ABSTRACT
Insects are the most species rich terrestrial taxa and their interactions with plants together comprise the majority of Earth’s biodiversity. Recent reports provide evidence for large climate driven declines in the abundance and diversity of insects- some suggest up to 40% of temperate species will go extinct in the next two decades. However, it has been less clear if substantive losses are occurring in intact low-latitude forests, where insect diversity is the highest. Further, the implications of reported declines on ecosystem services provided by these insects are speculated but not quantified. Using 22 years of plant-caterpillar-parasitoid data, we document the loss of entire genera of Lepidoptera and declines in parasitism in a protected tropical forest. Forty-percent of Lepidoptera genera studied are declining in frequency and reductions in parasitism events suggest a 30% drop over the next 100 years. These declines in parasitism represent a reduction in an important ecosystem service: enemy control of primary consumers. Reported reductions in diversity, density and parasitism appear to be partly driven by a changing climate and weather anomalies. Our results demonstrate the ecological costs of climate driven insect declines in intact tropical forests and support predictions that specialized parasitoids are likely to suffer. The consequences of these changes are in many cases irreversible, and declines in parasitism will likely negatively impact surrounding agriculture. The loss of important tropical taxa and erosion of associated ecosystem function underlines the apparent threat to global insect diversity and provides additional impetus for research on tropical diversity.
Significance Statement Recent reports suggest that there are substantial declines in insect abundance and diversity globally. The extent and severity of declines and their impact on ecosystem function remain to be quantified for low latitudes where the majority of diversity resides. From 22-years of data, we document losses of entire genera of lepidopteran larvae and declines in parasitoids of those larvae. Parasitism by host-specialists declined the most and is partly driven by increases in extreme precipitation events. These results show the ecological costs of current declines and support predictions that specialized insects are more vulnerable to changes in climate. Reported reductions in parasitism are likely to have economic consequences for tropical countries where conversion of intact forest to agriculture is highest.
INTRODUCTION
The impacts of global change are multi-faceted and ubiquitous (1) with ecological and evolutionary consequences (2) spanning aquatic and terrestrial ecosystems, and a wide diversity of taxa and species interactions (3). Much of global change research has focused on effects on single trophic levels, and despite an increased emphasis on interaction diversity in ecology (4), relatively few studies have linked climatic variability to interaction diversity, ecosystem stability, and services of parasitoids. Past studies have also been geographically and taxonomically biased towards temperate ecosystems (5) and the subset of tropical studies published are focused on vertebrates and focal tree species (6). Thus, although it has been clear for some time that a sixth mass extinction event is underway (7), only recently have studies attempted to document declines in insect abundance and diversity in intact tropical forests by examining broad guilds. Threats to insect diversity include climate change, habitat loss, fragmentation, invasive species, pesticides, and pollutants (9–11). The magnitude of these global change forces and the levels of ecosystem resilience vary considerably across biogeographic regions, and there has been a long-standing expectation that tropical communities are more stable. However, increases in extreme weather events will have complex and large effects on lowland tropical communities (12,13) where plant-insect food webs may be particularly sensitive because of highly-specialized trophic relationships relative to interactions at higher latitudes (14). Here we contribute to understanding species declines and loss of biological interactions in a protected and well-studied tropical wet forest.
RESULTS
Our study area is La Selva Biological Research Station, Heredia Costa Rica (10° 26′ N, 83° 59′ W), a ~1600-ha patch of forest bordered by agriculture (Fig. 1A). We used data from 1997-2018 to examine changes in taxonomic diversity among larval Lepidoptera (“caterpillars”) and associated parasitic Hymenoptera and Diptera (“parasitoids”). Our data reveal that declines in insect richness (Fig.1) and diversity (Fig. S1-S3) are widespread across the two consumer trophic levels. Extrapolation of estimated declines to the 1600 ha of La Selva yielded estimates for the number of species that have either been lost from the forest since the start of the study or have been reduced to sufficiently low density that they are no longer detected (which likely amounts to effective extirpation from the perspective of ecological interactions). We estimate 1056 fewer herbivore species (with 95% Bayesian credible intervals from 2112 to 352), and 704 fewer parasitoid species (from 1056 to 352). For the herbivores, for which we have the most data, we additionally used the first 5 years of data to estimate a baseline diversity (Chao estimator) from which the losses represent 38.8% reduction (with credible intervals from 77.6% to 12.9%).
Braulio Carillo National Forest (A.l) and surrounding areas, including La Selva (A.2), which has experienced declines in caterpillar diversity (β=−0.03, 95% credible intervals (CI) [−0.06,−0.01]) (B), associated parasitoid diversity (β =−0.02, [−0.03,−0.01]) (C) and interaction diversity (β =−0.07, [−0.13,−0.02]) (D) over the past 22 years (1997-2018). A large adjacent banana plantation is indicated by dashed white lines (A.3). Dotted lines on graphs are best fit lines from Bayesian regression, with 95% credible intervals in gray.
In addition to declines in caterpillar diversity, frequencies of encounter for entire genera of caterpillars are decreasing: out of the 64 genera studied, 41% (26 genera) have an 80% probability of being in decline (i.e. at least 80% of the mass of the Bayesian posterior distributions were less than zero for year coefficients in regressions for each of these genera) (Fig. 2, Table S1). These dramatic declines suggest that many caterpillars at La Selva will be losers and few will be winners in response to global change (15), resulting in an overall reduction in their roles as herbivores and food for other animals. Compelling examples of winners and losers include the success of Eucereon, which includes outbreak species (16) and the failure of formerly common genera such as Emesis (Fig.S4). Notably, declines in insectivorous vertebrate predators, including bats and birds within and near La Selva, have already been attributed to reductions in arthropod prey (17–19).
A) Point estimates for beta coefficients and associated 80% credible intervals (CI) for 64 genera that comprise a subset of all genera collected that met criteria for this analysis. Genus names are listed on the left margin and probabilities of a negative effect are on the right margin. Units of the year coefficient are the natural log of frequency per year. B) Frequency (untransformed) across years for select genera and representative larval and adult images.
One of the consequences of extirpation is the loss of interspecific interactions, which underlie ecosystem stability and ecosystem services (20), but questions about loss of interaction diversity are largely absent from global change literature, due to a dearth of quantitative empirical data (21–24). Along with taxonomic declines, interaction diversity at La Selva is decreasing: assemblages today have approximately 2,464 fewer unique interactions (30.9% reduction) than networks of interactions 22-year ago (Fig. 3A-B, Table S2, S3). Herbivore-enemy interactions were disproportionately affected, with over 77% of connections disappearing between herbivores and parasitoids when comparing networks of interactions in the first and last five years of the study. Reduction in species (25) and interaction diversity (26) can cause reduced ecosystem function via loss of functional redundancy, with likely cascading effects on natural biological control, pollination, plant diversity, primary productivity, and nutrient cycling. In fact, parasitism frequency, an important measure of natural biological control, also decreased over time (Fig.S5). Estimated parasitism decline was 3% per decade which represents a ~6.6% decline during the study period. Further, the probability of negative slope for overall parasitism across time was ~92%. Such declines in parasitism represent an impressive 30% drop in parasitism over the next 100 years and loss of a key ecosystem service that prevents damaging outbreaks of herbivorous insects(4). Losses of species and trophic interactions of this magnitude are particularly relevant in areas with intensified agriculture, where the global economic contribution of biological control is now estimated at $1.56 trillion per year (27, 28). Parasitoids are essential for biological control in banana, and over 10,000 ha of land surrounding La Selva are banana plantations with one of the largest plantations situated <3km from La Selva (Fig. 1A).
Tri-trophic networks illustrate host plants (green), caterpillars (blue), and associated parasitoids (yellow) for the first (A) and last 5 years (B). Nodes represent families within each trophic level and are grouped by suborder (Heterocera: light blue and Rhopalocera: dark blue) and order (Hymenoptera: light yellow and Diptera: mustard yellow) then ranked by node degree. Edge thickness represents relative link weights. Parasitism levels declined over time (C) (Hymenoptera: −0.001 [−0.004,0.002] and Diptera: −0.00007 [−0.003, 0.003] and these declines were associated with climatic changes (D). The structural equation model (good fit to the data: (χ2 =0.887, p=0.346, df=1) illustrating these associations estimates effects of time on climate and time and climate on percent parasitism. Models for species richness and interaction diversity yielded similar coefficients as did models for parasitism with more climate variables (Fig. S10-S11). Path coefficients are standardized and width of arrows are scaled based on magnitude of path coefficients. Standard errors are reported in brackets. Arrows represent positive associations and lines with circles represent negative associations. Parasitoid illustrations by M.L.F.
Environmental changes in and around the forest include land use, as well as annual temperature and precipitation which are both increasing at La Selva (Fig. S6-S9, Tables S4, S5). In the last five years, precipitation anomalies have been larger than previous years and the number of positive temperature anomalies are increasing, with the greatest Tmin and Tmax anomalies on record occurring during the study period. Structural equation models (SEM) provided support for the hypotheses that precipitation anomalies and their one year time lag are among the most important factors causing lower parasitism frequency, following predictions of Stireman et al. 2005 (29) (Fig. 3D). Declines in richness appear to be caused by changes in multiple climate variables (Fig.S10-11). With the exception of the effect of minimum temperature on caterpillar richness (Fig. S11A), time had the strongest direct negative association with richness and parasitism frequency compared to other predictors in all SEM models, suggesting that other unmeasured global change variables also contribute to insect declines at La Selva.
DISCUSSION
Declines in populations of plants and animals, extinctions, and loss of ecosystem function are defining features of the Anthropocene (7). From a general Bayesian perspective in which new results are used to update prior knowledge (30), additional corroborations of these Anthropocene-associated losses are useful in that they provide more precise estimates of decline probability for specific taxa, regions and ecosystems. Although insect declines have been the subject of recent high profile studies (8,31), the taxonomic and geographic breadth of the phenomenon is not without controversy (32) and reports have been rare from the planet’s most species-rich ecosystems. Thus we suggest that the results reported here strengthen the growing probability that insects are facing what indeed may be a global crisis. The hard work that still faces ecologists is to try to figure out which traits and habitats most expose species to risk, while the challenges for taxonomists and natural historians are to discover and describe new species and interactions before they disappear. All scientists should be considering how to use existing data to focus on the most imperiled taxa, ecosystems, and biogeographic regions. Tropical wet forests are clearly one biome requiring more precise estimates of species declines and a better understanding of determinants of these declines. For La Selva, we found that climate change is causing declines in species and entire genera of herbivores as well specialized parasitoids. Although such multi-trophic connections are not frequently studied in the context of global change, if results such as ours are widespread, then cascading results to other guilds and trophic levels can be expected (20) and warrant immediate concern and management effort.
METHODS
Study sites and sample methods
We collected plant-caterpillar-parasitoid interaction data within La Selva Biological Research Station located in Heredia Province, Costa Rica (10° 26′ N 83° 59′ W). La Selva is a 1600-ha patch of protected lowland tropical forest on the eastern Caribbean slope of the Cordillera Central connected via a corridor to the Braulio Carllio National Forest. Seasonality is marked by a wet season generally from May to December and a brief dry season beginning January to April. Peak rainfall occurs in June-July and March is the peak dry month. Samples were collected as a larger rearing program cataloguing plant-herbivore-parasitoid associations across the Americas17,25 from 1995 to present. We limited our results to records starting in 1997 up to 2018, and we excluded 2014 and 2016 because sampling days did not meet our minimum criteria of 20 sampling days/year. We sampled externally feeding immature Lepidoptera (caterpillars) from their host plant (including shelter builders) and reared them to adult moths or parasitoids(5). Caterpillars were located opportunistically by visual inspection along trail transects (distance varies between 50-3000m and select transects are continuously sampled across years), or in 10m diameter plots (149 plots total) by staff scientists, graduate students, parataxonomists and teams of Earthwatch volunteers and students. Due to varied sampling methods across years we weighted observed values by sampling effort. Sampling effort was calculated as the number of volunteer and staff days of sampling multiplied by the average area in square meters covered by each person in a 10-day sampling period (4000 m2). Hence, observed diversity and frequency is presented and analyzed in models as species equivalents or frequencies per hectare per year. We excluded Eois (Geometridae) and Quadrus (Hesperiidae) from all analyses because these focal genera present a bias in the rearing dataset due to focused collection for ancillary studies.
Rearing Methods & Processing Data
For our ongoing interaction diversity survey, collected larvae are given a unique voucher code that associate them with their host plant species. Caterpillars are reared individually in plastic containers or bags with a sample of hostplant. Species identifications are made initially by parataxonomists to lowest taxonomic level or morphospecies and verified by taxonomic experts or by referencing voucher specimens and image libraries. Some morphospecies are confirmed using mtDNA COI sequences, others by examining a mix of morphological characters, and others using genomic data. For the remaining species without morphospecies designations we assign morphotypes based on feeding relationships - consumers from the same family utilizing the same host family are designated a unique morphotype. This method is likely a conservative means to assigning species names, especially for tropical species31. Voucher specimens are sent to collaborating institutions including universities and museums (see www.caterpillars.org for a list of participating institutions).
Patterns in Diversity, Parasitism, Climate Variables
Abundance & Diversity
Taxonomic and interaction annual frequencies were obtained to understand patterns in diversity and abundance across time. Species-level and interaction frequencies were used for diversity analyses. Due to the rarity of species level data, we aggregated the data to examine genus-level frequencies of Lepidoptera to evaluate herbivore abundance patterns. Annual frequencies were calculated for each Lepidoptera genera. We limited genus-level abundance data analyses to genera with ≥5 years of data and sample points extending to 2010. Results are reported for the 64 Lepidoptera genera that met criteria. To obtain values of interaction diversity, we modified a community matrix such that rows were comprised of years and columns the unique interactions. Interactions were comprised of bi-trophic (plant-herbivore) and tri-trophic (plant-herbivore-enemy) interactions such that matrix cells represent annual frequencies of those interactions.
Diversity was calculated as Hill numbers, and values were interpreted as interaction or species equivalents(6). Hill numbers vary as a function of the parameter q and indicates the sensitivity of the index to rare species such that q=0, q=1 and q=2 are equivalent to species richness, Shannon’s diversity, and Simpsons diversity, respectively. We used functions provided in Chao et al. (2015)(7) to calculate Hill numbers. Results for q=0 are reported in text and q=1 & q=2 in the supplemental information. To obtain estimated percent herbivore loss we differenced total diversity (an estimate based on averaged Chao estimates of the first 5 years of data) and the observed species decline extrapolated from beta coefficients of the models predicting diversity across years.
Climate Variables
Climate variables were calculated as annual means of daily precipitation and average, minimum and maximum temperatures. We used meteorological data acquired from weather stations within La Selva from 1983-2018(8). Temperature is reported as degrees Celsius (°C) and precipitation millimeters (mm). To examine effects of extreme weather events and climate variability on patterns of diversity, we calculated anomalies and the coefficients of variation (CV) for each precipitation and temperature variable. Precipitation anomalies were calculated as the sum of daily values exceeding 2.5 standard deviations (sd) of the annual mean. Similarly, for temperature anomalies we used 2sd. The coefficient of variation was calculated as the ratio of standard deviation to the annual mean. We used simple linear regression to evaluate patterns among each climate variable across time and with respect to each season in the supplemental figures.
Evaluating Patterns in Network Structural Properties
We pooled interaction data to the family level for the first (1997-2001) and last (2012-2018) five years of collection to illustrate changes in tri-trophic network structure. For each network illustration we calculated node degrees and relative edge weights and reported link and node richness for each trophic level (Table S2).
Parasitism Frequency
Percent parasitism was calculated as the ratio of parasitism events to the sum of successfully emerged Lepidopteran adults and parasitized individuals for each month from 1997-2018. We examined monthly trends across time to account for intra-annual and seasonal variation in tropical population dynamics. Excluded from analyses were months with zero parasitism, zero eclosed caterpillars and those months without sufficient sample in the denominator (sum of adult eclosion and parasitism events). Sufficient sample was deemed as values exceeding the 1st quantile (Q1) of the distribution of denominators (IQR=12-103).
Statistical Models
We used Bayesian linear models to estimate coefficients associated with change through time on Lepidoptera, parasitoid and interaction diversity and parasitism frequency. This model was applied to total parasitism and separately for specialized (Hymenoptera) and non-specialized (Diptera) orders. Models were fit in JAGS (version 3.2.0) run with R and the rjags package(9) using (for each analysis) two Markov chains and 1,000,000 steps each; performance was assessed through examination of chain histories (burnin was not required), effective sample sizes and the Gelman and Rubin convergence diagnostic(10). Response variables were modeled as normal distributions with means dependent on an intercept plus predictor variables (either year alone, or year plus climatic variables), and vague or minimally-influential priors as follows: priors on beta coefficients (for year and climatic variables) were normal distributions with mean of zero and precision of 0.01 (variance = 100); priors on precisions were modeled as gamma distributions with rate = 0.1 and shape = 0.1. All data was z transformed prior to analysis.
An additional hierarchical model (with vague priors as already described) was used to estimate change across years in the frequency at which individual Lepidoptera genera were observed, with the year coefficients (and intercepts) estimated for each genus separately (as the lower level in the hierarchy) and simultaneously across all genera (the response variable for this analysis was log transformed prior to z transformation). For all models (simple and hierarchical) we retained point estimates from posterior distributions for beta coefficients, as well as 95% credible intervals for the diversity models and 80% intervals for the hierarchical model. We used the more liberal calculation of intervals for the latter in the interest of minimizing type II error in a situation involving the decline of entire genera (i.e., we would rather risk the possibility of erroneously inferring decline as opposed to mistakenly concluding that a declining taxon is stable). As a complementary measure of confidence not dependent on an arbitrary cutoff for importance, we calculated (for the beta coefficients estimated for each genus) the fraction of the posterior distribution less than zero, which can be interpreted as the probability that a genus has been observed with decreasing frequency over time.
We used Structural Equation Modeling (SEM) to test causal hypotheses that evaluated the effect of climate and time on taxonomic and interaction richness and parasitism. We used the global estimation method in the lavaan package v.0.6-3(11) in R v 3.5.3 to generate 3 models. Models evaluated casual relationships among caterpillar, parasitoid, interaction richness or parasitism and climate variables. Climate variables included: Tmin and Tmax and their anomalies and precipitation anomalies. A priori predictions facilitated the test of causal hypotheses that parasitism was modeled by precipitation and its one year lag. Models were assessed using the χ2 and model comparison using Akaike information criterion (AIC). We reported standardized path coefficients and illustrated the SEM results in a path diagram.
Funding
This study was supported by the NSF DGE-1447692, NSF DEB-1442103, Experiment.com.
Author Contributions
D.M.S., L.A.D., H.G.L. conceived project and supervised field work. D.M.S., L.A.D., M.F. analyzed data and wrote manuscript. Parasitoid illustrations by M.L.F.
Competing interests
None declared.
Data and materials availability
All data needed to evaluate the conclusions in the paper are available upon request.
ACKOWLEDMENTS
We thank Earthwatch Institute and volunteers, UNR Plant-Insect Group, G. Gentry, J.O. Stireman, S. Shaw, J. Whitfield, J. Miller, J. Brown, J. Elliot, R. Parry, L. Richards, A. Smilanich, T. Davis, Z. Bousum, B. Carranza, D. Brenes & O. Vargas for substantive contributions to this work.
Footnotes
We updated former document with a more detailed abstract. We added a significance statement. We changed reported effect sizes from per year % loss to percent decline per decade. We added two additional tables to the supplemental section (TableS5 and Table S6).
