepackage{multirow}
\def\nrow#1{\@tempcnta #1\relax%
\advance\@tempcnta by 1\relax%
\xdef\lenrow{\the\@tempcnta}}
\def\morerows#1#2{\nrow{#1}\multirow{\lenrow}{*}{#2}}
\let\MakeYrStrItalic\relax
\def\refinput#1{}
\def\Custref#1#2{\hyperlink{#2}{#1}}
\def\enlabel#1{\hypertarget{#1}{}}
\setlength{\fboxsep}{1em}
\newlength\boxedlength
\boxedlength=\textwidth
\advance\boxedlength-2em
\advance\boxedlength-2\fboxrule
\newsavebox{\mybox}
\newenvironment{frtextbox}[1]
{\begin{tcolorbox}[top=10pt,bottom=10pt,leftupper=10pt,rightupper=10pt,toptitle=11mm,breakable, size=fbox, 
boxrule=.6pt, pad at break*=1mm, arc=-1pt,titlerule=-1pt,boxsep=8pt, colback=white,colbacktitle=white,coltitle=black,  
title={\large\bf #1}%,code=\vskip -2\baselineskip
]}
{\end{tcolorbox}}
\newenvironment{Figure}{\begin{center}}{\end{center}}
\def\Caption#1{\captionof{figure}{#1}}
\let\ubreak\break
\csdef{Seqnsplit}{\\}
\def\botline{\\\hline}
\def\back#1{}
\newcommand*{\hyperlinkcite}[1]{\hyper@link{cite}{cite.#1}}
\DOI{10.5802/crgeos.303}
\datereceived{2024-10-13}
\daterevised{2025-06-30}
\dateaccepted{2025-07-28}
\ItHasTeXPublished
}
{
\PassOptionsToPackage{authoryear}{natbib}
\documentclass[crgeos]{article}
\usepackage[T1]{fontenc}
\def\enlabel#1{\label{#1}}
\def\CDRdoi{10.5802/crgeos.303}
\def\citelink#1#2{\citeyear{#1}}
\def\xcitealp#1#2{\citealp{#1}}
\def\xmorerows#1#2{\morerows{#1}{#2}}
\let\ubreak\relax
\makeatletter
\def\hyperlinkcite#1#2{\citeyear{#1}}
\newenvironment{Figure}{\begin{figure}}{\end{figure}}
\def\Caption#1{\caption{#1}}
}
\makeatother

\usepackage{upgreek}

\dateposted{2025-11-19}
\begin{document}

\begin{noXML}

\CDRsetmeta{articletype}{review}

\TopicFR{\'Ecosyst\`emes terrestres et aquatiques}
\TopicEN{Terrestrial and aquatic ecosystems}
\TopicFR{D\'eveloppement durable, am\'enagement}
\TopicEN{Sustainable development, landscaping}

\title{Ecosystem transformation by climate change and its consequences
for humans}

\alttitle{Transformation des \'{e}cosyst\`{e}mes par le changement
climatique et ses cons\'{e}quences pour les humains}

\author{\firstname{Sandra} \lastname{Lavorel}\CDRorcid{0000-0002-7300-2811}}
\address{Laboratoire d'Ecologie Alpine, Universit\'{e} 
Grenoble Alpes -- Universit\'{e} Savoie Mont Blanc, CNRS, CS 40700, 38058
Grenoble Cedex 9, France}
\email{sandra.lavorel@univ-grenoble-alpes.fr}

\keywords{\kwd{Climate extremes}
\kwd{Ecosystem transformation}
\kwd{Human nature-based adaptation}}

\altkeywords{\kwd{Extr\^{e}mes climatiques}
\kwd{Transformation des \'{e}cosyst\`{e}mes}
\kwd{Adaptation humaine fond\'{e}e sur la nature}}

\editornote{Article submitted by invitation}

\alteditornote{Article soumis sur invitation}

%\thanks{}

\begin{abstract}
The global climate and biodiversity crises are intimately linked.
Climate extremes and their combinations with other global change
factors are critical, yet uncertain drivers of changes in ecosystem
biodiversity and functioning. Here, I summarise the diverse dynamics of
ecosystems in response to climate extremes and their mechanisms. I then
highlight the consequences of these responses for humans, and how
changing biodiversity dynamics can be a resource for human adaptation
to global environmental crises. Complex systems theory underpins key
mechanisms of multiple types of ecosystem transformation: nonlinear
responses and hysteresis, the interplay of fast and slow variables and
the role of synergistic effects, cascading impacts and feedbacks. These
dynamics are underpinned by biotic mechanisms involving biodiversity at
all levels of organisation. Evidence on ecosystem transformation in
response to climate extremes highlights that beyond their direct
effects, climate extremes exacerbate other global change factors of the
biodiversity crisis. Then, by altering ecosystem structure,
biodiversity and functioning, climate change impacts the whole range of
nature's contributions to people, and decreases functions
underpinning adaptation and mitigation. There is increasing scientific
and operational evidence that ecosystems in good condition are however
critical to adaptation to climate change. Nature's contributions to
adaptation comprise climate change and risk mitigation, the resilience
and transformative ability of biodiverse and spatially connected
ecosystems, along with the social construction of new values for
transformed ecosystems. Processes of nature-based adaptation require a
transformation in the production and sharing of knowledge of relevant
ecological processes, in values for nature and in social\break relations.
\end{abstract}

\begin{altabstract} 
Les crises mondiales du climat et de la biodiversit\'{e} sont
intimement li\'{e}es. Les extr\^{e}mes climatiques et leurs
combinaisons avec d'autres facteurs de changement global constituent
des moteurs critiques, mais incertains, des modifications de la
biodiversit\'{e} et du fonctionnement des \'{e}cosyst\`{e}mes. Je
r\'{e}sume ici la diversit\'{e} des dynamiques des \'{e}cosyst\`{e}mes
en r\'{e}ponse aux extr\^{e}mes climatiques et leurs m\'{e}canismes. Je
mets ensuite en \'{e}vidence les cons\'{e}quences de ces r\'{e}ponses
pour les humains, ainsi que la mani\`{e}re dont les dynamiques
changeantes de la biodiversit\'{e} peuvent constituer une ressource
pour l'adaptation humaine aux crises environnementales mondiales. La
th\'{e}orie des syst\`{e}mes complexes sous-tend les m\'{e}canismes
cl\'{e}s de multiples types de transformations des \'{e}cosyst\`{e}mes
: r\'{e}ponses non lin\'{e}aires et hyst\'{e}r\'{e}sis, interactions
entre variables rapides et lentes, r\^{o}le des effets synergiques,
impacts en cascade et r\'{e}troactions. Ces dynamiques reposent sur des
m\'{e}canismes biotiques impliquant la biodiversit\'{e} \`{a} tous les
niveaux d'organisation. Les donn\'{e}es disponibles sur la
transformation des \'{e}cosyst\`{e}mes en r\'{e}ponse aux extr\^{e}mes
climatiques soulignent que, au-del\`{a} de leurs effets directs, les
extr\^{e}mes climatiques exacerbent d'autres facteurs du changement
global \`{a} l'origine de la crise de la biodiversit\'{e}. En
modifiant la structure, la biodiversit\'{e} et le fonctionnement des
\'{e}cosyst\`{e}mes, le changement climatique affecte ensuite
l'ensemble des contributions de la nature aux soci\'{e}t\'{e}s
humaines et r\'{e}duit les fonctions soutenant l'adaptation et
l'att\'{e}nuation. Il existe cependant de plus en plus de preuves
scientifiques et op\'{e}rationnelles indiquant que des
\'{e}cosyst\`{e}mes en bon \'{e}tat sont essentiels \`{a}
l'adaptation au changement climatique. Les contributions de la nature
\`{a} l'adaptation comprennent l'att\'{e}nuation du changement
climatique et des risques, la r\'{e}silience et la capacit\'{e} de
transformation des \'{e}cosyst\`{e}mes riches en biodiversit\'{e} et
spatialement connect\'{e}s, ainsi que la construction sociale de
nouvelles valeurs pour des \'{e}cosyst\`{e}mes transform\'{e}s. Les
processus d'adaptation fond\'{e}s sur la nature n\'{e}cessitent une
transformation dans la production et le partage des connaissances
relatives aux processus \'{e}cologiques, dans les valeurs
attribu\'{e}es \`{a} la nature et dans les relations sociales.
\end{altabstract}

\maketitle

\vspace*{1.3pc}

\twocolumngrid

\end{noXML}

\defcitealias{Wiens24}{ibid.}
\defcitealias{Vitasse21}{ibid.}

\section{Introduction}\label{sec1}

The global climate and biodiversity crises are intimately linked
through their shared societal causes and their dynamics
\citep{IPCC22,Portner21}. First, climate change already impacts
ecosystems and their biodiversity and is expected to become a major
driver of biodiversity change along with direct human use. Secondly,
these impacts will limit the ability of the biosphere to contribute to
\mbox{climate} mitigation and potentially feed forward to climate change
through greenhouse gas emissions from disturbed ecosystems (e.g.\ fires,
heat effects on peatland, melting permafrost) and alteration of
terrestrial and oceanic surface properties. Furthermore, some climate
mitigation and adaptation actions hold the potential to worsen harm to
ecosystems and biodiversity. Hence, the interlinkages between the two
crises are asymmetric, with climate change potentially worsening the
degradation of biodiversity.

{\advance\baselineskip.3pt

The drivers of biodiversity changes are well established globally
\citep{Diaz19}. Land and sea transformation for human use, habitat
degradation and direct use of species are currently the dominant
drivers, cumulating 80\% of impacts on species \citep{IPBES22,WWF14}.
Climate change ranks as the third current global driver of biodiversity
change, but is set to have increasing impacts under all future
greenhouse gas emission scenarios  \citep{Diaz19,IPCC22}, although
expected to be significantly less under more ambitious options like the
1.5~\textdegree C target \citep{IPCC18}. These impacts result from
climate trends including increasing temperatures, changed precipitation
regimes, sea level rise and melting of polar and mountain ice, along
with changes in natural disturbance regimes (e.g.\ storms, fire) and in
the frequency and intensity of climate extremes \citep{IPCC12,IPCC22}.
In this context, the expected increases in climate variability and
extreme events in particular heat waves, drought and extreme
precipitations \citep{IPCC21} are critical.

These direct effects of climate changes on biodiversity are observed on
multiple facets of ecosystem structure and function \citep{Lavorel17}.

First, changes in climate parameters directly affect the physiology and
behaviour of all organisms, through either increase (temperature) or
decrease (water and thereby nutrient availability) in fundamental
physical and chemical processes. Changes in phenology, the annual
sequence of an organism's developmental stages, are a prominent example
with documented advances in growth and reproduction across multiple
biota, and especially those with lower regulation capacity (e.g.\ 
insects, reptiles) or greater exposure (e.g.\ high arctic or alpine
snowbed plants and vertebrates) \citep{Vitasse21}. For example, across
the European Alps insect emergence has advanced by an average of 6 days
per decade since 1970, while singing and laying dates for birds have
not changed significantly \citepalias{Vitasse21}. Physiological and
behavioural change directly translate into changed survival and
reproduction, and in some cases critical threats to population or even
species persistence. For example, in Grisons, Switzerland, rock
ptarmigan (\textit{Lagopus muta}) has significantly shifted its entire
range towards by ${+}$33~m per decade over the past 30~years, while snow
hare (\textit{Lepus timidus}) shifted its minimum elevation by ${+}$33 m
per decade \citep{Schai21}. These elevational shifts were mostly
related to a reduced number of frost days. Because these direct
responses are heterogeneous across biota and species, important biotic
interactions can fail due to new mismatches (e.g.\ pollination,
predator--prey) or synchronies (e.g.\ diseases)
\citep{Abarca21,Carroll24}.\looseness=1

Secondly, climate change has already modified species distributions
polewards and towards higher altitudes. The current observed
distributional shifts are ca.\ 0.02~\textdegree C annually, which is three
to four times slower than current warming rates \citep{Wiens24}. As
with phenology, these adjustments are heterogeneous across biota and
species \citep[e.g.][]{Carroll24,Vitasse21}. Modelling of responses to
climate change scenarios estimates that on average 20 to 30\% of
species will be threatened by extinction by 2100 
\citep{Diaz19,IPCC22}. Specifically, most recent models based on
observed shifts in species bioclimatic niches estimate extinction rates
to average at 17\% \citep{Wiens24}. A recent modelling study based on
observed species responses at their upper thermal range with three
dispersal scenarios projected global extinction of
14\%--32\% species (potentially 3--6~million) across major
terrestrial and marine biota by 2070 under the intermediate climate
scenario RCP 4.5 \citepalias{Wiens24}. Overall these projections are
underpinned by large uncertainties relating especially to the lack of
consideration of specific physiological responses, phenology or of
biotic interactions that limit establishment into newly suitable
habitat, and how species dispersal abilities are incorporated
\citep{Higgins20,McMahon11,Morin09}. Furthermore, they are based on
average climate conditions and, with the exception of some
physiological models, unable to consider the effects of climate
variability and extremes and their consequences for natural disturbance
regimes \citep{Thonicke20,Turner20}.

Thirdly, changes in temperatures and precipitation directly impact
ecosystem functioning through their effects on organisms' physiological
activities (e.g.\ water and nutrient uptake and recycling),
biodisponibility of resources from water and soils or sediments, and
biochemical process rates. Most reported impacts include the
acceleration of carbon and water cycles from local to global scales,
which won't be detailed in this review. Importantly, ecosystem
functioning is also affected indirectly by climate change through its
impacts on the distribution of ecosystems from a global to regional
scale \citep{Sitch08} and on community functional composition from
local to global scales \citep{Bugmann22,Chang15,Sakschewski15}.
\looseness=1

Lastly, in most extreme cases climate change has already started to
transform ecosystem structure and functioning, sometimes referred to as
regime shifts, ecosystem collapse or tipping points
\citep{Berdugo20,Bergstrom21,Leadley14,Rocha15} 
(see Section~\ref{sec21}). Most
reported examples include arctic deglaciation and glacier margins, tree
colonisation of boreal tundra, rainforest to savanna transformation,
tropical shrubland to desert transformation or coral\break reefs.

There are many uncertainties about mechanisms underpinning all these
effects and their future projections which, among other limitations do
not account for climate extremes and climate tipping points. In this
context, this paper focuses on climate extremes and their combinations
with other global change factors as critical, yet uncertain drivers of
changes in ecosystem biodiversity and functioning. After briefly
summarising the rich theoretical basis, I review evidence on the
diverse dynamics of ecosystems in response to climate extremes and on
their mechanisms. I then present the consequences of these responses
for humans, and conversely how biodiversity and its responses to
climate-related changes can be a resource for human adaptation to
global environmental crises.

\vspace*{2pt}
\section{Conceptual and theoretical basis}\label{sec2}
\vspace*{2pt}

\subsection{Transformation of ecosystems and their biodiversity by
climate change and climate extremes}\label{sec21}

\vspace*{2pt}

Ecosystem transformation in response to climate change refers to a
common body of understanding from complex systems theory
\citep{Folke04}, and prolific theoretical work. This literature and
\mbox{associated} terminology address qualitative changes in ecosystems in
response to climate change. In the following I present the main
relevant terms and concepts, referring to \textit{transformation} as
their umbrella concept. Ecosystem transformation is defined as a change
in the set of variables that control the system's functioning,
including their physical structure and their biodiversity. Ecosystem
development at glacier forefronts is a well-known example of
transformation, shifting from mineral surfaces with minimal
biodiversity to multitrophic soil-vegetation-animal food webs
\citep{Ficetola21}. Transformation is often caused by changes in
regulating physiological, demographic or biogeochemical feedback loops
resulting in qualitatively different structure and function
\citep{Walker04}. Transformation contrasts with \textit{resilience},
where instead ecosystem structure, including biodiversity, and function
return to their baseline state through multiple regulating and
buffering mechanisms. Changes during transformation may operate at a
variety of interacting scales, and be observable through a variety of
relevant indicators. Multiple concepts and definitions of ecosystem
transformation share key mechanisms including: nonlinear responses and
hysteresis, the interplay of fast and slow variables and the role of
synergistic effects, cascading impacts and \unskip\break feedbacks.

Referring to complex systems dynamics, transformation is often referred
to as a \textit{regime shift} where the large, persistent
reorganisation of the structure and function results from the
reconfiguration of abiotic and biotic control variables and processes
\citep{Rocha15}. Regime shifts can be driven by relatively linear
processes like temperature-driven transitions from kelp forest to
seaweed meadows \citep{Wernberg16}, shrub expansion into boreal and
alpine tundra ecosystems \citep{Myers11}, and transitions from forest
to savanna in response to changing fire regimes and grazing
\citep{Leadley14,Turner20}.

In contrast, a \textit{tipping point} is qualitative change in
ecosystem structure, its biodiversity and/or its function in response
to a given value of a climatic variable considered as a
\textit{threshold}. To be considered as a tipping point, this change
needs to be nonlinear and hard to reverse (\textit{hysteresis}),
because it is maintained by positive feedbacks. A particular case of
ecological tipping point regards rate-tipping transformation, observed
in response to a given rate of climate change or \textit{climate
velocity}, not just to a certain climate value (threshold state)
\citep{Synodinos23}. The extinction of species unable to adapt locally
and/or to disperse and successfully establish at pace with changing
climate is the best known case for rate-tipping points, representing
priorities for conservation approaches such as translocation
\citep{Brito18,Butt21}. Large-scale, biome-level ecosystem
transformation has the potential for generating tipping points in
regional and global climate \citep{Leadley14,Lenton08}.

\textit{Ecosystem collapse} is an extreme case of ecosystem
transformation and regime shift with limited capacity to recover, where
``the ecosystem has lost key defining features and functions, and is
characterised by declining spatial extent, increased environmental
degradation, decreases in, or loss of, key species, disruption of
biotic processes, and ultimately loss of ecosystem functions'', usually
characterised by thresholds of transformation
\citep{Berdugo20,Bergstrom21}. While the speed and linearity of
collapse can vary across ecosystems, these transformations nearly
always involve combinations of multiple climate and human factors, each
of which act as press, i.e.\ continuous, or pulse, i.e.\ event-driven
disturbances \citep{Portner22}. For example, coral reef collapse is
driven by combined trend temperature changes, heatwaves and the
degradation of water quality through land use. Massive tree die-back
events and the subsequent transformation of forest or woodland into
degraded grassland combine long-term effects of increased temperature
and drought, with extreme drought and fire events, pest outbreaks on
already fragmented populations.

In practice, ecosystems can undergo multiple steps of transformation
over the long-term. The case of drylands systems illustrates how
transformation takes place as a series of quantitative and qualitative
changes. As aridity increases, abrupt decays in aboveground
productivity are first observed, \mbox{followed} by a loss of soil fertility
due to the combined lack of biomass input, water availability and
resulting disruption of biotic activity, and a final systemic collapse
with the loss of plant cover \citep{Berdugo20}.

The notion of ecosystem collapse is challenged by the conceptualisation
of transformative change. When transformation leads to ecosystems with
reassembled biotic communities with no current analogues and stable
processes maintained by reconfigured feedback loops, these are
considered as \textit{novel ecosystems} \citep{Ordonez16}. Post
glaciation cases of such ecosystem development have been well
documented \citep{Turner20}. Recent examples are emerging especially
when climate change favours the colonisation and expansion of exotic
species that alter biotic communities and induce self-reinforcing
feedbacks on disturbance regimes and/or biogeochemical cycles
\citep{Prober19}.\looseness=1

\subsection{Mechanisms underpinning ecosystem resilience and
transformation with climate change and climate extremes}\label{sec22}

Complex systems dynamics of ecosystem transformation in response to
climate trends and extremes are underpinned by a set of biotic
mechanisms relevant to understanding the role of different facets of
biodiversity. These have been supported by the combination of evidence
from theoretical modelling, field observation including long-term
physical, ecological and some cultural data series and experiments 
\citep{Henne18,Thonicke20}.\looseness=1


Stable ecosystems maintain their structural attributes and multiple
ecosystem functions such as productivity, carbon sequestration or
pollination in the face of inter-annual climate variability and
disruption by disturbances including climate extremes.
\textit{Ecosystem stability} can be assessed by referring to its
different components including resistance to change especially in
response to extreme events, and processes of recovery towards a
baseline (resilience) or an alternative state (transformation)
\citep{Falk19,Ingrisch18a,Oliver15}. Stability is maintained by a
multiplicity of interplaying mechanisms from individuals (physiological
and behavioural) to populations (demographic and evolutionary),
communities (biotic interactions and \mbox{dispersal}), ecosystems (e.g.\ 
biogeochemical) and landscapes (lateral flows of organisms, matter and
energy) that can buffer the effects of environmental variation
\citep{Felton17,Oliver15,Schlagel20}.

Consistent with resilience theory \citep{Walker04}, multiple lines of
evidence from experiments \citep{Isbell15} and long-term observations
at community \citep{Liang22} to continental level \citep{Oliveira22}
suggest that greater biodiversity stabilizes natural communities and
ecosystems. The long-running debate on the relationship between species
diversity and stability \citep{McCann00} is beyond the scope of this
paper, yet established mechanisms can be summarised as follows. First
greater biotic diversity, including genetic, phenotypic, species and
functional diversity, {allows} redundancy, the presence of alternative
biota for a same ecological function, and insurance, the
complementarity across biota with respect to preferred environmental
conditions (e.g.\ climate, resources, disturbances). Secondly, the
complexity of biotic interaction networks is expected to support these
two mechanisms, where intermediate interaction density and modularity
allow a trade-off between complementary local responses and propagation
between network elements \citep{Montoya06}. Thirdly, intermediate
physical connectivity is likewise expected to stabilize spatial
networks of communities and ecosystems through the benefits of
dispersal across landscape elements with different exposure to climate
and other disruptions \citep{Gravel16}. These mechanisms are combined
across spatial and temporal scales, governing trajectories of
biodiversity and ecosystem functioning \citep{Falk19,Schlagel20}.

Many of these mechanisms can be related to functional traits of biota
\citep{Bello21,Oliver15,Walker99}. Specifically, functional traits
underpin three interacting mechanisms of stability. First, the traits
of dominant (i.e.\ most abundant) species or genotypes determine many
ecosystem processes by their prevalent effects through their biomass
contribution, and hence resistance and recovery after perturbations.
Secondly, complementary trait values across species or genotypes,
broadly referred to as functional diversity, \mbox{allow} differing responses
to environmental variability and hence compensatory dynamics at the
ecosystem to landscape scale. Thirdly and conversely, similarity across
species and genotypes in structural, physiological, biochemical or
phenological characteristics affecting target ecosystem functions such
as biomass production and nutrient recycling, i.e.\ effect traits,
coupled with dissimilarity in their characteristics driving the effects
of climate and disturbances on their survival, growth and reproduction,
i.e.\ response traits, underpin redundancy \citep{Lavorel02}.
Ultimately, these three mechanisms operate concomitantly in specific
ecosystems through modifications by climate variability, climate trends
and extremes of the abundance distribution of trait values in biotic
communities \citep{Kohler17} (Box~\Custref{1}{box1}). The resulting trait value
distributions then either buffer or propagate environmental change
effects on ecosystem\break functions.

In addition to stability and resilience, which maintain ecosystem
structure and function over the long term, mechanisms of transformation
are essential when facing extreme climate change and/or novel climatic
and disturbance regimes. Transformability, the ability to transform, is
underpinned by specific mechanisms \citep{Lavorel15}. Response
diversity can mediate ecological transformation in several ways: (1)
through shifting contributions of different species groups, especially
across groups with differing ecological niches like seeder vs.\ 
resprouter trees under changing fire regimes; (2)~through response
diversity within keystone functional groups such as structuring tussock
grasses in grasslands and savannas, or coral vs.\ algae in tropical
reefs; and (3) through response diversity in hyperdiverse communities
like tropical rainforest. Landscape and seascape connectivity plays a
key role not only in resilience but also in the transformation of
fragmented systems through its effects on propagule flows that are
necessary for disturbance responses and for the migration of
climatically suitable species. Transformation depends on immigration
by, or dominance of, species with novel traits where previously
dominant traits have been lost. Novel ecosystems containing exotic
species, following extinction of dominant native species, are likely to
become commonplace under future climate\unskip\break \citep{Prober19}.

}

\pagebreak

\onecolumngrid
\bigskip
\unskip\noindent\begin{frtextbox}{Box 1---Functional dynamics in the transformation of
ecosystems \protect\citep[summarised from][]{Kohler17}.}
\enlabel{box1}
\mbox{}


\vspace*{1pc}

The dynamics of change in ecosystem biodiversity and functioning can be
assessed from the analysis of the functional structure of communities
(Figure~\ref{fig1}). 
\vspace*{2pc}

\begin{Figure}
\includegraphics{fig01}
\vspace*{6pt}
\Caption{\label{fig1}Abundance structure of a community. Dominant
species have a significantly greater abundance across all sampled
communities; sub-dominants have intermediate abundance which can
increase or decrease across communities; subordinates have a low
abundance across all communities (\protect\xcitealp{Grime98}{1998}). Each bar
represents a species, with colours representing different functional
groups based on traits for a function of interest.}
\vspace*{2pc}
\end{Figure}

{

\advance\baselineskip by .5pt

Ecosystem function is estimated from its quantitative relationship to
species traits and their abundances \citep{Lavorel11}. Natural
variation in space (multiple locations within the landscape or region)
or time (multiple years) of the ecosystem function is considered as the
stable range of variation, or normal operating range, within which the
ecosystem is considered to \textit{resist} environmental change.
Historical climate variability is compatible with the persistence of
species within each community, but drives changes in their relative
abundances within their respective dominance groups. The corresponding
range of variation of the ecosystem function, or community potential
operating range, is simulated for iterations of such community
reassembly, and considered as its \textit{resilience}. Extreme climate
events exceed the tolerance of some of the species from the community,
and open opportunities for colonisation by species from other
communities with a similar history (meta-community), leading to a new
community. The corresponding range of variation of the ecosystem
function, the meta-community operating range, is calculated for
iterations of such community reassembly, and considered as its
\textit{transformation}\break (Figure~\ref{fig2}).

}

\begin{Figure}
\includegraphics{fig02}
\Caption{\label{fig2}Variability of an ecosystem function according to
observed variability (Normal operating range---NOR), response of a
stable community to natural climate variability (Community operating
range---CPOR) and transformation by a climate extreme (Meta-community
operating range).}
\end{Figure}

\end{frtextbox}

\vspace*{17pt}
\twocolumngrid

\section{Evidence for nonlinear effects of climate extremes on ecosystems}\label{sec3.}

Section~\ref{sec2} showed how mechanisms of ecosystem transformation
are explained by complex systems theory and by the effects of
biodiversity on ecosystem functioning. Yet, we lack comprehensive
evidence on the range of transformations from direct effects of climate
extremes to more complex and impactful interactions. In line with
recent syntheses \citep{Turner20}, I hypothesise that ecosystem
transformation tends to unfold in these more complex situations. In the
following I summarise evidence for such nonlinear effects of climate
extremes on ecosystems.

\subsection{Observed effects of compound effects of climate extremes}\label{sec31}

Climate change is associated with the increasing frequency and
intensity of climate extremes, as well as with not only single but
sequences of extreme events \citep{IPCC12,IPCC22,Thonicke20}.
While observational, experimental and modelling evidence and
understanding of the effects of single extremes on ecosystems are
increasing and yet incomplete \unskip\break\citep{DeBoeck18,Frank15,Turner20},
combined effects of extremes with climate trends and compound extremes
are still a great source of uncertainty on future trajectories of
biodiversity and ecosystem functioning.

Rare experiments combining extreme events with modifications of
baseline climate show abrupt decreases in ecosystem functioning. For
example the combination of heatwaves with drought in temperate
grasslands has variable effects on plant growth and senescence
depending on plant traits, but result in both short- and medium-term
reductions in primary production \citep{Benot14,DeBoeck16,Zwicke13},
along with decreased carbon sequestration and increased greenhouse gas
emissions \citep{Fuchslueger16,Ingrisch18b,Karlowsky18,Knapp24}.  Such
effects result from physiological stress on photosynthesis, reduced
regulation by plant transpiration, and modified plant--soil interactions
through plant exudates and soil microbial community composition and
metabolism.

Even fewer experiments have addressed successive climate extremes,
showing shifts in responses from physiological resilience to
transformation of community composition to morphological, \unskip\break
\mbox{phenological}
and physiological avoidance and resistance strategies
\citep{Dreesen14,Knapp24}. Here, observation of recent events informs
about the cumulated impacts of successive climate extremes and their
abrupt effects on ecosystems. For~example, in central Europe the two
extreme summers of 2018 and 2019 with an exceptional drought and heat
waves showed variable, nonlinear effects depending on specific exposure
and ecosystems \citep{Bastos20}. Agricultural areas, and especially
grasslands, showed by 2019 exceptional reductions in plant cover and
activity resulting from lasting soil water reduction and cumulated
physiological effects. In contrast, forests showed dramatic senescence
only by 2019, due to delayed physiological effects. Yet, the synthesis
of observations across networks of long-term forest monitoring plots
shows that the structure of European temperate forests has been
resilient to past large and severe disturbances (windthrow, fire, bark
beetle outbreaks) and concurrent climate conditions \citep{Cerioni24}.
Still, long-term shifts in composition to early-successional species
may be promoted by shorter successions of events.

\subsection{Evidence for benefits of biodiversity for ecosystem
resilience to climate extremes}\label{sec32}

Experimental approaches provide mixed support for the benefits of
biodiversity for the resilience of ecosystem functioning to climate
extremes expected from theory. Across 46 grassland species richness
manipulation experiments experiencing inter-annual climate variability,
benefits of species richness to aboveground net primary production were
largest during extreme wet or extreme dry years, supporting the
insurance mechanism of greater likelihood of drought or wet adapted
species in more diverse mixtures \citep{Isbell15}. Yet, in a 14-year
combined grassland biodiversity and climate manipulation experiment,
species richness benefited resistance of aboveground net primary
production to extreme wet or dry treatments no more than to moderate
wet or dry treatments \citep{Hossain22}. However, species richness
decreased resilience to moderate or extreme dry treatments, and had no
effect on resilience to moderate or extreme wet conditions. A
meta-analysis comprising 28 experiments on high vs.\ low biodiversity
plant communities exposed to extreme drought (19 experiments of which
17 in grasslands and 2 in forests) or extreme precipitation (9
experiments in grasslands) concluded that while higher biodiversity
promoted higher aboveground biomass production and these extremes
significantly reduced that production, higher biodiversity did
significantly not increase either resistance or recovery in the face of
these simulated extremes \citep{DeBoeck18}. These conclusions likely
reflect the low relevance of species richness {per se} to
community responses to extremes, where instead responses are likely
driven by functional traits \citep{Bello21}. The relevant variation in
functional traits within communities is largely driven by interspecific
variation and community assembly mechanisms in response to
environmental drivers and biogeographic, site and evolutionary history
\citep{Cavender16,Kraft15}. Moreover, intraspecific trait variation,
reflecting genetic diversity and phenotypic responses, can also be a
significant contributor \citep{Sanderson23}. Such plant but also
microbial trait driven responses of ecosystem functioning are evidenced
in naturally species-rich grasslands where dominant species traits
determine both resistance and resilience to experimental climate
extremes  \citep{Karlowsky18,Piton20,Schuchardt23}. Also consistent
with trait-based mechanisms, biodiversity benefits to ecosystem
functioning in experimental communities are greatest when considering
multiple combined environmental changes, reflecting functional
complementarity through functional divergence \citep{Isbell11}. 

\subsection{Indirect effects of climate extremes on ecosystem responses
to other global change factors}\label{sec33}

Beyond their direct effects, climate extremes exacerbate the effects of
other global change factors responsible for contemporary modifications
of biodiversity: the human use of ecosystems and species, invasive
alien species and pollution  \citep{Diaz19}.

First, climate extremes can further undermine the persistence of
populations that are already threatened by land or sea use or direct
exploitation \citep{Foden19}. Conversely, landscape and seascape
fragmentation by human use reduces species movement abilities after
extreme storms or fires, thereby increasing local to regional
extinction risks and ecosystem transformation
\citep{Bergstrom21,Turner20}.

Secondly, climate change, but also climate extremes that destroy local
populations of native species favour the spread of invasive alien
species \citep{Diez12}. Yet, while mean climate change increases
favourable habitats, extreme heat, drought or precipitation may reduce
the spread of invasive species as shown for example in the case of six
vertebrates for the Iberian Peninsula \citep{Baquero21} or aquatic
invertebrates \citep{McDowell17}. By transforming ecosystem structure
and key attributes like biomass or water regimes, invasive species can
amplify the effects of climate extremes. This is particularly
significant through changes in disturbance regimes. For example, in
Australia the continental scale expansion of introduced buffel grass
(\textit{Cenchrus ciliaris}) into semi-arid and arid rangelands
strongly increases fuel quantity and continuity, fostering larger and
more intense fires \citep{Schlesinger13}. In the same way, invasive
species colonising riparian corridors and floodplains alter
hydrological regimes during floods through changing biomass and
sediment accumulation \citep{Kiss19}.

Thirdly, climate extremes combine with other disturbances, like
herbivory and pests to transform ecosystem structure and biodiversity.
Modelling suggests that in European forests, extensive browsing by
extant ungulate populations is likely to critically limit spontaneous
adaptation of forest composition to climate scenarios through setting
back the recruitment of warm-adapted species like oaks (\textit{Quercus
robur} and \textit{Q. petraea}) \citep{Dobor24}. While knowledge is
still incomplete on how and when extreme drought favours tree mortality
through subsequent bark beetle outbreaks \citep{Jaime24}, those forests
submitted to their combination can catalyse vulnerability to subsequent
disturbances like windthrow and fire \citep{Senf21,Turner20}.\looseness=-1
\vspace*{-5pt}

\subsection{When resilience is breached: systemic effects, cascades and
interactions}\label{sec34}
\vspace*{-5pt}

Together these cases support the emergence of novel disturbance regimes
under climate change and climate extremes that may challenge ecological
resilience and promote novel ecosystems \citep{Turner23}. These
transformations can in turn propagate to further regime shifts through
cascading effects. The analysis of 30 characteristic regime shifts at a
global scale revealed that these are most frequently driven by shared
drivers, of which climate (especially for aquatic and marine systems)
and agriculture or land cover conversions were \mbox{prevalent}
\citep{Rocha18}. Effects of shared drivers are then compounded by
domino effects across adjacent or connected ecosystems at ecotones
(e.g.\ coastal systems including mangroves or tidal systems), or
emerging feedbacks in space and/or time through changed climate
regimes, fire, agriculture or urbanisation. In particular, the
conversion of grassland to desert or arid systems can impact
precipitation regimes and accelerate aridification. Likewise, poleward
or altitudinal migration of tree lines in boreal and mountain regions
can enhance the increase in temperature due to large increases in
albedo.

\section{Implications for humans and nature-based adaptation}\label{sec4}

By altering ecosystem structure, biodiversity and functioning, climate
change impacts the whole range of nature's contributions to people
\citep{Runting17}. Specifically increasing climate extremes have large
impacts on all material benefits of food, raw materials (including
wood) and freshwater, on local to global climate regulation and on
mitigation of natural risks \citep{Thonicke20}. Furthermore, climate
change decreases these very functions that are critical to climate
adaptation and mitigation. This includes the weakening of terrestrial
carbon sinks \citep{Lauerwald24,Seidl14} and the degradation of
ecosystems with risk protection benefits like forests regulating
erosion and floods at the head of catchments or in flood-prone valleys
\citep{Pramova12}, protection forests \citep{Stritih24} or wetlands
\citep{Xi21}.

Ecosystems in good condition are however critical to social adaptation
to climate change \citep{Cohen16,Colloff20,Lavorel15}. Such
\textit{nature's contributions to adaptation} comprise not only climate
change and risk mitigation, but also the resilience and transformative
ability of genetically, taxonomically or functionally diverse and
spatially connected ecosystems (see Section~\ref{sec21}), along with
the social construction of new values for transformed ecosystems. As an
example, emergent ecosystems at glacier forefronts reduce risks from
landslides and floods, sequester carbon in vegetation and soil, provide
new grazing grounds and forest products and offer new cultural
contributions to art, education or research, as well as spiritual
connections \citep{Khedim21,Zimmer22}. Multifunctionality is typical of
nature-based solutions, which often deliver multiple co-benefits beyond
their primary objectives of climate mitigation or risk mitigation
\citep{Chausson20,Gonzalez25,Lavorel20}.

Furthermore, the concept of nature's contributions to adaptation
recognises that human connections to nature, capitals, and social
processes are necessary for people to implement and benefit from these
contributions \citep{Lavorel20,Locatelli24}. For example, in the
Grenoble region (France) nature-based solutions of ecosystem
conservation, restoration and management are central to policy and
planning documents. They comprise among others: increasing urban
vegetation, restoration of wetlands and floodplains, restoring
landscape connectivity to support species migration, agroforestry or
adaptive management of forests. Their implementation and scaling
requires a combination of levers, including knowledge production and
sharing, changing and supporting nature's values and perception, local
governance, supportive policies, financial support, and the
reinforcement of the landscape planning culture  
\citep{Bruley}.\looseness=-1
\vspace*{-3pt}

\section{Conclusion}\label{sec5}
\vspace*{-3pt}

Climate extremes, often when combined with climate trends and other
disturbances and global change factors can transform the structure,
biodiversity and functioning of ecosystems. Some of these changes are
qualitative, and for some abrupt, contributing to the global and local
loss of biodiversity, and to critical changes in ecosystem functions
and feedbacks to the global climate. Such transformations can imperil
human livelihoods and societies. Yet, biodiversity provides an
insurance for ecosystems and humans in the face of these
transformations. Processes of social adaptation enable the mobilisation
of biodiversity for adaptation, requiring in turn a transformation in
the production and sharing of knowledge of relevant ecological
processes, in values for nature and in social relations.
\vspace*{-5pt}

\section*{Acknowledgments}
\vspace*{-3pt}

This paper is a contribution from the Transformative Adaptation
Research Alliance (TARA, \url{https://research.csiro.au/tara/}), an
international network of researchers and practitioners dedicated to the
development and implementation of novel approaches to transformative
adaptation to global change. This paper contributes to the Programme on
Ecosystem Change and Society (PECS, a Future Earth core project) and
its working group on ``Nature-based transformations: Evolving
human--nature interactions under changing climate''
(\url{https://pecs-science.org/nature-based-transformations/}).

\section*{Declaration of interests}

The author does not work for, advise, own shares in, or receive funds
from any organization that could benefit from this article, and has
declared no affiliations other than their research organization.

\back{}

\printbibliography
\refinput{crgeos20240844-reference.tex}

\end{document}