Predicting Phenological Shifts in a Changing Climate

Predicting phenological shifts in a changing climate

Katherine Scrantona,1,2 and Priyanga Amarasekarea,1

aDepartment of Ecology and Evolutionary Biology, University of California, Los Angeles, CA 90095

Edited by Nils Chr. Stenseth, University of Oslo, Oslo, Norway, and approved October 26, 2017 (received for review June 21, 2017) Phenological shifts constitute one of the clearest manifestations selection drives thermal optima of lower-latitude ectotherms to of climate warming. Advanced emergence is widely reported in coincide with the mean temperature while that of higher-latitude high-latitude ectotherms, but a significant number of species ectotherms evolves to exceed the mean habitat temperature (20). exhibit delayed, or no change in, emergence. Here we present a This difference in thermal optima is well documented in the mechanistic theoretical framework that reconciles these disparate empirical literature (12, 15, 20, 21). observations and predicts population-level phenological patterns These latitudinal differences in thermal adaptation suggest based solely on data on temperature responses of the underly- that warming should have differential effects on lower-latitude ing life history traits. Our model, parameterized with data from and higher-latitude species. Because tropical ectotherms’ ther- insects at different latitudes, shows that peak abundance occurs mal optima coincide with the mean habitat temperature, an earlier in the year when warming increases the mean environ- increase in the mean temperature will cause these optima to mental temperature, but is delayed when warming increases the fall below the mean, with the result that the birth rate will be amplitude of seasonal fluctuations. We find that warming does below its optimal value, and the mortality rate will be higher, for not necessarily lead to a longer activity period in high-latitude most of the year. If the mean temperature increases to a level species because it elevates summer temperatures above the upper at which the mortality rate exceeds the birth rate, the species limit for reproduction and development. Our findings both con- will go extinct. In contrast, because temperate ectotherms’ ther- firm and confound expectations for ectotherm species affected mal optima exceed the mean habitat temperature, a moder- by climate warming: an increase in the mean temperature is more ate increase in the mean temperature is likely to be beneficial. detrimental to low-latitude species adapted to high mean tem- Hence, warming involving an increase in the mean tempera- peratures and low-amplitude seasonal fluctuations; an increase ture is likely to be more detrimental to tropical ectotherms in seasonal fluctuations is more detrimental to high-latitude than to temperate ectotherms. In contrast, warming involving an species adapted to low mean temperatures and high-amplitude increase in seasonal temperature fluctuations [e.g., an increase fluctuations. in warm and cold extremes (22)] is likely to be more detrimental to temperate ectotherms than to tropical ectotherms. This is because the concave-up nature of the mortality response dic- phenological shifts | stage-structured population model | variable tates that the average mortality rate must be higher when sea- developmental delay | climate change | life history traits sonal fluctuations are larger. At the same time, the concave- down nature of the birth rate response means a lower average emperature is the major abiotic factor that affects phenol- birth rate. If warming causes the magnitude of fluctuations to Togy, the seasonal timing of life history events. Climate warm- exceed the species’ response breadths, the average mortality rate ing is increasingly disrupting natural phenological patterns, but will exceed the average birth rate, predisposing the species to the consequences of such disruptions on population dynamics and species interactions are poorly understood (1, 2). Given Significance that ectotherms (microbes, plants, invertebrates, fish, amphibians, and reptiles) constitute the vast majority of biodiversity Changes in species’ phenology, the seasonal timing of life his- on the planet, elucidating the connection between their phe- tory events, constitute one of the most unambiguous conse- nology and population dynamics is a crucial research priority quences of climate warming and one of the least understood. (3–5). Ectotherm life history traits such as fecundity, develop- As our climate continues to warm and become more vari- ment, and survivorship exhibit plastic responses to tempera- able, we need theory that can explain the current phenologi- ture variation (6): when the temperature changes, the response cal patterns and predict future changes. We present a mathe- changes accordingly (7, 8). These responses are the result of tem- matical framework that translates temperature effects on the perature effects on the underlying biochemical processes (e.g., phenotypic traits of individual organisms to the population- reaction kinetics, enzyme inactivation, hormonal regulation) and level phenological patterns observed in ectotherms. It is suf- take qualitative forms (monotonic, unimodal) that are conserved ficiently mechanistic to yield accurate predictions and suffi- across ectotherm taxa (9–15). It is these trait responses that we ciently broad to apply across ectothermic taxa. Its power lies must focus on if we are to predict how warming-induced pheno- in generating predictions based solely on life history trait logical changes influence ectotherm population dynamics. responses to temperature and hence completely independent The nature of thermal adaptation dictates that trait responses of the population-level observations of phenological changes. be concave-up or concave-down functions of temperature (7, 16– 18). Based on Jensen’s inequality (19), we know that a tem- Author contributions: K.S. and P.A. designed research, performed research, analyzed perature response function that is concave up (e.g., exponen- data, and wrote the paper. tial) will yield a higher average response in a more variable The authors declare no conflict of interest. environment while a function that is concave down (e.g., uni- This article is a PNAS Direct Submission. modal) will yield a lower average response in a more variable Published under the PNAS license. environment. Per capita birth rate of all ectotherms exhibits Data deposition: Python code used to create and analyze the DDE model and R code a unimodal (i.e., concave-down) response to temperature. A used to fit thermal response curves and seasonal temperature fluctuations are available concave-down response means that higher-latitude (e.g., tem- at https://github.com/drscranto/doi.10.1073.PNAS.1711221114. perate) species, which have wider response breadths (greater 1K.S. and P.A. contributed equally to this work. plasticity), will exhibit lower average performance than lower- 2To whom correspondence should be addressed. Email: [email protected]. latitude (e.g., tropical) species, a necessary cost of thermal adap- This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. tation to higher-latitude environments. In this case stabilizing 1073/pnas.1711221114/-/DCSupplemental.

13212–13217 | PNAS | December 12, 2017 | vol. 114 | no. 50 www.pnas.org/cgi/doi/10.1073/pnas.1711221114 Downloaded by guest on September 28, 2021 Downloaded by guest on September 28, 2021 tr eed togyo h hra aiblt ftehabitat. the of variability thermal temper- the the habitat on mean responses; strongly the depends to trait relative ature these optima thermal in of pre- location generates differences regimes bio- quantitative thermal Adap- underlying in taxa. dictable variation ectotherm the latitudinal across to on conserved tation down), effects is concave processes, temperature or chemical from up concave arise (i.e., which temperature a to in responses even apparent be will it, abundances. itself, species’ since of abundance quantify census peak to regular the easy as relatively the well is contrast, as abundance In peak the shifts. of in phenological timing error timing of observation the quantification assess increases accurately empirical which to emergence, difficult sur- earliest is our it of on literature, the Based of phenology. that vey in to patterns shifts want abundance warming-induced we seasonal signify because species’ so in do signatures We identify Methods). peak devel- and and of (Materials reproduction for timing opment periods activity and as well abundance, as peak abundance, abundance, the annual quantifying dynamics, mean conse- population the on on changes well investigation such our of are quences focus emergence We 26). of 25, timing (4, documented the in changes Warming-induced Results ani- ectotherm on globe. warming the across climate mals of framework effects the predictive elucidating powerful Our for a data. temper- field provides to of approach responses independent trait-based trait completely phenological history therefore, based of and, life consequences, ature on duration dynamical information This and on population patterns. nature solely their the phenological and predict alters changes to warming us how allows of under- mechanistic standing a providing regimes, dif- under climate-warming change ferent themselves delays developmental how on focus and (Materials fluctuations seasonal of tempera- mean amplitude Methods the the in and increases both ture considering cold by continued dimensions warming two and of these events accommodates heat analysis tem- Our in mean (22). increase the extremes an that in also increase evidence but an increasing only perature not is involves There warming animals. climate ectotherm life of complex the Methods cycles captures and realistically that (Materials Methods) and dynamics Materials construct population to of equations model differential a delay all use to We applicable universally taxa. is ectotherm therefore thermody- and of 24) principles 16–18, (7, first namics from derived responses life trait of history descriptions mechanistic incorporates that dynamics lation responses trait dynamics. from population up scale to can developing that requires framework traits mathematical this history a Doing life outcomes. species’ population-level on into effects metric temperature the (fitness), translates rates rates mortality that growth capita and per birth species’ of determine responses to temperature the inter- with development to acts of need population response we temperature the phenology, the how at influences understand warming observe how we predict timing that To (e.g., level. breeding) events and delays phenological emergence developmental the of these determine temperature- is ultimately by It that characterized delays. (23) developmental cycles driven life complex exhibit tr fet ndvlpet Over development. on temper- effects considering of ature involves viability it because the proposition complicated on have trait can differen- latitudes. history different warming the inhabiting life species climate anticipate ectotherm to that of us effects nature allows tial temperature qualitative to the responses Thus, extinction. catnadAmarasekare and Scranton sntdaoe h ulttv aueo iehsoytrait history life of nature qualitative the above, noted As explicitly we that is framework our of strength particular A popu- ectotherm for framework theoretical a present we Here rdcigwrigefcso hnlg,hwvr samore a is however, phenology, on effects warming Predicting and IMtrasadMethods). and Materials SI 80% fmtza animals metazoan of and SI ntema eprtr assteotmmt alblwthe increase below fall An to 21). optimum the 15, causes (12, temperature temperature mean the habitat coincides in mean that reproduction the for with temperature optimal fluctuations, low-amplitude an and exhibit because temperatures ensue mean character- high outcomes regimes by thermal these ized to outset, adapted are the which at species, tropical noted As 4). the (Fig. in a earlier 2); occurs (Fig. also year which abundance, peak in and increase tem- abundance moderate The a 4). (3−6 temperature from and mean a benefit the 2 contrast, and (Figs. by y species, species, 80 perate tropical within the extinction causes in tem- increase abundance environmental annual mean lati- the mean of in function increase of a An perature as 2–4). increases opposite (Figs. the differences tude has that these and patterns fluctuations) of activity magnitude and in abundances species’ increase on effects vs. temperature mean 1, trop- (Fig. from and latitudes Methods, species temperate and insect and Hemipteran Mediterranean, from ical, data ther- with with response theory ectotherm model dynamics general mal of population this the dynamics link parameterize we To population data, warm- latitude. and of any inhabiting phenology effects species qualitative the the on adapta- predict ing thermal therefore in should variation tion, latitudinal reflect trait in that differences descriptions quantitative parameters with mechanistic combined responses, incorporates trait that of model dynamical A rprintl ihrltv otetmeauernei xeine nits in dis- experiences is it species environment. range thermal tropical temperature typical the the that to by relative Note exhibited high species. plasticity proportionately phenotypic each of by degree experienced the range variation the temperature regions line shaded seasonal vertical the The of and mechanistic Methods). temperature of and habitat mean Materials fits the (SI the represents data are the curves to functions the response and experiments, temperature from (Clavigralla tropical and shadabi histrionica), (Murgantia (A–C Mediterranean mortality lucorum), adult of (G–I rates tion capita per the of 1. Fig. motnl,w n httewrigrgm ices in (increase regime warming the that find we Importantly, aiue.I ahpnl h ice ersn epnedata response represent circles the panel, each In latitudes. ) hra epne flf itr ris (A–I traits. history life of responses Thermal o he eitrnisc pce rmtmeae(Apolygus temperate from species insect Hemipteran three for ) 3−6 PNAS 10 ◦ ◦ IMtrasadMethods). and Materials SI vr10ydcesspa bnac and abundance peak decreases y 100 over C nraecue eln u o extinction not but decline a causes increase C | eebr1,2017 12, December ◦ )wt nices nma annual mean in increase an with C) | ,brh(D–F birth ), o.114 vol. eprtr responses Temperature ) | o 50 no. ,admatura- and ), Materials | 10 13213 ◦ C

ECOLOGY mean temperature, the temperate species exhibits an advance in the timing of peak abundance and a longer activity period; when warming involves an increase in seasonal fluctuations, it exhibits a delay in the timing of peak abundance and a shorter activity period (Figs. 2 and 3). The tropical species, which is active year- round and exhibits delayed feedback cycles in its typical thermal environment (SI Materials and Methods), shows a decline in its reproductive period but no change in its developmental period under both warming regimes (Fig. 3). This is at least in part due to warming leading to faster development and hence a damping of delayed feedback cycles (SI Materials and Methods), resulting in a transition from abundances changing due to both intrinsic cycling and seasonal forcing to seasonal forcing only. A striking finding is that warming involving an increase in seasonal fluctuations, the evidence for which is accumulating (22), is more detrimental to high-latitude species already adapted to thermal regimes with large seasonal fluctuations. This finding runs counter to intuition, but can be explained in terms of the differential selection pressures faced by low- and high- latitude species in their respective thermal environments. Trop- ical species, because they experience thermal environments that deviate little from the high mean temperature, maximize their reproductive output at or near the mean temperature. The temperature at which fitness is maximized is therefore critically close Fig. 2. Effects of warming on phenological abundance patterns. Shown to the upper temperature limit for viability (12, 15). Any further are warming effects on mean annual abundance (Top row), peak abundance increase in the mean will push the species into a realm in which (Middle row), and timing of peak abundance (Bottom row) for insect species mortality exceeds reproduction (Fig. 1). High-latitude species, from different latitudes: temperate (Left column), Mediterranean (Center because they experience thermal environments that deviate column), and tropical (Right column). In each panel, solid bars represent greatly from the mean temperature, maximize fitness by evolving the effects of increasing the mean habitat temperature, and the hatched wider response breadths that match the temperature ranges they bars are effects of increasing the amplitude of seasonal fluctuations (3 ◦C increase, yellow; 6 ◦C increase, red; and 10 ◦C increase, maroon). The height of each bar depicts the proportional change in abundance or timing relative to that observed in the species’ ambient thermal environment (Materials and Methods). Extinct populations are marked E∗.

mean, causing a decrease in the birth rate relative to the mortality rate. Once the mean temperature increases to a level at which the birth rate falls below the mortality rate, extinction will occur. Hence, an increase in the mean temperature is more detrimental to tropical ectotherms than to temperate ectotherms. Temperate species, which are adapted to thermal regimes characterized by low mean temperatures and high-amplitude seasonal fluctuations, exhibit optimal temperatures for reproduction that are well above the mean habitat temperature (12, 15, 21). A moderate increase in warming is therefore beneficial (Fig. 2). What is unexpected is the difference in response when warming involves an increase in seasonal fluctuations. Now the tropical species exhibits an increase in both peak and mean abundance as warming increases, and there is no extinction at 10 ◦C of warming (Fig. 2). The temperate species, in contrast, exhibits a decline in both peak and mean abundance. Importantly, peak abundance is delayed rather than advanced (Fig. 2). These outcomes ensue because an increase in seasonal fluctuations causes reproduction to cease once the minimum and maximum temperatures exceed the species’ response breadth. Colder temperatures will have little net effect because they reduce both birth and mortality rates. However, warmer temperatures will decrease the Fig. 3. Effects of warming on activity periods. The reproductive activity birth rate and increase the mortality rate, leading to a strong neg- period (Top row) is the fraction of the year during which the per capita ative effect on the temperate species’ per capita growth rate and birth rate exceeds 10% of its physiological optimum. The developmental hence its abundance. The key insight to emerge from our analy- activity period (Bottom row) is the fraction of the year during which the per sis is that the antagonistic effects of birth and mortality rates at capita maturation rate exceeds 25% of its maximum. The height of each bar high temperatures make temperate ectotherms highly vulnerable depicts the proportional change in the activity period length relative to that observed in the ambient thermal environment for insect species from dif- to the predicted increase in the frequency and duration of heat ferent latitudes: temperate (Left column), Mediterranean (Center column), waves (22). and tropical (Right column). In each panel, solid bars represent the effects of Of note, warming-induced changes in the temperate species’ increasing the mean habitat temperature, and the hatched bars are effects timing of peak abundance are reflected in the changes in its activ- of increasing the amplitude of seasonal fluctuations (3 ◦C increase, yellow; ity period (Fig. 3). When warming involves an increase in the 6 ◦C increase, red; and 10 ◦C increase, maroon).

13214 | www.pnas.org/cgi/doi/10.1073/pnas.1711221114 Scranton and Amarasekare Downloaded by guest on September 28, 2021 Downloaded by guest on September 28, 2021 catnadAmarasekare and inhab- Scranton species between even vary magnitude to tend the which shifts, of phenological duration and causing is warming Climate Discussion (temperate temperature repro- habitat for mean 1). Fig. temperature the attribute; above optimum well an development duction and of attributes) sensitivity temperature (tropical high for breadth and response narrow reproduction thermal a and attributes: high- temperate individual-level the and combine of tropical responses the those trait Mediterranean to that The intermediate species. are low-latitude fact Mediterranean traits the its the on of responses from warming of arise 2 effects (Figs. species mixed entirely These almost 3). species response and species’ Mediterranean temperate the the warm- fluctuations, mimics When 3). seasonal and 2 increases (Figs. period ing activity the for at of (except length abundance extinction and peak ing) of for timing of (except terms in in abundance species mean tropical and the 10 peak to of similarly and changes terms tropical tempera- population the mean its the on ture, increases effects warming the When of species. mixture temperate a are dynamics lation tem- the with compared much is species species. fluctuations tropical perate seasonal the in to increase species detrimental an less why temperate is the This with 1). (Fig. compared range experience temperature seasonal typically more the they deal to great relative a exhibit plasticity possi- species phenotypic tropical been that quantifying is have to effects, approach not warming trait-based would mechanistic, our which without insight, ble crucial The species. pat- abundance the the respectively, and depict, pattern, 3 curves abundance under maroon terns typical and the red, depicts yellow, curve black The column). (Center species Mediterranean column), (Bottom in fluctuations sonal ( increase temperature an habitat involves mean warming the when abundance adult (log-transformed) 4. Fig. ne yteseisi hi yia hra niomns This environments. thermal typical experi- is their range in ratio temperature species the the by to fit- response enced relative the maximize quantifying fecundity to from for plasticity comes breadth insufficient this to is of them Confirmation there push ness. which will in fluctuations realm seasonal a in constraints. other increase phenotypic and further energetic of Any given limit achieve upper can the they at plasticity therefore are They experience. year. the of months hottest the not, during does abundance fluctuations in seasonal decline in large increase a an despite but 10 species by tropical the temperature of mean tion that the Note half species. in tropical and increase warmer the temperate in an the the season in rainy represents months the summer and panel species the Mediterranean to each corresponds in which year, region the of shaded The year. 100th h fet fwrigo h eierna pce’popu- species’ Mediterranean the on warming of effects The ◦ fwrig i.2 n iial otetmeaespecies temperate the to similarly and 2) Fig. warming; of C fet fwrigo nulaudnepten.Soni annual is Shown patterns. abundance annual on warming of Effects 1. 14 ◦ o h eprt pce and species temperate the for ,6 C, ◦ ,ad10 and C, o,dse ie)frtetmeaeseis(Left species temperate the for lines) dashed row, ◦ Top fwrig l iesre r o the for are series time All warming. of C o,sldlns n mltd fsea- of amplitude and lines) solid row, oun,adtoia pce ( species tropical and column), ◦ assdtriitcextinc- deterministic causes C 5.68 o h tropical the for 10 ◦ fwarm- of C Right itn fjuveniles of sisting Model. Population Methods and Materials dis- earth’s to the threatening constitute that are interactions biodiversity. of that webs enemies) phe- complex mutu- the observed natural rupt (e.g., widely and species the prey, interacting address alists, between climate- to extended mismatches of be nological kind easily both). or any can fluctuations or It mean and in increase mortality) (e.g., scenario warming vs. density- (e.g., fecundity response that mechanisms dependent temperature regulatory flexible population observed sufficiently and empirically is Finally, functions when incorporate species. developed can focal species have a it related we for framework distantly unavailable param- are the from be data can even response model our trait data 14), 13, with 10, (9, eterized values a of within range lie to narrow constrained of thermodynamically parameters are series the responses time because trait And, of data. independent his- population-level other completely life and therefore underlying and the traits of tory responses infor- temperature on about solely based mation patterns phenological influences climate how warming of Most ectotherms predictions niches. generates animal framework trophic our and importantly, to habitats, applicable latitudes, 13–15), different is inhabiting (9–11, framework taxa trait-based ectotherm our across conserved is temperature emergence. in species, delay high-elevation a exhibit in fluctuations, while seasonal advance higher-amplitude warming, experience which an under exhibit time emergence fluctuations, experience seasonal which lower-amplitude species, Low-elevational potentially species’ patterns. difference grasshopper of emergence This observations (27) period. al.’s an et activity Buckley experience explains shorter abun- and peak species a in delay and abundance such a dance show when peak they fluctuations, but seasonal in in period, increase advance activity mean an longer the a in show increase they fluctu- an in temperature, experience instance, increase species For an high-latitude dynamics. vs. when population temperature ectotherm mean influence the ations) in be (e.g., increase can regimes an warming differences different emer- such how that in understanding exhibit find by advance explained also We number 25). an substantial (2, exhibit a emergence delayed warming, species to of response in 27). number gence (25, recon- large also effects a we climate-warming so While of doing observations In from disparate nature. up in cile observed scale phenological are population-level that to the underlie patterns to way that traits processes history a life biochemical provide ectotherm the we on complex effects ectotherms, the model temperature of captures stage-structured realistically cycles a history that life into dynamics life population temperature of of understand- to descriptions mechanistic responses mechanistic a trait incorporating a such By discrep- present just which ing. we these provides by Here that Explaining mechanisms patterns. framework the 26). phenological understand 5, affects we temperature (2, that latitude requires ancies same the iting .Tenme fdy ttksfrjvnlst aueit dlsis adults function into the through mature variation temperature to seasonal incorporate in juveniles We seasonally. Meth- varies for and temperature takes when delay Materials year it developmental (SI the days (28) by of insects given number in The observation common ods). a the rate, on rates. operates birth competition, mortality intraspecific by and mediated maturation, dependence, Density birth, capita per functions temperature-dependent the where eas h ulttv aueo iehsoytatrsossto responses trait history life of nature qualitative the Because dA dt dJ dt = = PNAS B(T R emdltednmc fa cohr ouaincon- population ectotherm an of dynamics the model We J A n adults and (T (t ), (t B, | A(t − R eebr1,2017 12, December A ))A(t τ and , (t ))) A, ) − − τ D R D (t A A X ,wihdpnso h ieo the of time the on depends which ), (T (T (X T (t (t (t = ,wihtksasnsia form sinusoidal a takes which ), ))A(t − J , τ | eit epciey the respectively, depict, A) (t ), ))) o.114 vol. − D J (T | (t ))J o 50 no. (t ) | 13215 [1]

ECOLOGY (SI Materials and Methods). This formulation yields a set of differential The Harlequin bug (Murgantia histrionica, Hemiptera: Pentatomidae) is a equations with variable delays: delay differential equations (DDEs) (SI Mate- Mediterranean species from coastal southern California (39), and the pod- rials and Methods). sucking bug (Clavigralla shadabi, Hemiptera: Coreidae) is a tropical species from Benin (32). We use these particular species because they have the Temperature Responses of Life History Traits. Temperature responses of most complete data on the temperature responses of fecundity, matura- reproduction, development, and mortality are determined by temperature tion, and mortality rates, allowing us to fit the mechanistic temperature effects on the underlying biochemical processes [e.g., reaction kinetics, hor- response functions to the data and estimate key parameters (SI Materials monal regulation (7, 14, 16, 17, 24, 29)]. A plethora of empirical studies and Methods). We characterize the typical thermal environment of each (9–15) show that the qualitative nature of these responses (e.g., Gaussian, species by fitting a sinusoidal model of seasonal fluctuations to temperature exponential, left skewed) is conserved across ectotherm taxa (7, 24) (inverte- data collected from each species’ habitat: Langfang, China (39.32◦1609700 N) brates, amphibians, reptiles), because temperature effects on the biochem- for the green plant bug; Irvine, CA (33◦370800 N) for the Harlequin bug; ical processes that underlie these trait responses are themselves conserved and Cotonou, Benin (8◦200 N) for the pod-sucking bug (SI Materials (7, 24). and Methods). Temperature-driven biochemical rate processes (e.g., reaction kinetics and enzyme inactivation) yield trait responses at the phenotypic level that Model Analysis. We numerically integrate the DDE model (Eq. 1), using the are monotonic or left skewed. Temperature-driven biochemical regula- Python package PyDDE (40). We generate time series of abundances for each tory processes [e.g., neural and hormonal regulation (29–31)] yield trait species under the typical seasonal temperature regime and the two warm- responses that are unimodal and symmetric (e.g., Gaussian). The underlying regimes: increases in the mean habitat temperature and increases in the ing biochemical basis of trait responses allows us to mechanistically derive amplitude of fluctuations by 3 ◦C, 6 ◦C, and 10 ◦C. We incorporate climate the mathematical forms from first principles of thermodynamics. warming as a gradual increase in the mean temperature or amplitude of x The per capita mortality rate of all ectotherms exhibits a monotonic tem- fluctuations over a period of 100 y. The per day increase is given by 100×365 , perature response that is well described by the Boltzmann–Arrhenius func- where x◦ = 3, 6, 10 is total amount of warming after a century (SI Materials tion for reaction kinetics (7, 9, 10, 18). The maturation rate of ectotherms and Methods). exhibits a left-skewed temperature response (7, 16–18) that results from the We calculate metrics of population dynamics for each warming scenario: reduction in reaction rates at low and high temperature extremes due to mean annual abundance, peak annual abundance, and the timing of peak enzyme inactivation. This response is well described by a thermodynamic abundance (the day of the year on which the peak is observed). We do so rate process model (16, 17, 24). The per capita birth rate of most ectotherms using the time series for the 100th year, to ensure that the metrics reflect the exhibits a symmetric, unimodal temperature response (15, 32–35) that is well effects of a full century of warming. We also calculate the activity period of described by a Gaussian function. Based on empirical evidence (36, 37) we each species in terms of reproduction and development. We define repro- consider the temperature response of intraspecific competition to also be ductive activity period as the fraction of the year during which the per Gaussian, with the strongest competition occurring at the physiological opti- capita birth rate exceeds 10% of its physiological optimum and developmen- mum for the birth rate, at which temperature the demand for resources is tal activity period as the fraction of the year during which the maturation the greatest. rate exceeds 25% of its maximum in the ambient environment. We quantify the effects of warming by calculating the proportional change in each Study Species. We link the theory with data by parameterizing the popula- of the five metrics relative to the ambient thermal environment, i.e., the tion model (Eq. 1), using temperature response data from three Hemipteran difference between abundance (timing) in the species’ typical thermal envi- insect species from different latitudes. The green plant bug (Apolygus luco- ronment and a given warming scenario divided by the abundance (timing) rum, Hemiptera: Miridae) is a temperate species from northern China (38). in the typical thermal environment.

1. Miller-Rushing AJ, Høye TT, Inouye DW, Post E (2010) The effects of phenological 20. Amarasekare P, Johnson C (2017) Evolution of thermal reaction norms in seasonally mismatches on demography. Philos Trans R Soc Lond B Biol Sci 365:3177–3186. varying environments. Am Nat 189:E31–E45. 2. Forrest JR (2016) Complex responses of insect phenology to climate change. Curr Opin 21. Tewksbury JJ, Huey RB, Deutsch C (2008) Climate warming puts the heat on tropical Insect Sci 17:49–54. ectotherms. Science 320:1296–1297. 3. Walther GR, Post E, Convery P, Menzel A, Parmesan C (2002) Ecological responses to 22. Field C, et al., eds (2014) Climate Change 2014 - Impacts, Adaptation and Vulnerabil- recent climate change. Nature 416:389–395. ity: Part A: Global and Sectoral Aspects (Cambridge Univ Press, New York). 4. Root TL, et al. (2003) Fingerprints of global warming on wild animals and plants. 23. Minelli A, Fusco G (2010) Developmental plasticity and the evolution of animal com- Nature 421:57–60. plex life cycles. Philos Trans R Soc B Biol Sci 365:631–640. 5. Parmesan C (2007) Influences of species, latitudes and methodologies on estimates of 24. Ratkowsky D, Olley J, Ross T (2005) Unifying temperature effects on the growth rate phenological response to global warming. Glob Change Biol 13:1860–1872. of bacteria and the stability of globular proteins. J Theor Biol 233:351–362. 6. Roff D (1992) The Evolution of Life Histories: Theory and Analysis (Chapman and Hall, 25. Parmesan C (2006) Ecological and evolutionary responses to recent climate change. New York). Annu Rev Ecol Syst 37:637–669. 7. Van der Have TM, de Jong G (1996) Adult size in ectotherms: Temperature effects on 26. Diez JM, et al. (2012) Forecasting phenology: From species variability to community growth and differentiation. J Theor Biol 183:329–340. patterns. Ecol Lett 15:545–553. 8. Kingsolver J (2009) The well-temperatured biologist. Am Nat 174:755–768. 27. Buckley LB, Ehrenberger JC, Angilletta MJ (2015) Thermoregulatory behaviour limits 9. Gillooly JF, Brown JH, West GB, Savage VM, Charnov EL (2001) Effects of size and local adaptation of thermal niches and confers sensitivity to climate change. Funct temperature on metabolic rate. Science 293:2248–2251. Ecol 29:1038–1047. 10. Gillooly JF, et al. (2002) Effects of size and temperature on developmental time. 28. Murdoch W, Briggs CJ, Nisbet RM (2003) Consumer-Resource Dynamics (Princeton Nature 417:70–73. Univ Press, Princeton). 11. Huey R, Berrigan D (2001) Temperature, demography, and ectotherm fitness. Am Nat 29. Nijhout H (1994) Insect Hormones (Princeton Univ Press, Princeton). 158:204–210. 30. Hochachka P, Somero G (2002) Biochemical Adaption (Oxford Univ Press, Oxford). 12. Deutsch CJ, et al. (2008) Impacts of climate warming on terrestrial ectotherms across 31. Long M, Fee M (2008) Using temperature to analyse temporal dynamics in the song- latitude. Proc Natl Acad Sci USA 105:6668–6672. bird motor pathway. Nature 456:189–194. 13. Dell A, Pawar S, Savage VM (2011) Systematic variation in the temperature 32. Dreyer H, Baumgartner¨ J (1996) Temperature influence on cohort parameters and dependence of physiological and ecological traits. Proc Natl Acad Sci USA 108: demographic characteristics of the two cowpea coreids Clavigralla tomentosicollis 10591–10596. and C. shadabi. Entomol Exp Appl 78:201–213. 14. Kingsolver J, et al. (2011) Complex life cycles and the responses of insects to climate 33. Morgan D, Walters KFA, Aegerter JN (2001) Effect of temperature and cultivar on the change. Integr Comp Biol 51:719–732. pea aphid, Acyrthosiphon pisum (Hemiptera: Aphididae) life history. Bull Entomol Res 15. Amarasekare P, Savage V (2012) A mechanistic framework for elucidating the tem- 91:47–52. perature dependence of fitness. Am Nat 179:178–191. 34. Jandricic SE, Wraight SP, Bennett KC, Sanderson JP (2010) Developmental times and 16. Sharpe P, DeMichele D (1977) Reaction kinetics of poikilotherm development. J Theor life table statistics of Aulacorthum solani (Hemiptera: Aphididae) at six constant Biol 64:649–670. temperatures, with recommendations on the application of temperature-dependent 17. Schoolfield R, Sharpe J, Magnuson C (1981) Non-linear regression of biological development models. Environ Entomol 39:1631–1642. temperature-dependent rate models based on absolute reaction-rate theory. J Theor 35. Dannon EA, Tamo M, van Huis A, Dicke M (2010) Functional response and life history Biol 88:719–731. parameters of Apanteles taragamae, a larval parasitoid of Maruca vitrata. BioControl 18. Van der Have T (2002) A proximate model for thermal tolerance in ectotherms. Oikos 55:363–378. 98:141–155. 36. Johnson CA, et al. (2016) Effects of temperature and resource variation on insect 19. Jensen JLWV (1906) Sur les fonctions convexes et les in’egalites´ entre les valeurs population dynamics: The bordered plant bug as a case study. Funct Ecol 30: moyennes. Acta Mathematica 30:175–193. French. 1122–1131.

13216 | www.pnas.org/cgi/doi/10.1073/pnas.1711221114 Scranton and Amarasekare Downloaded by guest on September 28, 2021 Downloaded by guest on September 28, 2021 7 a Z ta.(03 fet ftmeaueaddniyo h otlt n repro- and mortality the on density and temperature of Effects (2013) al. et GZ, Gao 37. 8 uY uK ykusKG u 21)Tmeauedpnetlf itr fthe of history life Temperature-dependent (2010) Y Guo KAG, Wyckhuys K, Wu Y, Lu 38. 6 ono ,LwnI(96 h rwhrt of rate growth The (1946) I Lewin F, Johnson 46. single- tractable (2009) of M formulation Angilletta systematic The 45. (1983) J Lawton struc- R, physiologically Nisbet W, a Gurney Simplifying 44. (2008) T Kooten van for T, models Schellekens population of A, formulation deRoos systematic 43. The (1983) W Gurney RM, Nisbet 42. populations. structured for equations Delay-differential (1997) R Nisbet 41. (2015) Team Development Core distributed Python patchily a 40. on parasitoids competing of Coexistence (2000) P Amarasekare 39. 7 aaeV,Glol F rw H etG,CanvE 20)Efcso oysize body of Effects (2004) EL Charnov GB, West JH, Brown JF, Gillooly VM, Savage 47. 8 olrK ebaceJ(08 nettemltlrne hti h oeo ontogeny, of role the is What tolerance: thermal Insect (2008) J Terblanche K, Bowler 48. catnadAmarasekare and Scranton uto fcto pi pi gossypii. Aphis aphid cotton of duction Zool bug, plant green n coenzyme. and Oxford). Press, Univ structure. age incorporating models population species model. biomass stage-structured a 47–62. to model population tured duration. instar varying dynamically with insects 89–116. pp York), S, New Tuljapurkar Hall, eds and Systems, (Chapman Freshwater H and Caswell Terrestrial, Marine, in Models Population OR). Beaverton, Foundation, ware mechanisms. spatial vs. Local host: n eprtr npplto growth. population on temperature and gigadsenescence? and ageing 45:387–393. elCm Physiol Comp Cell J plgslucorum Apolygus hra dpin hoeia n miia Synthesis Empirical and Theoretical A Adaption: Thermal ilRev Biol 8:339–355. Ecology 28:47–75. (Meyer-D hnJAp Ecol Appl J Chin yhn344Documentation 3.4.4 Python mNat Am 81:1286–1296. .coli E. r Hmpea Miridae). (Hemiptera: ur) ¨ ho ou Biol Popul Theor 163:429–441. nrlto otmeaue quinine temperature, to relation in nmEcol Anim J 24:1300–1304. ho ou Biol Popul Theor 23:114–135. 52:479–495. plEntomol Appl Pto Soft- (Python Structured- (Oxford 73: 0 oeTa (2016) Team Core R community. host-multiparasitoid 60. a in dynamics An Spatial systems: (2000) P multi-trophic Amarasekare in 59. relationships Diversity-stability (2003) P com- Amarasekare in coexistence 58. species and variation temporal Trade-offs, (2007) Examining P change? Amarasekare climate 57. to respond species will How (2010) G Belovsky A, Laws 56. dynamics: population in resources and temperature of Interaction (1996) ME Ritchie 55. biol- on temperature of Effect (2006) V Baniameri Y, Fathipour AA, Talebi AA, of Zamani parameters history 54. life on temperatures constant of Inﬂuence (2007) parame- MN table Parajulee life 53. and development Temperature-dependent (2010) Z Weng Y, Hou 52. utility. and Concepts time: Thermal Carri (2005) 51. N Straalen van A, Honek D, Trudgill temperature maximum and 50. minimum the between Relationship (2009) al. et A, Dixon 49. miia exploration. empirical predation. intraguild with munities insect. herbivorous an on density population Entomol and temperature of effects the 79–89. pp Melbourne), Publishing, (CSIRO theory. of test A cucumber. of greenhouse parameters on growth population and ogy aphid, cotton the of ters parasitoids. egg in Biol Appl insects. in development for thresholds onainfrSaitclCmuig Vienna). Computing, Statistical for Foundation Ecol 69:201–213. r ,Bii 19)Eouino hra estvt fprstzto capacity parasitization of sensitivity thermal of Evolution (1997) G Boivin Y, ere ` coot nipae Octodonta 39:312–319. 146:1–14. rnir fPplto Ecology Population of Frontiers PNAS netn cotton. infesting gossypii, Aphis Evolution :ALnug n niomn o ttsia Computing Statistical for Environment and Language A R: nmEcol Anim J Clotr:Chrysomelidae). (Coleoptera: | plEntomol Appl J eebr1,2017 12, December 51:2028–2032. 72:713–724. Ecology uc Ecol Funct 130:453–460. pi gossypii Aphis 88:2720–2728. d ly ,Sepr ,D ar P Barro De A, Sheppard R, Floyd eds , 23:256–264. | nio Entomol Environ o.114 vol. nio Entomol Environ lvr(o. Aphididae) (Hom., Glover | o 50 no. 36:666–672. 39:1676–1684. | Environ Anim J 13217 Ann (R

ECOLOGY