Estimating Density Dependence, Environmental Variance, and Long-Term Selection on a Stage-Structured Life History.

AbstractA method for analyzing long-term demographic data on density-dependent stage-structured populations in a stochastic environment is derived to facilitate comparison of populations and species with different life histories. We assume that a weighted sum of stage abundances, N, exerts density dependence on stage-specific vital rates of survival and reproduction and that N has a small or moderate coefficient of variation. The dynamics of N are approximated as a univariate stochastic process governed by three key parameters: the density-independent growth rate, the net density dependence, and environmental variance in the life history. We show how to estimate the relative weighs of stages in N and the key parameters. Life history evolution represents a stochastic maximization of a simple function of the key parameters. The long-term selection gradient on the life history can be expressed as a vector of sensitivities of this function with respect to density-independent, density-dependent, and stochastic components of the vital rates. To illustrate the method, we analyze 38 years of demographic data on a great tit population, estimating the key parameters, which accurately predict the observed mean, coefficient of variation, and fluctuation rate of N; we also evaluate the long-term selection gradient on the life history.

A high-density SNP chip for genotyping great tit (Parus major) populations and its application to studying the genetic architecture of exploration behaviour.

High-density SNP microarrays ("SNP chips") are a rapid, accurate and efficient method for genotyping several hundred thousand polymorphisms in large numbers of individuals. While SNP chips are routinely used in human genetics and in animal and plant breeding, they are less widely used in evolutionary and ecological research. In this article, we describe the development and application of a high-density Affymetrix Axiom chip with around 500,000 SNPs, designed to perform genomics studies of great tit (Parus major) populations. We demonstrate that the per-SNP genotype error rate is well below 1% and that the chip can also be used to identify structural or copy number variation. The chip is used to explore the genetic architecture of exploration behaviour (EB), a personality trait that has been widely studied in great tits and other species. No SNPs reached genomewide significance, including at DRD4, a candidate gene. However, EB is heritable and appears to have a polygenic architecture. Researchers developing similar SNP chips may note: (i) SNPs previously typed on alternative platforms are more likely to be converted to working assays; (ii) detecting SNPs by more than one pipeline, and in independent data sets, ensures a high proportion of working assays; (iii) allele frequency ascertainment bias is minimized by performing SNP discovery in individuals from multiple populations; and (iv) samples with the lowest call rates tend to also have the greatest genotyping error rates.

Heritable variation in maternally derived yolk androgens, thyroid hormones and immune factors.

Maternal reproductive investment can critically influence offspring phenotype, and thus these maternal effects are expected to be under strong natural selection. Knowledge on the extent of heritable variation in the physiological mechanisms underlying maternal effects is however limited. In birds, resource allocation to eggs is a key mechanism for mothers to affect their offspring and different components of the egg may or may not be independently adjusted. We studied the heritability of egg components and their genetic and phenotypic covariation in great tits (Parus major), using captive-bred full siblings of wild origin. Egg mass, testosterone (T) and androstenedione (A4) hormone concentrations showed moderate heritability, in agreement with earlier findings. Interestingly, yolk triiodothyronine hormone (T3), but not its precursor, thyroxine hormone (T4), concentration was heritable. An immune factor, albumen lysozyme, showed moderate heritability, but yolk immunoglobulins (IgY) did not. The genetic correlation estimates were moderate but statistically nonsignificant; a trend for a positive genetic correlation was found between A4 and egg mass, T and lysozyme and IgY and lysozyme, respectively. Interestingly, phenotypic correlations were found only between A4 and T, and T4 and T3, respectively. Given that these egg components are associated with fitness-related traits in the offspring (and mother), and that we show that some components are heritable, it opens the possibility that natural selection may shape the rate and direction of phenotypic change via egg composition.

Disrupted seasonal biology impacts health, food security and ecosystems.

The rhythm of life on earth is shaped by seasonal changes in the environment. Plants and animals show profound annual cycles in physiology, health, morphology, behaviour and demography in response to environmental cues. Seasonal biology impacts ecosystems and agriculture, with consequences for humans and biodiversity. Human populations show robust annual rhythms in health and well-being, and the birth month can have lasting effects that persist throughout life. This review emphasizes the need for a better understanding of seasonal biology against the backdrop of its rapidly progressing disruption through climate change, human lifestyles and other anthropogenic impact. Climate change is modifying annual rhythms to which numerous organisms have adapted, with potential consequences for industries relating to health, ecosystems and food security. Disconcertingly, human lifestyles under artificial conditions of eternal summer provide the most extreme example for disconnect from natural seasons, making humans vulnerable to increased morbidity and mortality. In this review, we introduce scenarios of seasonal disruption, highlight key aspects of seasonal biology and summarize from biomedical, anthropological, veterinary, agricultural and environmental perspectives the recent evidence for seasonal desynchronization between environmental factors and internal rhythms. Because annual rhythms are pervasive across biological systems, they provide a common framework for trans-disciplinary research.

Replicated high-density genetic maps of two great tit populations reveal fine-scale genomic departures from sex-equal recombination rates.

Linking variation in quantitative traits to variation in the genome is an important, but challenging task in the study of life-history evolution. Linkage maps provide a valuable tool for the unravelling of such trait-gene associations. Moreover, they give insight into recombination landscapes and between-species karyotype evolution. Here we used genotype data, generated from a 10k single-nucleotide polymorphism (SNP) chip, of over 2000 individuals to produce high-density linkage maps of the great tit (Parus major), a passerine bird that serves as a model species for ecological and evolutionary questions. We created independent maps from two distinct populations: a captive F2-cross from The Netherlands (NL) and a wild population from the United Kingdom (UK). The two maps contained 6554 SNPs in 32 linkage groups, spanning 2010 cM and 1917 cM for the NL and UK populations, respectively, and were similar in size and marker order. Subtle levels of heterochiasmy within and between chromosomes were remarkably consistent between the populations, suggesting that the local departures from sex-equal recombination rates have evolved. This key and surprising result would have been impossible to detect if only one population was mapped. A comparison with zebra finch Taeniopygia guttata, chicken Gallus gallus and the green anole lizard Anolis carolinensis genomes provided further insight into the evolution of avian karyotypes.

Heritability of gonad size varies across season in a wild songbird.

Many organisms advance their seasonal reproduction in response to global warming. In birds, which regress their gonads to a nonfunctional state each winter, these shifts are ultimately constrained by the time required for gonadal development in spring. Gonadal development is photoperiodically controlled and shows limited phenotypic plasticity in relation to environmental factors, such as temperature. Heritable variation in the time required for full gonadal maturation to be completed, based on both onset and speed of development and resulting in seasonally different gonad sizes among individuals, is thus a crucial prerequisite for an adaptive advancement of seasonal reproduction in response to changing temperatures. We measured seasonal gonadal development in climate-controlled aviaries for 144 great tit (Parus major) pairs, which consisted of siblings obtained as whole broods from the wild. We show that the extent of ovarian follicle development (follicle size) in early spring is highly heritable (h(2) = 0.73) in females, but found no heritability of the extent of testis development in males. However, heritability in females decreased as spring advanced, caused by an increase in environmental variance and a decrease in additive genetic variation. This low heritability of the variation in a physiological mechanism underlying reproductive timing at the time of selection may hamper genetic adaptation to climate change, a key insight as this great tit population is currently under directional selection for advanced egg-laying.

Genetic background, and not ontogenetic effects, affects avian seasonal timing of reproduction.

Avian seasonal timing is a life-history trait with important fitness consequences and which is currently under directional selection due to climate change. To predict micro-evolution in this trait, it is crucial to properly estimate its heritability. Heritabilities are often estimated from pedigreed wild populations. As these are observational data, it leaves the possibility that the resemblance between related individuals is not due to shared genes but to ontogenetic effects; when the environment for the offspring provided by early laying pairs differs from that by late pairs and the laying dates of these offspring when they reproduce themselves is affected by this environment, this may lead to inflated heritability estimates. Using simulation studies, we first tested whether and how much such an early environmental effect can inflate heritability estimates from animal models, and we showed that pedigree structure determines by how much early environmental effects inflate heritability estimates. We then used data from a wild population of great tits (Parus major) to compare laying dates of females born early in the season in first broods and from sisters born much later, in second broods. These birds are raised under very different environmental conditions but have the same genetic background. The laying dates of first and second brood offspring do not differ when they reproduce themselves, clearly showing that ontogenetic effects are very small and hence, family resemblance in timing is due to genes. This finding is essential for the interpretation of the heritabilities reported from wild populations and for predicting micro-evolution in response to climate change.

APOE1 mutation in a patient with type III hyperlipoproteinaemia: detailed genetic analysis required.

We present the case of a patient with clinical features of familial dysbetalipoproteinaemia (FD) including high levels of total cholesterol, hypertriglyceridaemia and the presence of palmar xanthomas. Whereas genotype analysis identified the APOE3E3 isoform, sequence analysis revealed the presence of one APOE1 allele due to a mutation, p.Lys164Glu, which leads to loss of function of apolipoprotein E (ApoE), a rare cause of dominant FD.

Mutations in LPL, APOC2, APOA5, GPIHBP1 and LMF1 in patients with severe hypertriglyceridaemia.

OBJECTIVES: The severe forms of hypertriglyceridaemia (HTG) are caused by mutations in genes that lead to the loss of function of lipoprotein lipase (LPL). In most patients with severe HTG (TG > 10 mmol L(-1) ), it is a challenge to define the underlying cause. We investigated the molecular basis of severe HTG in patients referred to the Lipid Clinic at the Academic Medical Center Amsterdam. METHODS: The coding regions of LPL, APOC2, APOA5 and two novel genes, lipase maturation factor 1 (LMF1) and GPI-anchored high-density lipoprotein (HDL)-binding protein 1 (GPIHBP1), were sequenced in 86 patients with type 1 and type 5 HTG and 327 controls. RESULTS: In 46 patients (54%), rare DNA sequence variants were identified, comprising variants in LPL (n = 19), APOC2 (n = 1), APOA5 (n = 2), GPIHBP1 (n = 3) and LMF1 (n = 8). In 22 patients (26%), only common variants in LPL (p.Asp36Asn, p.Asn318Ser and p.Ser474Ter) and APOA5 (p.Ser19Trp) could be identified, whereas no mutations were found in 18 patients (21%). In vitro validation revealed that the mutations in LMF1 were not associated with compromised LPL function. Consistent with this, five of the eight LMF1 variants were also found in controls and therefore cannot account for the observed phenotype. CONCLUSIONS: The prevalence of mutations in LPL was 34% and mostly restricted to patients with type 1 HTG. Mutations in GPIHBP1 (n = 3), APOC2 (n = 1) and APOA5 (n = 2) were rare but the associated clinical phenotype was severe. Routine sequencing of candidate genes in severe HTG has improved our understanding of the molecular basis of this phenotype associated with acute pancreatitis and may help to guide future individualized therapeutic strategies.

Hepatic steatosis does not cause insulin resistance in people with familial hypobetalipoproteinaemia.

AIMS/HYPOTHESIS: Hepatic steatosis is strongly associated with hepatic and whole-body insulin resistance. It has proved difficult to determine whether hepatic steatosis itself is a direct cause of insulin resistance. In patients with familial hypobetalipoproteinaemia (FHBL), hepatic steatosis is a direct consequence of impaired hepatic VLDL excretion, independently of metabolic derangements. Thus, patients with FHBL provide a unique opportunity to investigate the relation between increased liver fat and insulin sensitivity. METHODS: We included seven male participants with FHBL and seven healthy matched controls. Intrahepatic triacylglycerol content and intramyocellular lipid content were measured using localised proton magnetic resonance spectroscopy (¹H-MRS). A two-step hyperinsulinaemic-euglycaemic clamp, using stable isotopes, was assessed to determine hepatic and peripheral insulin sensitivity. RESULTS: ¹H-MRS showed moderate to severe hepatic steatosis in patients with FHBL. Basal endogenous glucose production (EGP) and glucose levels did not differ between the two groups, whereas insulin levels tended to be higher in patients compared with controls. Insulin-mediated suppression of EGP during lower dose insulin infusion and insulin-mediated peripheral glucose uptake during higher dose insulin infusion were comparable between FHBL participants and controls. Baseline fatty acids and lipolysis (glycerol turnover) at baseline and during the clamp did not differ between groups. CONCLUSIONS/INTERPRETATION: In spite of moderate to severe hepatic steatosis, people with FHBL do not display a reduction in hepatic or peripheral insulin sensitivity compared with healthy matched controls. These results indicate that hepatic steatosis per se is not a causal factor leading to insulin resistance. TRIAL REGISTRATION: ISRCTN35161775.

Characterisation of non-obese diabetic patients with marked insulin resistance identifies a novel familial partial lipodystrophy-associated PPARγ mutation (Y151C).

AIMS/HYPOTHESIS: Familial partial lipodystrophy (FPLD) is a rare metabolic disorder with clinical features that may not be readily recognised. As FPLD patients require a specific therapeutic approach, early identification is warranted. In the present study we aimed to identify cases of FPLD among non-obese patients with type 2 diabetes mellitus and marked insulin resistance. METHODS: We searched the databases of three diabetic outpatient clinics for patients with marked insulin resistance, arbitrarily defined as the use of ≥100 U insulin/day, and BMI ≤ 27 kg/m(2). In all patients, metabolic variables and anthropomorphic measurements were evaluated and DNA was sequenced for mutations in the genes encoding lamin A/C (LMNA), peroxisome proliferator-activated receptor γ (PPARγ) and cell death-inducing DFFA-like effector c (CIDEC). RESULTS: Out of 5,221 diabetic individuals, 24 patients fulfilled all criteria. Twelve patients were willing to participate, of whom five showed clinical features of lipodystrophy. In three of these patients the clinical diagnosis of FPLD was confirmed by the presence of mutations in LMNA or PPARG; one patient harboured a novel heterozygous mutation (Y151C) in PPARG. The Y151C mutant displayed impaired DNA-binding capacity and hence reduced transcriptional activity compared with wild-type PPARγ. Dominant-negative activity was absent. CONCLUSION/INTERPRETATION: The combination of BMI ≤ 27 kg/m(2) and the use of >100 U insulin/day increases the chance of identifying lipodystrophy. Thus careful assessment of clinical features of FPLD should be considered in these patients, allowing earlier therapeutic interventions.

Triglycerides and cardiovascular risk.

In 1996 a meta-analysis was published showing that an increase in plasma triglyceride (TG) levels was associated with an increase in CHD risk, even after adjustment for high density lipoprotein cholesterol (HDL-C) levels. Very recently, two studies were published that further extent the early observation and showed the importance of nonfasting plasma triglyceride (TG) levels in the prediction of risk on coronary heart disease (CHD). In the current review we have summarized all available evidence obtained in clinical studies showing that treatment guidelines should reconsider to include nonfasting TG in their risk assessments as nonfasting TG levels may better predict CVD risk.

Nutritional interventions for reducing morbidity and mortality in people with HIV.

BACKGROUND: Adequate nutrition promotes and maintains optimal immune function. Dietary support may, therefore, improve clinical outcomes in HIV-infected individuals by reducing the incidence of HIV-associated complications and attenuating progression of HIV disease, thereby improving quality of life and ultimately reducing disease-related mortality. OBJECTIVES: To evaluate the effectiveness of various macronutrient interventions, such as a balanced diet or high protein, high carbohydrate, or high fat diets given orally, in reducing morbidity and mortality in adults and children living with HIV infection. SEARCH STRATEGY: We searched CENTRAL (up to March 2006), MEDLINE (1966 to March 2006), EMBASE (1988 to March 2006), LILACS (up to March 2006), and AIDSearch (up to March 2006). We also scanned reference lists of articles and contacted authors of relevant studies and other researchers. SELECTION CRITERIA: Randomised controlled trials evaluating the effectiveness of macronutrient interventions compared with no nutritional supplements or placebo in the management of adults and children infected with HIV. DATA COLLECTION AND ANALYSIS: Three reviewers independently applied study selection criteria, assessed study quality, and extracted data. Effects were assessed using weighted mean difference and 95% confidence intervals. Meta-analysis employed a fixed-effect model, except when the chi-square test for heterogeneity was significant (p<0.1). MAIN RESULTS: Eight trials (with a total of 486 participants), met the criteria for inclusion in our review. None of the studies reported on mortality, morbidity, or disease progression. Overall, macronutrient supplementation (with or without nutritional counselling) significantly improved energy intake (5 trials; n=254; WMD 367 kcal.day-1; 95% CI: 217 to 516) and protein intake (3 trials; n=128; WMD 17 g.day-1; 95% CI: 8 to 26) compared with no nutritional supplementation or placebo. There was no evidence of an effect on body weight (8 trials; n=423; WMD 0.24 kg; 95% CI: -0.6 to 1.1), fat mass (6 trials; n=305; WMD -0.73 kg; 95% CI: -1.83 to 0.37), fat-free mass (5 trials; n=311; WMD 0 kg; 95% CI: -2.3 to 2.4) or CD4 count (6 trials; n=271; WMD 0.23 cells.mm-3; 95% CI: -40.2 to 40.6). AUTHORS' CONCLUSIONS: Given the current evidence base, which is limited to a few small trials in high-income countries, no firm conclusions can be drawn about the effects of macronutrient supplementation on morbidity and mortality in people living with HIV.

Micronutrient supplementation in children and adults with HIV infection.

BACKGROUND: The scale and impact of the HIV/AIDS pandemic has made the search for simple, affordable, safe, and effective public health interventions all the more urgent. Micronutrient supplements hold the promise of meeting these criteria, but their widespread use needs to be based on sound scientific evidence of effectiveness and safety. OBJECTIVES: To assess whether micronutrient supplements are effective in reducing morbidity and mortality in adults and children with HIV infection. SEARCH STRATEGY: The Cochrane Library (CENTRAL), EMBASE, MEDLINE, AIDSearch, CINAHL, and conference proceedings were searched, and pharmaceutical manufacturers and researchers in the field were contacted to locate any ongoing or unpublished trials. SELECTION CRITERIA: Randomised controlled trials comparing the effects of micronutrient supplements (vitamins, trace elements, and combinations of these) with placebo or no treatment on mortality and morbidity in HIV-infected individuals. DATA COLLECTION AND ANALYSIS: Two reviewers independently appraised trial quality and extracted data. Study authors were contacted for additional data where necessary. A meta-analysis was not deemed appropriate due to significant heterogeneity between trials. MAIN RESULTS: Fifteen trials were included. Six trials comparing vitamin A/beta-carotene with placebo in adults failed to show any effects on mortality, morbidity, CD4 and CD8 counts, or on viral load. Four trials of other micronutrients in adults did not affect overall mortality, although there was a reduction in mortality in a low CD4 subgroup. In a large Tanzanian trial in pregnant and lactating women, daily multivitamin supplementation was associated with a number of benefits to both mothers and children: a reduction in maternal mortality from AIDS-related causes; a reduced risk of progression to stage four disease; fewer adverse pregnancy outcomes; less diarrhoeal morbidity; and a reduction in early-child mortality among immunologically- and nutritionally-compromised women. Vitamin A alone reduced all-cause mortality and improved growth in a small sub-group of HIV-infected children in one hospital-based trial, and reduced diarrhoea-associated morbidity in a small HIV-infected sub-group of infants in another trial. AUTHORS' CONCLUSIONS: There is no conclusive evidence at present to show that micronutrient supplementation effectively reduces morbidity and mortality among HIV-infected adults. It is reasonable to support the current WHO recommendations to promote and support adequate dietary intake of micronutrients at RDA levels wherever possible. There is evidence of benefit of vitamin A supplementation in children. The long-term clinical benefits, adverse effects, and optimal formulation of micronutrient supplements require further investigation.

The short-term effects of anti-tuberculosis therapy on plasma pyridoxine levels in patients with pulmonary tuberculosis.

Plasma levels of pyridoxal phosphate (PLP) were determined in 20 patients with pulmonary tuberculosis before and after one week of drug therapy including isoniazid. At baseline, body mass index and PLP levels were reduced in 10 and 18 patients, respectively. After 7 days of therapy, PLP levels decreased (P < 0.001) in all but one subject who inadvertently received pyridoxine supplementation. The decreased PLP levels occurred despite a significant improvement in the acute phase response (increased albumin [P < 0.001] and reduced C-reactive protein levels [P < 0.01]). This study indicates the need for possible routine pyridoxine supplementation in patients with newly diagnosed tuberculosis.

Plasma vitamin A and zinc levels in HIV-infected adults in Cape Town, South Africa.

A cross-sectional study of 132 adults attending an HIV clinic in Cape Town, South Africa, was conducted to determine predictors of low plasma vitamin A and Zn levels. No patients were on antiretroviral therapy. The possible confounding effect of the acute-phase response was controlled by including C-reactive protein levels in multivariate analysis and by excluding active opportunistic infections. Retinol levels were low (<1.05 micromol/l) in 39 % of patients with early disease (WHO clinical stages I and II) compared with 48 and 79 % of patients with WHO stage III and IV respectively (P<0.01). Plasma Zn levels were low (<10.7 micromol/l) in 20 % of patients with early disease v. 36 and 45 % with stage III and IV disease respectively (P<0.05). C-reactive protein levels were normal in 63 % of subjects. Weak, positive associations were found between CD4+ lymphocyte count and plasma levels of retinol (r 0.27; 95 % CI 0.1, 0.43) and Zn (r 0.31; 95 % CI 0.25, 0.46). Multivariate analysis showed the following independent predictors of low retinol levels: WHO stage IV (odds ratio 3.4; 95 % CI 2.1, 5.7) and body weight (odds ratio per 5 kg decrease 1.15; 95 % CI, 1.08, 1.25), while only body weight was significantly associated with low Zn levels (OR per 5 kg decrease 1.19; 95 % CI 1.09, 1.30). CD4+ lymphocyte count <200/microl was not significantly associated with either low retinol or Zn levels. In resource-poor settings, simple clinical features (advanced disease and/or weight loss) are associated with lowered blood concentrations of vitamin A and/or Zn. The clinical significance of low plasma retinol and/or Zn levels is unclear and more research is required to establish the role of multiple micronutrient intervention strategies in HIV disease.

The costs of egg production and incubation in great tits (Parus major).

The costs of egg production and incubation may have a crucial effect on avian reproductive decisions, such as clutch size and the timing of reproduction. We carried out a brood-size enlargement experiment on the great tit (Parus major), in which the birds had to lay and incubate extra eggs (full costs), only incubate extra eggs (free eggs) or did not pay any extra cost (free chicks) in obtaining a larger brood. We used female fitness (half the recruits produced plus female survival) as a fitness measure because it is the female which pays the costs of egg production and incubation, and because clutch size is under female control. Female fitness decreased with increasing costs (fitness of free chicks females is higher than that of free eggs females which is higher than that of full costs females). These fitness differences were due to differences in female survival rather than in the number of recruits produced. This is the first time that the costs of egg production and incubation have been estimated using such a complete fitness measure, including, as our measure does, the local survival to the following year of both the female and her offspring. Our results emphasize that reproductive decisions cannot be understood without taking egg production and incubation costs into account.

Adjustment to climate change is constrained by arrival date in a long-distance migrant bird.

Spring temperatures in temperate regions have increased over the past 20 years, and many organisms have responded to this increase by advancing the date of their growth and reproduction. Here we show that adaptation to climate change in a long-distance migrant is constrained by the timing of its migratory journey. For long-distance migrants climate change may advance the phenology of their breeding areas, but the timing of some species' spring migration relies on endogenous rhythms that are not affected by climate change. Thus, the spring migration of these species will not advance even though they need to arrive earlier on their breeding grounds to breed at the appropriate time. We show that the migratory pied flycatcher Ficedula hypoleuca has advanced its laying date over the past 20 years. This temporal shift has been insufficient, however, as indicated by increased selection for earlier breeding over the same period. The shift is hampered by its spring arrival date, which has not advanced. Some of the numerous long-distance migrants will suffer from climate change, because either their migration strategy is unaffected by climate change, or the climate in breeding and wintering areas are changing at different speeds, preventing adequate adaptation.

Warmer springs disrupt the synchrony of oak and winter moth phenology.

Spring temperatures have increased over the past 25 years, to which a wide variety of organisms have responded. The outstanding question is whether these responses match the temperature-induced shift of the selection pressures acting on these organisms. Organisms have evolved response mechanisms that are only adaptive given the existing relationship between the cues organisms use and the selection pressures acting on them. Global warming may disrupt ecosystem interactions because it alters these relationships and micro-evolution may be slow in tracking these changes. In particular, such shifts have serious consequences for ecosystem functioning for the tight multitrophic interactions involved in the timing of reproduction and growth. We determined the response of winter moth (Operophtera brumata) egg hatching and oak (Quercus robur) bud burst to temperature, a system with strong selection on synchronization. We show that there has been poor synchrony in recent warm springs, which is due to an increase in spring temperatures without a decrease in the incidence of freezing spells in winter. This is a clear warning that such changes in temperature patterns may affect ecosystem interactions more strongly than changes in mean temperature.