Climate change has desynchronized insect and vegetation phenologies across Europe

Yanru Huang, Chaoyang Wu, Wenjiang Huang, Yang Liu, Man Qi, Jie Bai, Yingying Dong, Samuel J. L. Gascoigne, Philippe Ciais, Josep Peñuelas, Roberto Salguero‐GómezbioRxiv 2023 · DOI

TL;DR

Climate change is causing a growing mismatch between when plants bloom and when insects emerge across Europe. Vegetation is advancing its phenology four times faster than insects, leading to a significant desynchronization that could disrupt ecosystems and agriculture. This mismatch is driven by vegetation's higher sensitivity to climate change.

What Problem Does This Paper Address?

The paper tackles the critical issue of phenological mismatch – the decoupling of the timing of biological events between interacting species – driven by climate change. Specifically, it focuses on the synchrony between vegetation phenology (when plants leaf out) and insect phenology (when insects emerge or become active). This synchrony is vital for many ecological processes, including pollination and food webs. A mismatch can lead to a "trophic cascade," where disruptions at one level cascade through the ecosystem, potentially impacting food security and biodiversity. While previous studies have hinted at this problem, a comprehensive, large-scale understanding of the trends, drivers, and geographical variations across Europe was lacking. The authors aim to fill this gap by quantifying the extent of this desynchronization and identifying its underlying mechanisms.

Methods

The researchers employed a multi-faceted approach combining remote sensing and citizen science data:

  • Satellite-Derived Vegetation Phenology: Leaf unfolding dates (LUD) were extracted from the GIMMS NDVI3g dataset (1982-2015) using three different methods (dynamic-threshold, piecewise-logistic, and modified double-logistic functions) to ensure robustness. This involved processing satellite imagery to track vegetation greenness over time.
  • Citizen Science Insect Phenology: Insect occurrence dates (IOD) for 1,584 herbivorous insect species were gathered from crowdsourced data, primarily from the Global Biodiversity Information Facility (GBIF).
  • Quantile Regression: This statistical technique was used to model the trends in vegetation phenology, insect phenology, and their difference (Vegetation-Insect Phenological Difference, VID) over time and across geographical locations. This allowed for the estimation of rates of change.
  • Statistical Analysis: Various statistical methods, including Wilcoxon signed-rank tests, Kruskal-Wallis tests, partial correlation analysis, ridge regression, and random forests, were employed to:
  • Compare phenological shifts between different insect orders and life-cycle types (voltinism).
  • Analyze the spatial patterns of VID along latitudinal gradients.
  • Identify the environmental factors driving phenological mismatch.

Datasets

  • GIMMS NDVI3g dataset: A satellite-derived dataset providing Normalized Difference Vegetation Index (NDVI) data from the Advanced Very High-Resolution Radiometer (AVHRR) from 1982-2015. This was used for vegetation phenology.
  • Global Biodiversity Information Facility (GBIF): A global repository of biodiversity data, used to source occurrence records for 1,584 herbivorous insect species across Europe (1982-2015).
  • E-OBS dataset: A daily gridded dataset of terrestrial observations for Europe, providing meteorological data (temperature, precipitation, wind speed, humidity, solar radiation) from 1982-2015.
  • ERA5-Land reanalysis data set: A high-resolution dataset providing monthly soil temperature, soil-moisture content, and evapotranspiration data from 1982-2015.

Key Insights and Ideas

  • Vegetation is "outrunning" insects: The most striking finding is that vegetation phenology is advancing significantly faster than insect phenology. This suggests a fundamental shift in the timing of resource availability for insects.
  • Higher sensitivity of vegetation: Vegetation appears to be more sensitive to climate change than insects, leading to its more rapid phenological advance. This is a crucial driver of the observed mismatch.
  • Latitudinal gradients matter: The desynchronization is not uniform across Europe. Higher latitudes are experiencing a more pronounced decoupling, suggesting that regional climate change impacts and species' adaptive capacities vary geographically.
  • Insect traits influence adaptation: The ability of insects to keep pace with vegetation changes is not uniform. Different insect orders and life-cycle strategies (e.g., univoltine vs. multivoltine) exhibit varying degrees of adaptation, with some orders like Lepidoptera showing greater asynchrony.
  • Environmental factors are key drivers: Specific environmental factors, particularly temperature and solar radiation, are identified as major drivers of the phenological mismatch.

Key Results

  • Widespread Asynchrony: 61.2% of the studied vegetation and insect phenological patterns showed significant asynchrony.
  • Rate of Advance: Vegetation phenology advanced by 12 days between 1982-2015, while insect phenology advanced by only 3 days over the same period – a four-fold slower rate for insects.
  • VID Increase: 79.6% of insect phenological patterns showed an increase in VID (meaning greater asynchrony), with 61.2% being statistically significant. This translates to an exacerbation of phenological asynchrony at a rate of nearly three days per decade.
  • Latitudinal Differences: The rate of VID change was significantly higher in high latitudes (0.38 d/y) compared to low latitudes (-0.02 d/y).
  • Order-Specific Shifts: Lepidoptera showed the fastest rate of VID shift (0.34 d/y), indicating greater asynchrony, while Coleoptera adjusted their phenology more rapidly, leading to fewer significant VID shifts.
  • Voltinism Impact: Univoltine insects showed faster VID shifts than multivoltine insects, suggesting that multivoltine insects might adapt more quickly due to having more generations to adjust their life history strategies.

Limitations and Future Work

  • Data Limitations: The authors acknowledge potential biases in citizen science data due to limitations in spatial and temporal coverage, as well as species representation. They also note the need to be cautious with analytical outcomes due to inevitable taxonomic and biogeographical sampling biases.
  • Focus on Adult Phenology: The study focused on the mean date of adult insect occurrence, acknowledging that different developmental stages (egg, larval, pupal) might have distinct phenological responses.
  • Future Research Directions: The paper emphasizes the need for future research to:
  • Focus on the complex interactions within interspecific phenological networks.
  • Investigate the impacts of both biotic and abiotic factors on phenological changes.
  • Pay close attention to high-latitude areas where phenological changes are more pronounced.
  • Develop large-scale, integrative approaches to provide a robust scientific basis for ecological conservation, resource management, and policy development.

Summary updated

Abstract

Summary Climate change is drastically affecting the phenologies of species worldwide, including shifts in development 1–3 . The potential impact of climate change on the synchronicity of plant-insect phenology is particularly concerning since the stability of pollination networks and food chains depends on their tight temporal dependencies 4–6 . Furthermore, the recently reported “insect armageddon” 7 makes it urgent to establish trends and identify primary drivers of plant-insect phenological synchrony. Here, coupling high-resolution remote sensing and citizen science data across Europe over 34 years, we examine the differences in occurrence dates of 1,584 herbivorous insects across four orders and the corresponding dates of leaf unfolding. We find that 61.2% of the vegetation and insect phenologies have become highly asynchronous, with vegetation phenology advancing four-fold faster than insect phenology. These trends were modulated by insects’ life-cycles and taxonomic order. A primary driver of this phenological mismatch is the higher sensitivity of vegetation phenology than insect phenology to climate, which has prevented insects from matching the pace of plant phenological advance in the growing season. Our analyses provide a unique continental overview and underlying mechanisms of the asynchronicity between vegetation and insect phenologies, thus enhancing our ability to predict and manage its potential cascading ecological effects.
BibTeX
@misc{huang2023climate,
  author = {Yanru Huang and Chaoyang Wu and Wenjiang Huang and Yang Liu and Man Qi and Jie Bai and Yingying Dong and Samuel J. L. Gascoigne and Philippe Ciais and Josep Peñuelas and Roberto Salguero‐Gómez},
  title = {{Climate change has desynchronized insect and vegetation phenologies across Europe}},
  year = {2023},
  doi = {10.1101/2023.12.11.571152},
  howpublished = {bioRxiv},
}
Figure