Climate change vulnerability of Arctic char across Scandinavia

Abstract

Climate change is anticipated to cause species to shift their ranges upward and poleward, yet space for tracking suitable habitat conditions may be limited for range‐restricted species at the highest elevations and latitudes of the globe. Consequently, range‐restricted species inhabiting Arctic freshwater ecosystems, where global warming is most pronounced, face the challenge of coping with changing abiotic and biotic conditions or risk extinction. Here, we use an extensive fish community and environmental dataset for 1762 lakes sampled across Scandinavia (mid‐1990s) to evaluate the climate vulnerability of Arctic char (Salvelinus alpinus), the world's most cold‐adapted and northernly distributed freshwater fish. Machine learning models show that abiotic and biotic factors strongly predict the occurrence of Arctic char across the region with an overall accuracy of 89 percent. Arctic char is less likely to occur in lakes with warm summer temperatures, high dissolved organic carbon levels (i.e., browning), and presence of northern pike (Esox lucius). Importantly, climate warming impacts are moderated by habitat (i.e., lake area) and amplified by the presence of competitors and/or predators (i.e., northern pike). Climate warming projections under the RCP8.5 emission scenario indicate that 81% of extant populations are at high risk of extirpation by 2080. Highly vulnerable populations occur across their range, particularly near the southern range limit and at lower elevations, with potential refugia found in some mountainous and coastal regions. Our findings highlight that range shifts may give way to range contractions for this cold‐water specialist, indicating the need for pro‐active conservation and mitigation efforts to avoid the loss of Arctic freshwater biodiversity.

Full text

Glob Change Biol. 2024;30:e17387.  
|
1 of 10
https://doi.org/10.1111/gcb.17387
wileyonlinelibrary.com/journal/gcb
Received:4March2024
|
Revised:8May2024
|
Accepted:21May2024
DOI : 10.1111/gcb .17387
RESEARCH ARTICLE
Climate change vulnerability of Arctic char across Scandinavia
Clint C. Muhlfeld1 | Timothy J. Cline2 | Anders G. Finstad3,4 | Dag O. Hessen5 |
Sam Perrin4 | Jens Thaulow6 | Diane Whited7 | Leif Asbjørn Vøllestad5
This is an op en access arti cle under the ter ms of the CreativeCommonsAttribution License, which permits use, distribution and reproduction in any medium,
provide d the original wor k is properly cited.
©2024TheAuthor(s).Global Change Biology published by J ohn Wil ey & Sons Ltd . This ar ticle h as been contributed to by U. S. Gover nment employee s and the ir
workisinthepublicdomainintheUSA .
1U.S. Geo logical Sur vey, Northern Rocky
Mountain Science Center, West Glacier,
Montana,USA
2Depar tment of Ecology, Montana St ate
University,Bozeman,Montana,USA
3Depar tment of Natura l Histor y, NTNU
University Museum, Norwegian University
of Science and Technology, Trondheim,
Norway
4Gjærevoll Center for Biodiversity
ForesightAnalyses,NorwegianUniversity
of Science and Technology, Trondheim,
Norway
5Department of Biosciences, University of
Oslo, Oslo, Norway
6Formerly Employed at Norwegian
Instit ute for Water Research, Oslo,
Norway
7Flathead Lake Biological Station,
University of Mo ntana , Polson , Monta na,
USA
Correspondence
Clint C. Muhlfeld, U.S. Geological Survey,
Northern Rocky Mountain Science Center,
Glacie r National Park , 38 Mather Drive ,
WestGlacier,MT59936USA .
Email: cmuhlfeld@usgs.gov
Funding information
U.S. Ge ologic al Sur vey, Nort hern Rocky
Mountain Science Center; University of
Oslo, Department of Bio sciences; U.S.
Fulbright Specialist Program
Abstract
Climate change is anticipated to cause species to shift their ranges upward and pole-
ward, yet space for tracking suitable habitat conditions may be limited for range-
restricted species at the highest elevations and latitudes of the globe. Consequently,
range-restrictedspeciesinhabitingArcticfreshwaterecosystems,whereglobalwarm-
ing is most pronounced, face the challenge of coping with changing abiotic and biotic
conditions or risk extinction. Here, we use an extensive fish community and environ-
mentaldatasetfor1762lakessampledacrossScandinavia(mid-1990s)toevaluatethe
climatevulnerabilityofArcticchar(Salvelinus alpinus),theworld'smostcold-adapted
and northernly distributed freshwater fish. Machine learning models show that abi-
oticandbioticfactorsstronglypredicttheoccurrenceofArcticcharacrosstheregion
withanoverallaccuracyof89percent.Arcticcharislesslikelytooccurinlakeswith
warm summer temperatures, high dissolved organic carbon levels (i.e., browning),
andpresenceofnorthernpike(Esox lucius).Importantly,climatewarmingimpactsare
moderatedbyhabitat(i.e., lakearea) andamplifiedby the presence ofcompetitors
and/orpredators(i.e.,northernpike).ClimatewarmingprojectionsundertheRCP8.5
emission scenario indicate that 81% of extant populations are at high risk of extir-
pation by 2080. Highly vulnerable populations occur across their range, particularly
near the southern range limit and at lower elevations, with potential refugia found in
some mountainous and coastal regions. Our findings highlight that range shifts may
give way to range contractions for this cold- water specialist, indicating the need for
pro-activeconservationandmitigationeffortstoavoidthelossofArcticfreshwater
biodiversity.
KEYWORDS
Arcticchar,Arcticfreshwaterecosystems,climatevulnerability,extinctionrisk,range
contractions, Scandinavia

2 of 10
|
MUHLFELD et al.
1 | INTRODUC TION
Arcticfreshwaterecosystemsareexperiencingprofoundenviron-
mental changes due to climate change and multiple anthropogenic
stressors(Heino etal., 2009; Li et al., 2022; Sala et al., 2000; Su
et al., 2021).TheArcticregion iswarmingfourtimesfaster than
the global average, altering water temperature, hydrological re-
gimes, water qualit y, and food webs within freshwater ecosystems
(Fengetal.,2021; Saros et al., 2023; Wrona et al., 2016).Astem-
peratures increase and exceed thermal limits, many cold- water
species are experiencing declines in distribution and abundance,
while cool- and warm- water species are expanding into higher el-
evations and latitudes, potentially displacing cold- adapted species
(Barbarossaetal.,2021; Reist, Wrona, Prowse, Power, Dempson,
King, et al., 2006).H uman activ itie s ,su cha slan d-usec hang es,p ol-
lution, and introduction and spread of invasive species, are further
acceleratingfreshwaterbiodiversityloss(Perrinetal.,2021; Reid
et al., 2019).Climatechangeandlandscapealterationsareincreas-
ingprecipitation andforest cover (Heinoetal.,2009),leadingto
permafrostthaw(Vonketal.,2015)andelevateddissolvedorganic
carbonrunoff,resultinginthe“browning”(Crapartetal.,2023; de
Wit et al., 2016; Finstad et al., 2016; Larsen et al., 2 011)anddis-
ruption of f reshwater ecosy stems (Finsta d et al., 2014; Hayden
et al., 2019; Karlsson et al., 2009).Thesecombined stressors are
posingsignificant threats toArcticfreshwaterspecies and biodi-
versity, warranting broad- scale research to understand and mit-
igate their ecological impact, particularly on climate- sensitive,
cold- water species.
Salmonid fishes are especially sensitive to climate change and
altered environmental conditions because they require cold- water
habitat s that are increasingly fragmented by human activities,
thereby forcing populations to tolerate environmental conditions
in situ (Kovach et al., 2016). Consequently, many native trout and
char species and lineages are endangered across the Northern
Hemisphere (Muhlfeld et al., 2018; Muhlfeld et al., 2 019). A rctic
char (Salvelinus alpinus) is the most cold-adapted and northerly
distributed freshwater fish globally that may be especially sensi-
tive to climate change (Layton et al., 2021; Reist, Wrona, Prowse,
Power, Dempson, Beamish, et al., 2006).Itisalso a socioeconomi-
cally important species for both recreational fishing and consump-
tion (Klemetsen, 2013). However, populations have significantly
declined, particularly in the southern part of their Holarctic range,
with peripheral populations persisting in cold, deep lakes at tem-
perate latitudes(Kelly etal., 2020). The decline of Arctic char has
been attributed to various human- driven impact s, including climate
change, habitat loss, overfishing, pollution, invasive species, and
complexinteractionsamongthesestressors(Weinsteinetal.,2024).
GlobalclimatechangeisanticipatedtofurtherendangerArcticchar
bywarminghabitatsbeyondtheirthermalpreference(i.e.,0–10°C)
(Hein et al., 2012; Larsson, 2005).As a result, Arcticcharpopula-
tions may face increased vulnerability to decline or extirpation in the
face of ongoing climate change and other anthropogenic pressures.
Thus, uncovering complex relationships between environmental
conditionsandArcticchardistributionisparticularlyimportantfor
unders tanding how future climate ch ange may affect the p ersistence
of this cold-ada pted species a nd biodiversi ty of Arcti c freshwater
ecosystems.
Climate change vulnerabilit y assessments are valuable tools for
identif ying species that are most likely to be vulnerable to the im-
pactsofclimatechange(Fodenetal.,2018; Pacifici et al., 2015).By
evaluating the sensitivit y and exposure of species to various climatic
and environmental changes, vulnerability assessments can help
assess species' risks of extinction or decline, identify geographic
areas or populations of concern, and guide conser vation efforts to
mitigate climate change impacts. However, in recent decades there
has been a growing interest in assessing the climate vulnerability of
freshwater species based on downscaled models of temperature
that predict habitats where temperatures will be within the thermal
limitsofcold-waterfishes(Kovachetal.,2016).Yet,suchapproaches
fail to consider complex interactions between multiple environmen-
tal stressors and their combined effects on the persistence of spe-
ciesunderfuture climaticconditions(Pacificietal.,2015).Machine
learningtechniques(e.g., random forest, neuralnetworks, etc.)are
increasingly used in ecological research for identifying the environ-
mental factors that influence species distribution across diverse
landscapes(Lucas,2020).Machine learning algorithms, trainedon
large and complex datasets, capture non- linear relationships be-
tween species occurrence and environmental variables, improving
prediction accuracy and potentially revealing complex ecological
interactions among variables (Breiman, 2001). As such, t hese ap-
proaches may provide powerful insights intospecies' vulnerability
to climate change and for guiding effective climate adaptation and
conservationstrategies(Cutleretal.,20 07).
Studies examining how climate change and environmental con-
ditions influence the vulnerabilit y of cold- adapted species across
high- latitude landscapes are needed for predicting the future of
Arcticfreshwater biodiversity. Here, we quantify the vulnerability
of Arcti c char to future cli mate change across S candinavia. Us ing
an extensive dataset of fish species occurrence and environmental
information from 1762lakes sampled inthe mid-1990s (herein re-
ferredtoas“baseline”conditions),weusearandomforestmodelto
predictArcticchardistributionunderfutureclimatescenarios(mid
andlate21stcentury).Resultsprovideacomprehensiveassessment
ofthe environmental factors influencing the distribution of Arctic
characrossdiverseArcticlandscapesandidentifypotentialrefugia
for persistence of this cold- adapted species under future climate
warming.
2 | MATERIALS AND METHODS
2.1 | Fish community and environmental data
We combined ex tensive datasets on fish communities, climate,
and limnological parameters to assess the environmental factors
influen cing the distr ibution of Arc tic char in 1762 Scandinavia n

|
3 of 10
MUHLFELD et al.
lakes (Nor way, Sweden, and Fi nland) samp led in the mid-1990s.
Lakes were selected from the 1995 Northern European Lake
Survey, which aimed to evaluate water quality (Henriksen
et al., 1998).Fishcommunitydatawereobtainedbyco-authorsin
Scandinaviafromthe1995–1997NordicLakesFishSurvey,which
aimed to assess the status of fish populations in Fennoscandian
lakes (≥0.04 km2) (Tammi et al., 2003). Fish presence–absence
datawereobtainedusingstandardizedquestionnaires(Hesthagen
et al., 1993; Hesthagen et al., 1999; Tammi et al., 2003),targeting
local experts like landowners and municipal environmental man-
agers. Method validity was confirmed by cross- referencing with
test- fishing data, which have proven highly reliable for Norwegian
lakeswithlimitedfishspecies(Hesthagenetal.,1993).Inadditio n,
we restricted the geographical area to the known historical dis-
tributionofArcticchartoavoidfalseabsences,usingeithermaps
georeferenced from literature sources (Daverdin et al., 2019;
Huitfeldt- Kaas, 1918) or,since Arctic charisananadromous fish
originally immigrating to Scandinavia from the sea after the last
ice- age, the historical high sea level delineation. Historical sea lev-
els were compiled from the Finland Geological Survey, Geological
Survey of Sweden, and Norway. Predictor variables for modeling
the dist ribution of Arc tic char (see Se ction 2.2 below)included
water chemistry attributes (total phosphorus (P), total organic
carbon(TOC),andpH;Henriksenetal.,1998),bioticinteractions
(occurrenceofbrowntroutandnorthernpike;Tammietal.,2003),
human disturbance (i.e., Human Footprint estimated in 1993)
(Venter et al., 2016), and lake area (Henriksen et al., 19 98).
Additionally,weusedendofthe20thcenturyclimatesimulations
(1961–1990)ofmeansummerairtemperatureandmeansummer
precipitation to characterize baseline climatic conditions for each
lake, which allowed us to consistently projec t potential changes in
Arctic char distribution under future climate warming scenarios
(seeSection2.3below)(Navarro-Racinesetal.,2020).
2.2 | Occurrence modeling
Weusedrandomforestmodels(Cutleretal.,2007 )toquantifythe
importance and estimate the par tial dependence of several abiotic
and biotic f actors on the p resence and abse nce of Arctic cha r in
lakes(Figure S1).Toincreasethepredictiveaccuracyoftheanalysis,
random forest models were trained on half of the dataset and fitted
to the other half of the data. In our dataset, absences are much more
common (1433absencesand329 presences); therefore, weuseda
stratified random forest where each tree was fit to a random sample
of 150 presences and 150 absences. We assessed a range of strati-
fied sample sizes, and the choice of stratification sample size did not
change the accuracy of the model or covariate effects. We fit ran-
dom forest models including 500 0 trees using the “randomForest”
packageinR(RCoreTeam,2023).
To assess the str ength of covaria te effect s on Arctic char o c-
currence, we calculated variable importance using the mean de-
creaseintheGiniimpuritymetric.Thismetricmeasuresthemodel's
ability to correctly classify presence or absence for each covariate
(nodepurity)andisvaluableforuseinclassificationanalyses(Cutler
et al., 20 07). Alarger number indicatesthat when a variable is in-
cluded in a tree the rate of correc t classification is increased. Other
measures of variable importance (e.g., mean accuracy decrease)
yielded similar results, except for the presence of brown trout being
animportant predictor of Arctic char presence (seeSection 4). To
assess the direction and overall shape of each covariate effect and
interactions betweenvariables on Arcticchar presence,we calcu-
lated the partial dependence using the “pdp” package in R. While
variables in a random forest model do not need to be transformed to
meet parametric assumptions, we log- transformed total organic car-
bon, precipitation, total phosphorus, lake area, and human footprint
because of their skewed distributions to make the interpretation of
partial effects easier.
2.3 | Future predictions
We used climate projections from CMIP5 to assess the potential im-
pactoffutureclimatechangeonArcticcharlakehabitats(Navarro-
Racines et al., 2020). An ensemble of three General Circulation
Models(GCMs)(GFDL-ESM2M,BCC-CSM1,andMPI-ESM-LR)was
employed to optimize the simulation of mean summer air tempera-
tures (July through September) across Scandinavia (1 km2 resolu-
tion). This simulation covered historical conditions (1961–1990)
andfutureclimatescenariosforthe2050s(2040–2069)and2080s
(2070–2099) under representative concentration pathways(RCPs)
4.5 and 8.5. RCP 4.5 represents a future with moderate greenhouse
gas emissions, resulting in a radiative forcing increase of 4.5 watts
per square meter by 2100. Conversely, RCP 8.5 depicts a high-
emission future where greenhouse gas concentrations continue to
rise unabated, leading to a radiative forcing of 8.5 watts per square
meter by the year 2100.
TopredictthefutureoccurrenceofArctic charunder different
climate scenarios, we used the fitted random forest model with pre-
dictions of future temperatures under two future climate scenarios
andtwodifferenttimeperiods:RCP4.5(2050and2080)andRCP
8.5(2050and2080).ThresholdsofamodeledprobabilityofArctic
char occurrence from the fit ted random forest model were used to
determineriskcategoriesforfuturepredictionsofArcticcharpres-
enceundervariousclimatescenarios(Figure S2). Thevastmajority
ofobservedArctic charpresences(87%) occurredwherethemod-
eled probability of occurrence was greater than .8, and no presences
were obser ved in lakes with a modeled probability less than .6.
Therefore, we used these values to set vulnerability thresholds for
future presence. Lakes where the probabilit y of occurrence in a fu-
ture scenario was greater than .8 were considered low risk, lakes
where the future probability of occurrence was less than .6 were
considered high risk, and lakes between .6 and .8 were considered
moderate risk. All other variables were assumed tobe unchanged
in future scenarios, yet their effects moderate risk through interac-
tions within the model.

4 of 10
|
MUHLFELD et al.
3 | RESULTS
Environmental conditions strongly influenced the spatial distribu-
tionofArcticcharacrossScandinavianlakes(Figure 1; Figure S1).
The full random forest model correctly classified the presence or
absenc eofA rcticch arin89%ofs ample dlakes(Table S1).Th ep rob-
ability of Arctic charpresence decreased withincreased summer
temperatures, TOC concentrations, and the presence of northern
pike(Figure 2a).Amongthesefactors,temperaturehadthestrong-
est eff ect on the dist ribution of Arc tic char, as they rare ly were
predicted to occur in lakes with mean summer air temperatures
above13°C(Figure 2b).Moreover,Arcticcharwerenotablyabsent
from lakes w ith TOC concentr ations exceedi ng 4.5 mg/L , indicat-
ing a strong negative effect of water browning on their occurrence
(Figure 2c). The presence of northern pike showed a significant
bioticinteractionwithArctic char,reducingtheirlikelihoodofoc-
currence by approximately half in lakes where northern pike were
present(Figure 2d).
There were also important interac tive effects among some of
theabiotic and biotic variables influencing Arctic char occurrence.
Specifically, lake area moderated the negative ef fect s of warm tem-
peraturesandinteractionswithnorthernpike(Figure 3a,b).Inlakes
whereairtemperaturesexceeded13°C,Arcticchar were1.5times
morelikelytooccurin largerlakeswithan area greaterthan3 km2,
likely due to the increased availability of deep cold- water refuges
(Figure 3a). Additionally, larger lakes (Figure 3b) and those with
coldertemperatures(Figure 3c)demonstratedahigherprobabilityof
Arcticcharcoexistencewithnorthernpike.Thesefindingshighlight
theimportanceoflakesize(andvolume)inmoderatingtheimpacts
oftemperatureandbioticinteractionsonArcticchardistribution.
FIGURE 1 SpatialdistributionofArcticcharacrossScandinavia.(a–c)MapsshowingthesamplinglocationsandoccurrenceofArctic
char(Tammietal.,2003)inrelationtoelevation(a),totaldissolvedcarbon(TOC)(Henriksenetal.,19 98)andmeansummertemperature
(b)(Navarro-Racinesetal.,2020),andnorthernpikeoccurrence(c)(Tammietal.,2003).SummarydataareincludedinFigure S1. Map lines
delineate study areas and do not necessarily depict accepted national boundaries.

|
5 of 10
MUHLFELD et al.
Future climate change is predicted to significantly reduce the
extent of suitable lake habitatssupporting Arcticchar.Wemod-
eled future habitatconditions and theoccurrence of Arctic char
under future temperature warming (RCP4.5 and RCP8.5) sce-
narios. O ur projections suggest a 4 0%–80% decline in s uitable
habitat s that are likely to support populations persisting on the
landscape by the end of the 21st century. Warming projections
under the RCP4.5 emission scenario suggest that 29% of extant
populations (baseline) are at high risk of extirpation by 2050
(Figure 4a)and 42% by 2080 (Figure 4b),while 45% have ame-
dium risk of ex tirpation by 205 0 (Figure 4a)a nd 44% by 2080
(Figure 4b). Under the RCP8.5 scenario, warming projections
indicate a more pronounced trend: by 2050, 52% of populations
face a high riskofextirpation (Figure 4c),rising to 81%by2080
(Figure 4d),while40%facemediumriskby 2050 (Figure 4c),de-
cliningto16%by2080(Figure 4d).These resultssuggestsignifi-
cant range contractions in Scandinavian lakes for this cold- adapted
species, primarily in warm, lower- elevation lakes with high TOC
concentrations, predominantly at the southern range limit. Under
the RCP4.5 scenario, populations with a low risk of extirpation are
projected to decrease from 26%by 2050 (Figure 4a) to 14%by
2080(Figure 4b),whileunder themoresevereRCP8.5scenario,
thosenumbersdropfrom8%by2050(Figure 4c)toamere2%by
2080(Figure 4d).Theselow-riskpopulationsarelikely to persist
FIGURE 2 Environmentalfactors
influencingthedistributionofArcticchar.
(a)Variableimportanceforrandomforest
regressionofArcticcharoccurrence
againstenvironmentalvariables.Variables
are ordered from highest to lowest
by random forest variable importance
(seeSection2).Thefullrandomforest
regression model is 88.5% accurate
in predicting the presence or absence
ofArcticchar(a).Partialdependence
plotsoftemperature(b),totalorganic
carbon(naturallog)(c),andnor thernpike
(presence/absence)(d)ontheoccurrence
ofArcticchar.Boxplotsalongthex-axes
denote the range of observed presences
(blue)andabsences(orange)foragiven
predictor. Pie chart s in panel b show
theproportionofArcticcharpresences
relative to northern pike presences
(orange)andabsences(blue).
FIGURE 3 InteractiveeffectsofabioticandbioticfactorsinfluencingtheoccurrenceofArcticchar.(a–c)Partialdependenceofthe
interactiveeffectsbetweenlakeareaandsummertemperature(a),lakeareaandnorthernpikeoccurrence(b),andsummertemperatureand
northernpikeoccurrence(c)ontheoccurrenceofArcticchar.

6 of 10
|
MUHLFELD et al.
at higher latitudes and elevations in the mountainous regions and
along the coastal areas of the Norwegian Sea.
4 | DISCUSSION
Species distribution models have been widely applied as decision-
support tools for strategic adaptation and conservation planning for
freshwater species. However, most approaches are based on down-
scaled models of water temperature that predict habitats where
temperatures will be within the thermal limit s of cold- water fishes
(Arms trong et al., 2021). Suc h climate-envelop e approache s often
neglect how changes in temperature interact with other environ-
mentalfactorstoaffectspecies'distribution(Fodenetal.,2018).We
used an extensive environmental monitoring dataset and advanced
machine learning to evaluate the complex interplay of environmen-
tal, physical, and physiological factors influencing the occurrence
ofArcticchar.Randomforestmodelsdemonstratethatabiotic and
biotic factors strongly influenceArcticcharoccurrence, withlakes
experiencing warm summer temperatures, high TOC levels, and
northernpikebeinglesslikelytosupportArcticcharunderbaseline
(1990 s)andfuturewarming(2050and2080)co nditions.Thesefind-
ings underscore the importance of considering these complexities
in climate vulnerability assessments and conservation planning for
freshwater species.
Waterbrowninghasastr ongn eg ativeeffectonArcticcharpres-
ence in Scandinavian lakes. Water browning can disrupt lake food
webs by decreasing water transparency, benthic primary produc-
tion,andthusdissolvedoxygenconcentrations(Thraneetal.,2014;
Vasconcelos etal., 20 19). Water browning likelyaffects theforag-
ing ability (i.e.,searchfield)and food resources available to Arctic
char (Karlsson etal., 2009),with the potential to ultimatelyaffect
population production (Finstad etal., 2 014; Karlsson et al., 2009).
Browningalsoaffectsthermalstratification(i.e.,moreheattrapped
intheupperpartofthewatercolumn),causingincreasedresistance
towardmixing(Palmeret al., 20 14).Inaddition,increased inputsof
organic C not only reduce photosynthesis with depth, but also in-
crease microbial respiration. The sum of these ef fect s is likely to re-
duce deep- water oxygen concentrations, posing a specific challenge
toanoxygen-demandinglakespawnerlikeArcticchar.
FIGURE 4 Projectedextirpationrisk
ofArcticcharunderfutureclimates.
Projectedextirpationrisk(vulnerability)
forArcticcharpopulationsunderfuture
climate warming scenarios: RCP4.5 2050
(a)and2080(b)andRCP8.52050(c)and
2080(d).Extirpationriskiscalculated
from the future probabilit y of occupancy
(p)forArcticcharundereachscenario.
Reddotsindicatehighrisk(p < .6),yellow
mediumrisk(p = .6–.8),andbluearelow
risk(p > .8).Maplinesdelineatestudy
areas and do not necessarily depict
accepted national boundaries.

|
7 of 10
MUHLFELD et al.
Understanding how ecosystems respond to climate change
depends on examining how habitat conditions interact with the
prevailing climate (Parmesan, 2006). Our study revealed that
larger lakes can mitigate the adverse impacts of warm tempera-
turesandinteractionswithnorthernpikeonArcticcharpresence.
Wefound thatlarger lakes (>3 km2) are1.5 times morelikely to
host Arctic char in areas with average air temperatures exceed-
ing 13°C. These larger lakes also facilitate the coexistence of
Arcticcharwithnorthernpike,whichtypicallyaffectcharoccur-
rence.NorthernpiketendtooutcompeteandpreyonArcticchar
inwarmer,moreproductivelakes,whileArcticchartendtothrive
insmaller,colder,oligotrophiclakeswithextendedicecover(Hein
et al., 2012). T hese deep, larg e lakes create colde r, oxygen-rich
layers through thermal stratification, providing a stable thermal
environment for cold-adapted fishes like Arc tic char. As global
temperatures rise, these deep, cold lakes will serve as critical ref-
uges for Ar ctic char, enablin g them to coexis t with competit ors
and predators like northern pike.
The presence of brown trout did not strongly influence the
occurrence of Arctic char,likely due to their coexistence in larger
lakes and shared preference for cold temperatures. While brown
troutpresencewasagoodpredictorofArcticcharpresence,itper-
formed poorly in discriminating between presences and absences.
These results suggest that at a broad scale, the geographic distribu-
tion of these species is similar, making brown trout presence a good
indicat or of suitable ha bitat for Arc tic char. At smaller sc ales, the
geographical distributions of these cold- water species are primarily
influenced by ecosystem productivity and competitive interactions,
with Arc tic char favoring cold, low-produ ctivity lakes and brown
troutfavoringwarmer,moreproductivelakes(Finstadetal.,2011).
In sympat ric conditions, interspecific competitio n can lead to the dis-
placementofArcticcharfromlittoralhabitats(Elliott&Elliott,2010 ;
Eloranta et al., 2013), while warming temperatures and decreased
oxygenlevelsmayfurtherlimitsuitableArcticcharhabitats(Elliott&
Elliott, 2010).Climatechangeandanthropogenicstressorsmaydis-
proportionatelyreducelakehabitatnichesavailableforArcticchar,
potentiallyallowingbrowntrouttoexpandintovacantniches(Hein
et al., 2012).Thisexpansionmaynegativelyim pa ctArcticcharab un-
dance, a dynamic not fully captured in our presence- only analysis.
OurresultsportendsignificantrangecontractionsofArcticchar
across Scandinavia due to future global warming, particularly near
the southern range limit and at lower elevations. Under a conserva-
tiveemissionscenario(RCP4.5),42%ofpopulationsfacehighriskof
extirpation by the end of the 21st century, increasing to 81% under
ahigh-emissionscenario(RCP8.5).Theseresultssubstantiateother
studies projecting significant climate- induced range contractions at
smallergeographicalscales.Forexample,Heinetal.(2012)predicted
a 73% range reduction in Swedish lakes by 2100. Range contractions
are expected primarily in warm, lower elevation lakes with high TOC
concentrations, mostly at the southern range limit. Projections indi-
cate that only 14% and 2% of populations are estimated to face low
risk of extirpation by 2080 under RCP4.5 and 8.5 scenarios, respec-
tively. Low- risk populations are likely to persist at higher latitudes
and elevations in the mountainous regions and coastal areas of the
Norwegian Sea. Notably, our “baseline” datasets originate from the
1990s, and current fish distribution and environmental conditions
may have changed since then. These results can inform management
strategies to restore and protect critical habitats, and to identify and
prioritize “climate refugia” that support species persistence as the
climate cont inues to warm and tr ansform the Ar ctic's freshwater
ecosystems.
The persistence of many species is ultimately linked to whether
they can adapt in place to rapid environmental changes or shif t their
distribution to track suitable habitats. For range- restricted freshwa-
ter speci es like Arctic c har,t he greatest cha llenges of climat e ad-
aptation will be related to adaptive capacity, dispersal ability, and
habitatalterations.WhileArcticcharhavedemonstratedphenotypic
adaptabilitytorapidlychangingtemperatures(Hookeretal.,2023),
their potential for northward and upward expansion is limited unless
newhabitatsemergefromretreating glaciers (Pitmanetal., 2020),
especially in polar regions such as Svalbard. Consequently, shifts
in habitat conditions and connectivity are anticipated to reduce
overall distribution and abundance, increase isolation, diminish
gene flow, and e rode genetic and eco logical diversit y–critical for
adaptat ion and resilie ncy. Additionall y,Ar ctic char disp lay consid-
erable intra- species diversity, with individuals within lakes varying
in morph ology, behavior, and life h istory (Kle metsen et al., 20 03;
Weinstein et al., 2024). This diver sity may stem fro m differences
in genetic populations or phenotypic plasticity, related to feeding
niches, s pawning locations, and ph enology (Brun ner et al., 2001;
Klemetsen, 2010 ). While our study assumes uniform responses to
environmental stressors within each lake, it is important to acknowl-
edge that the loss of genetically distinct populations may exceed our
predictions, thus magnifying the predicted loss of diversity.
These findings enhance our understanding of the abiotic and bi-
oticfactorsinfluencingArcticcharpopulationsandtheirvulnerability
toclimatechangeacrossArcticlandscapes.Theinteractionbetween
climatic and anthropogenic stressors underscores the urgency of de-
veloping proactive climate adaptation and mitigation strategies to
protect populations and diverse habitats in high- latitude landscapes.
Conser vation strategies might include protecting climate refugia,
restoring habitat diversity and connectivity, translocating imper-
iled populations, establishing native fish reserves, and minimizing
anthropogenic impacts such as pollution, habitat destruction, and
exotic species introductions. Such conser vation measures may hold
promise forenhancingtheadaptation andresilienceofArcticchar
and other cold- water species to impending climate warming across
high- latitude landscapes.
AUTHOR CONTRIBUTIONS
Clint C. Muhlfeld: Conceptualization; data curation; formal analysis;
funding acquisition; investigation; methodology; project administra-
tion;resources;writing–originaldraft;writing–reviewandediting.
Timothy J. Cline: Form al analysis; me thodolog y; writing – o riginal
draft;writing – review andediting. Anders G. Finstad: Data cura-
tion; meth odology; wr iting – review and edi ting. Dag O. Hessen:

8 of 10
|
MUHLFELD et al.
Conceptualization; data curation; formal analysis; funding acquisi-
tion;investigation;projectadministration;resources;writing–origi-
nal draft; writing – review andediting. Sam Perrin: Data curation.
Jens Thaulow: Data curatio n; writing – revi ew and editing . Diane
Whited:Datacuration;formalanalysis;methodology;writing–re-
view and editing. Leif Asbjørn Vøllestad: Conceptualization; data
curation; formal analysis; funding acquisition; investigation; project
administration;resources;writing – originaldraft;writing–review
and editing.
ACKNOWLEDGEMENTS
This research was supported by the US Fulbright Program, US
Geological Sur vey Nor thern Rocky Mountain Science Center, and
theUniversity ofOslo. Any use of trade,firm,orproductnamesis
for descriptive purposes only and does not imply endorsement by
the US government.
CONFLICT OF INTEREST STATEMENT
The authors declare no conflict of interest present.
DATA AVA ILAB ILITY STATE MEN T
The data and code that suppor t the findings of this study are openly
available in Zenodo at https:// doi. org/ 10. 5281/ zenodo. 11459009.
Lake location and fish occurrence was accessed from the Norwegian
Institute for Nature Research and are available upon request.
Lake area and biogeochemistry was accessed from the Norwegian
Institute for Water Research Repository at https:// niva. brage. unit.
no/ niva- xmlui/ handle/ 11250/ 209342. Human Footprint maps were
accessed from htt p s: // da tad r y ad. org / s t ash / da t as e t/ do i: 10 . 50 61/
dryad. 052q5 . Climate data were accessed from http:// ccafs - clima te.
or g / .
ORCID
Clint C. Muhlfeld https://orcid.org/0000-0002-4599-4059
Timothy J. Cline https://orcid.org/0000-0002-4955-654X
Anders G. Finstad https://orcid.org/0000-0003-4529-6266
Dag O. Hessen https://orcid.org/0000-0002-0154-7847
Sam Perrin https://orcid.org/0000-0002-1266-1573
Jens Thaulow https://orcid.org/0000-0002-4063-6738
Diane Whited https://orcid.org/0000-0002-6255-715X
Leif Asbjørn Vøllestad https://orcid.org/0000-0002-9389-7982
REFERENCES
Armstrong,J.B.,Fullerton,A.H.,Jordan,C.E.,Ebersole,J.L.,Bellmore,J.
R.,Arismendi,I.,Penaluna,B.E.,&Reeves ,G.H.(2021).Theimpor-
tance of warm habit at to the growth regime of cold- water f ishes.
Nature Climate Change, 11(4), 354–361. http s:// doi. org / 10. 103 8/
s4155 8- 021- 00994 - y
Barbar ossa, V., Bosmans , J., Wanders, N ., King, H. , Bierkens, M. F. P.,
Huijbre gts, M. A . J., & Schipper, A . M. (2021). Threat s of global
warmingto the world'sfreshwaterfishes. Nature Communications,
12(1),1701.htt ps:// doi. org/ 10. 10 38/ s4146 7- 021- 21655 - w
Breiman,L.(20 01).Randomforests.Machine Learning, 45,5–32.
Brunner,P.C.,Douglas,M.R., Osinov,A ., Wilson,C. C.,& Bernatchez,
L. (2001). Holarctic phylogeography of Arctic charr (Salvelinus
alpinus L.) infer red from mitoc honrial DNA s equences. Evolution,
55,573–586.
Crapar t, C., F instad, A . G., He ssen, D. O., Vogt, R . D., & Ander sen, T.
(2023). Spa tial predict ors and tempor al forecas t of total organi c
carbon levels in boreal lakes. Science of the Total Environment, 870,
161676. https:// doi. org/ 10. 1016/j. scito tenv. 2023. 161676
Cutler,D.R.,Edwards,T.C.,Beard,K.H.,Cutler,A.,&Hess,K.T.(2007).
Random forest s for classification in ecology. Ecology, 88(11),
2783–2792.
Daverdin , M., Finstad, A . G., & Blumen trath, S. (2019). Freswater fish
native distribution map transcriptions from Huit feldt- Kaas, H. (1918).
Norges Teknisk- Naturvitenskapelige Universitet. https:// doi. org/
10. 21400/ 1MW T3950
de Wit, H. A., Valinia, S., Weyhenmeyer, G. A., Futter, M. N.,
Kortelainen,P.,Austnes,K.,Hessen,D.O.,Räike,A.,Laudon,H.,
&Vuorenmaa,J.(2016).Currentbrowningofsurfacewaterswill
Be further promoted by wet ter climate. Environmental Science &
Technology Letters, 3(12),430–435.https:// doi. org/ 10. 1021/ acs.
estle tt. 6b00396
Elliott,J.M.,&Elliott,J.A.(2010).TemperaturerequirementsofAtlantic
salmon Salmo salar, brown trout Salmo trutta and Arctic charr
Salvelinus alpinus: Predicting the effects of climate change. Journal
of Fish Biology, 77(8), 1793 –1817. htt ps:// doi. org/ 10 . 1111/j. 10 95-
8649. 2010. 02762. x
Eloranta,A.P.,Knudsen,R.,&Amundsen,P.-A.(2013).Nichesegregation
ofc oexis tingA rct iccharr (Salvelinus alpinus)a ndbro wn trout(Salmo
trutta)constrains food web couplinginsubarcticlakes.Freshwater
Biology, 58(1),207–221.ht tp s:// doi. o rg / 10. 1111 / f wb. 120 52
Feng,D.,Gleason,C.J.,Lin,P.,Yang, X.,Pan,M.,& Ishitsuka,Y.(2021).
Recent changes to Arctic river discharge. Nature Communications,
12(1),6917.https:// doi. org/ 10. 10 38/ s4146 7- 021- 27228 - 1
Finstad, A., Andersen, T., Larsen, S., Tominaga, K.,Blumentrath, S., de
Wit, H. , Tømmerv ik, H., & Hessen, D. (2016). From green ing to
browning: Catchment vegetation development and reduced S-
deposition promote organic carbon load on decadal time scales in
Nordic lakes. Scientific Reports, 6, 31944. ht tps:// doi. o rg/ 10. 103 8/
srep3 1944
Finstad,A.G.,Forseth,T.,Jonsson,B.,Bellier,E.,Hesthagen,T.,Jensen,
A. J., He ssen, D. O., & Fold vik, A. (2011). Com petitive excl usion
along climate gradients: Energy efficiency influences the distri-
bution of t wo salmonid fishes. Global Change Biology, 17(4), 1703–
1711. https:// doi. org/ 10. 1111/j. 1365- 2486. 2010. 02335. x
Finstad,A.G., Helland,I.P.,Ugedal,O., Hesthagen,T.,&Hessen, D.O.
(2014).Unimodalresponseoffishyieldtodissolvedorganiccarbon.
Ecology Letters, 17(1),36–43.htt ps:// doi. or g/ 10 . 1111/ ele. 122 01
Foden, W. B., Young , B. E., Akçak aya, H. R., G arcia, R. A ., Hoffman n,
A. A.,Stein, B.A., Thomas,C. D., Wheatley, C. J.,Bickford, D. P.,
Carr, J. A.,Hole, D. G ., Martin, T. G.,Pacifici, M., Pearce-Higgins,
J. W., Platts, P. J., Visconti, P., Watson, J. E. M., & Huntley, B.
(2018).Climatechange vulnerability assessment ofspecies.Wiley
Interdisciplinary Reviews: Climate Change, 10, e551.
Hayden, B.,Harrod,C., Thomas, S.M., Eloranta,A. P.,Myllykangas, J.-
P., Siwertsson, A., Præbel, K., Knudsen, R., Amundsen, P.-A., &
Kahilainen,K.K.(2019).Fromclearlakestomurkywaters—Tracing
the functional response of high- latitude lake communities to con-
current “greening” and “browning”. Ecology Letters, 22(5),807–816.
https :// doi. or g/ 10. 1111/ e le. 1323 8
Hein,C.L.,Öhlund,G.,&Englund,G.(2012).FuturedistributionofArctic
char Salvelinus alpinus in Sweden under climate change: Effects of
temperature, lake size and species interactions. Ambio, 41(3),303–
312. https:// doi. org/ 10. 1007/ s1328 0- 012- 0308- z
Heino,J.,Virkkala,R.,&Toivonen,H.(2009).Climatechangeandfresh-
water biodiversity: Detected patter ns, future trends and adapta-
tions in northern regions. Biological Reviews, 84,39–54.ht t p s : // doi .
org / 10. 1111/j . 1469- 185X . 2008 . 0 0 06 0. x

|
9 of 10
MUHLFELD et al.
Henriksen, A.,BritLisa, S., Jaakko,M.,Wilander,A., Ron,H.,Curtis,C.,
Jensen,J.P.,Erik,F.,&Tatyana,M.(1998).NorthernEuropeanLake
survey, 1995: Finland, Nor way, Sweden, Denmark, Russian Kola,
Russian Karelia, Scotland and Wales. Ambio, 27(2), 80–91.ht t p ://
www. jstor. org/ stable/ 4314692
Hesthagen, T., Rosseland, B.O., Berger,H. M., & L arsen, B. M.(1993).
Fish community s tatus in Norwegian lakes in relation to acidifica-
tion:Acomparisonbetweeninterviewsandactualcatchesbytest-
fishing. Nordic Journal of Freshwater Research, 68,34–41.
Hesthagen, T., Sevaldrud, I. H., & Berger, H. M.(1999).Assessment of
damage of fish populations in Norwegian lakes due to acidification.
Ambio: A Journal of the Human Environment, 28,112–117.
Hooker,O.E., Adams, C. E.,&Chavarie,L. (2023).Arcticcharr pheno-
typic responses to abrupt gener ational scale temperature change:
Aninsight into howcold-water fishcould respondtoextremecli-
matic events. Environmental Biology of Fishes, 106(5), 909–922.
https:// doi. org/ 10. 1007/ s1064 1- 022- 01363 - 0
Huitfeldt-Kaas, H.(1918).Fer skvandsfiskenes utbredelse og indvandring i
Norge: Med et tillæg om krebsen.Centraltrykkeriet.(inNor wegian).
Karlsson, J., Byström, P., Ask, J., Ask, P., Persson, L., & Jansson, M.
(2009). L ight limita tion of nutri ent-p oor lake ecos ystems. Nature,
460(7254),506–509.https :// doi. org/ 10 . 103 8/ natur e0 8179
Kelly, S., Moore, T. N., de Eyto, E ., Dillane, M., Goulon, C., Guillard, J.,
Lasne, E., McGinnity, P., Poole, R ., Winf ield, I . J., Woolway, R. I.,
&Jennings, E.(2020).WarmingwintersthreatenperipheralArctic
charr populations of Europe. Climatic Change, 163 (1), 599–618.
https:// doi. org/ 10. 1007/ s1058 4- 020- 02887 - z
Klemet sen, A. (2010). The charrproblem revisited: Exceptional phe-
notypic plasticity promotes ecological speciation in postglacial
lakes. Freshwater Reviews, 3, 49–74. https:// doi. org/ 10. 16 08/
FRJ- 3.1 . 3
Klemet sen, A.(2013). Themost variablevertebrateonEarth.Journal of
Ichthyology, 53(10),781–791.https:// doi. org/ 10 . 1134/ S 0032 94521
3100044
Klemet sen, A., Amundsen, P.-A., Dempson, J.B., Jonsson, B., Jonsson,
N.,O'Connell,M.F.,&Mortensen,E.(2003).AtlanticsalmonSalmo
salar L., brown trout Salmo truttaL. and Arctic charr Salvelinus al-
pinus (L.): A review of aspects of their life histories. Ecolog y of
Freshwater Fish, 12,1–59.
Kovach,R.P.,Muhlfeld,C.C.,Al-Chokhachy,R.,Dunham,J.B.,Letcher,
B.H.,&Kershner,J.L.(2016).Impactsofclimaticvariationontrout:
A global synthesis andpath forward [ journalar ticle]. Reviews in
Fish Biology and Fisheries, 26 (2),135–151.htt ps:// doi. or g/ 10. 1007/
s1116 0 - 015- 9414- x
Lars en, S., Ande rson, T., & Hesse n, D. O. (2011). Climate cha nge pre-
dicted to cause severe increase of organic carbon in lakes. Global
Change Biology, 17(2), 118 6–1192.ht tps:// doi . org / 10. 1111 /j. 1365 -
2486. 2010. 02257. x
Larsson,S.(2005).ThermalpreferenceofArcticcharr,Salvelinus alpinus,
and brown trout, Salmo trutta—Implicationsfortheirnichesegrega-
tion. Environmental Biology of Fishes, 73(1),89–96.https:// doi. org/
10. 1007/ s1064 1- 004- 5353- 4
Layton,K.K.S.,Snelgrove,P.V.R.,Dempson,J.B.,Kess,T.,Lehnert,S.J.,
Bentzen,P.,Duffy,S.J.,Messmer,A.M.,Stanley,R.R.E.,DiBacco,
C., Salisbur y, S. J., Ruzzante, D. E., Nugent, C . M., Ferguson, M . M.,
Leong,J.S.,Koop,B.F.,&Bradbury,I.R.(2021).Genomicevidence
of past an d future climate -linke d loss in a migrator y Arctic f ish.
Nature Climate Change, 11(2), 158–165. htt ps: // doi. or g/ 10. 1038/
s4155 8- 020- 00959 - 7
Li, X., P eng, S., Xi , Y.,Wool way,R . I., & Liu, G . (2022). Ea rlier ice loss
accelerates lake warming in the Northern Hemisphere. Nature
Communications, 13(1), 5156. https:// doi. or g/ 10. 1038/ s4146 7-
022- 328 30 - y
Lucas,T.C.D.(2020).Atranslucentbox:Interpretablemachinelearning
in ecolog y. Ecological Monographs, 90(4), e01422. https:// doi. org/
10. 1002/ ec m. 1422
Muhlfeld, C., Dauwalter, D., D'Angelo, V., Ferguson, A., Giersch, J.,
Impson, N., Koizumi, I., Kovach, R., McGinnity, P., Schöffmann,
J., Vøllest ad, L., & Epif anio, J. (2019). Globa l status of trou t and
char: Conservation challenges in the twenty- first century. In J. L.
Kershner,J. E. Williams,R. E. Gresswell,&J.Lobon-Cervia(Eds.),
Trout and char of the world (pp. 717–760). American Fisheries
Societ y.
Muhlfeld, C. C., Dauwalter, D. C., Kovach, R. P., Kershner, J. L., Williams,
J.E.,& Epifanio, J. (2018). Troutinhotwater: A callforglobal ac-
tion. Science, 360(6391), 866–867. https:// doi. org/ 10. 1126/ scien
ce. a at8 455
Navarro -Rac ines, C., Tarapu es, J., Thornto n, P., Jarvis , A., & Ramire z-
Villegas,J.(2020).High-resolutionandbias-correctedCMIP5pro-
jections for climate change impact assessments. Scientific Data,
7(1),7.htt ps:// doi. or g/ 10. 10 38/ s 4159 7- 019- 03 43- 8
Pacifici,M.,Foden,W.B.,Visconti,P.,Watson,J.E.M.,Butchart,S.H.M.,
Kovacs,K.M.,Scheffers,B.R.,Hole,D.G.,Martin,T.G.,Akçakaya,
H.R.,Corlett,R.T.,Huntley,B.,Bickford,D.,Carr,J.A.,Hoffmann,
A.A.,Midgley,G.F.,Pearce-Kelly,P.,Pearson,R.G .,Williams,S.E.,
…Rondinini,C. (2015).Assessingspecies vulnerability toclimate
change. Nature Climate Change, 5(3), 215–224.https:// doi. org/ 10.
1038/ nclim ate2448
Palmer,M.E.,Yan,N.D.,&Somers,K.M.(2014).Climatechangedrives
coherenttrendsinphysicsandoxygencontentin NorthAmerican
lakes. Climatic Change, 124(1), 285–299.h ttps:// doi. org/ 10. 10 07/
s1058 4- 014- 1085- 4
Parme san, C. (20 06). Ecologic al and evoluti onary resp onses to rece nt
climate change. Annual Review of Ecological Systems, 37,637–669.
https:// doi. org/ 10. 1146/ annur ev. ecols ys. 37. 091305. 110100
Perrin, S. W., Bærum, K. M., Helland, I. P., & Finstad, A. G . (2021).
Forecasting the future establishment of invasive alien freshwater
fish species. Journal of Applied Ecology, 58(11),2404–2414.h tt ps: //
doi . org/ 10. 1111/ 1365 - 266 4. 1399 3
Pitman,K.J.,Moore,J.W.,Sloat,M.R.,Beaudreau,A.H.,Bidlack,A.L.,
Brenner,R.E.,Hood,E.W.,Pess,G.R.,Mantua,N.J.,Milner,A.M.,
Radić,V.,Reeves, G. H., Schindler,D.E., & Whited, D.C. (2020).
Glacier retreat and Pacific Salmon. BioScience, 70 (3), 220–236.
https:// doi. org/ 10. 1093/ biosci/ biaa015
R Core Team. (2023). R: A Language and Environment for Statistical
Computing. R Foundation for Statistic al Comp uting. ht tp s://
www.R- proje ct. org/
Reid,A.J.,Carlson,A.K., Creed,I.F.,Eliason, E.J.,Gell, P.A., Johnson,
P.T.J.,Kidd, K.A.,MacCormack,T.J.,Olden,J.D.,Ormerod,S.J.,
Smol,J.P.,Taylor,W.W.,Tockner,K.,Vermaire,J.C.,Dudgeon,D.,
&Cooke,S. J. (2019).Emerging threats and persistent conserva-
tion challenges fo r freshwater biodiversit y. Biological Reviews, 94(3),
849–873.htt ps:// doi. org/ 10 . 1111/ br v. 124 80
Reist, J. D., Wrona, F. J., Prowse, T. D., Power, M., Dempson, J. B., Beamish,
R.J.,King,J.R.,Carmichael,T.J.,&Sawatzky,C.D.(20 06).General
effectsofclimatechangeonArcticfishesandfishpopulations.Ambio:
A Journal o f the Human Environment, 35(7),370–380,311.ht t ps : // doi.
o r g / 1 0 . 1 5 7 9 / 0 0 4 4 - 7 4 4 7 ( 2 0 0 6 ) 3 5 [ 3 7 0 : G E O C C O ] 2 . 0 . C O ; 2
Reist, J. D., Wrona, F. J., Prowse, T. D., Power, M., Dempson, J. B., King, J.
R.,&Beamish,R.J.(2006).Anoverviewofeffectsofclimatechange
on selec ted Arctic fre shwater and anadro mous fishes. Ambio: A
Journal of the Human Environment, 35(7),381–387,387.ht t p s : // d o i .
o r g / 1 0 . 1 5 7 9 / 0 0 4 4 - 74 4 7 ( 2 0 0 6 ) 3 5 [ 3 8 1 : A O O E O C ] 2 . 0 . C O ; 2
Sala, O. E. , Stuart Ch apin, F., Armes to, J. J., Berlo w,E ., Bloomfi eld, J.,
Dirzo, R ., Huber- Sanwald, E., Huenneke, L. F., Jack son, R . B., Kinzig,
A., Leemans, R., Lodge, D. M.,Mooney,H. A., Oesterheld, M.n.,
Poff, N. L., Sykes, M. T., Walker, B. H., Walker, M., & Wall, D. H.
(200 0). Global bio diversity s cenarios for th e year 2100. Science,
287(5459),1770 –1774. https:// doi. org/ 10. 1126/ scien ce. 287. 5459.
1770
Saros, J.E.,Arp, C. D.,Bouchard,F.,Comte,J., Couture, R.-M.,Dean,J.
F., Lafrenière, M., MacIntyre, S., McGowan, S., Rautio, M., Prater,

10 of 10
|
MUHLFELD et al.
C., Tank, S. E., Walvoord, M., Wickland, K. P., Antoniades, D.,
Ayala-Borda,P.,Canario,J.,Drake,T.W.,Folhas,D.,…Vincent,W.
F.(2023). Sentinel responses of Arctic freshwater systems to cli-
mate: Link ages, evidence, and a roadmap for future research. Arctic
Science, 9(2),356–392.ht tps:// doi. org/ 10. 1139/ as- 2022- 0021
Su,G.,Logez,M.,Xu,J.,Tao,S.,Villéger,S.,& Brosse,S.(2021).Human
impacts on global freshwater fish biodiversit y. Science, 371(6531),
835–838.https:// doi. org/ 10. 1126/ scien ce. abd3369
Tammi, J., App elberg, M., B eier, U., Hesthagen , T., Lapp alainen, A ., &
Rask,M.(2003).FishstatussurveyofNordiclakes:Effectsofacid-
ification, eutrophication and stock ing activity on present fish spe-
cies composition. Ambio, 32(2), 98–105. https:// doi. org/ 10. 1579/
0044- 7447- 32.2. 98
Thrane, J.-E., Hessen, D. O., &Andersen, T.(2014). The absorption of
light in lakes: Negative impact of dissolved org anic carbon on pri-
mary productivity. Ecosystems, 17(6), 104 0–1052. https:// doi. org/
10. 1007/ s1002 1- 014 - 9776- 2
Vasconcelos,F.R., Diehl,S., Rodríguez, P.,Hedström,P.,Karlsson, J., &
Byström,P.(2019).Bottom-upandtop-downef fectsofbrowning
and warming on shallow lake foo d webs. Global Change Biology,
25(2),504–521.https: // doi. org/ 10. 1111/ gc b. 14521
Venter,O.,Sanderson,E.W.,Magrach,A.,Allan,J.R.,Beher,J.,Jones,K .
R., Possingham, H. P., Laurance, W. F., Wood, P., Fekete, B. M., Lev y,
M.A.,&Watson,J.E.M.(2016).Globalterrestrialhumanfootprint
maps for 1993 and 2009. Scientific Data, 3(1),160067.h t t p s : // d o i .
org / 10. 103 8/ sda ta. 2016. 67
Vonk, J. E., Tank, S. E., Bowden, W. B., Laurion, I., Vincent, W. F.,
Aleks eychik, P., Amyot, M., B illet, M. F., Caná rio, J., Cor y,R . M.,
Deshpande, B. N ., Helbig, M., Jammet, M., Karlsson, J., Larouche,
J.,MacMillan, G.,Rautio,M.,Walter Anthony,K. M.,&Wickland,
K. P. (2015). Reviews a nd synthes es: Effec ts of permaf rost thaw
on Arct ic aquatic e cosystem s. Biogeosciences, 12(23 ), 7129–7167.
https:// doi. org/ 10. 5194/ bg- 12- 7129- 2015
Weinstein,S.Y.,Gallagher,C.P.,Hale,M.C., Loewen,T.N.,Power,M.,
Reist,J.D.,&Swanson,H.K.(2024).Anupdatedreviewofthepost-
glacialhistor y,ecology,anddiversityofArcticchar(Salvelinus alpi-
nus) and Dolly Var den (S. malma). Environmental Biology of Fishes,
107(1),121–154.https:// doi. org/ 10. 1007/ s1064 1- 023- 01492 - 0
Wrona, F. J., Johan sson, M., Cul p, J. M., Jenkins , A., Mård, J ., Myers-
Smith, I. H.,Prowse, T.D., Vincent, W.F.,&Wookey,P.A. (2016).
Transitions in Arctic ecosystems: Ecological implications of a
changing hydrological regime. Journal of Geophysical Research:
Biogeosciences, 121(3), 650–674. https:// doi. org/ 10 . 1002/ 2015J
G003133
SUPPORTING INFORMATION
Additional supporting information can be found online in the
Suppor ting Information section at the end of this article.
How to cite this article: Muhlfeld, C. C ., Cline, T. J., Finstad,
A.G.,Hessen,D.O.,Perrin,S.,Thaulow,J.,Whited,D.,&
Vøllestad,L.A.(2024).ClimatechangevulnerabilityofArctic
char across Scandinavia. Global Change Biology, 30, e 173 87.
https://doi .org /10.1111/gcb.173 87