Disease surveillance by artificial intelligence links eelgrass wasting disease to ocean warming across latitudes

Abstract

Ocean warming endangers coastal ecosystems through increased risk of infectious disease, yet detection, surveillance, and forecasting of marine diseases remain limited. Eelgrass (Zostera marina) meadows provide essential coastal habitat and are vulnerable to a temperature-sensitive wasting disease caused by the protist Labyrinthula zosterae. We assessed wasting disease sensitivity to warming temperatures across a 3500 km study range by combining long-term satellite remote sensing of ocean temperature with field surveys from 32 meadows along the Pacific coast of North America in 2019. Between 11% and 99% of plants were infected in individual meadows, with up to 35% of plant tissue damaged. Disease prevalence was 3Â higher in locations with warm temperature anomalies in summer, indicating that the risk of wasting disease will increase with climate warming throughout the geographic range for eelgrass. Large-scale surveys were made possible for the first time by the Eelgrass Lesion Image Segmentation Application, an artificial intelligence (AI) system that quantifies eelgrass wasting disease 5000Â faster and with comparable accuracy to a human expert. This study highlights the value of AI in marine biological observing specifically for detecting widespread climate-driven disease outbreaks.Disease outbreaks frequently cause rapid declines of host populations, transforming community structure and ecosystem functioning. Outbreaks that affect foundation or keystone species have particularly widespread and long-lasting consequences. Prominent examples include the ecological extinction of chestnut trees in eastern U.S. forests from chestnut blight (Ellison et al. 2005); decimation of at least 20 species of sea-stars in the eastern Pacific due to sea-star wasting disease

Predictable Changes in Eelgrass Microbiomes with Increasing Wasting Disease Prevalence across 23° Latitude in the Northeastern Pacific

“…Recently, Groner et al ( 23 ) linked warmer temperatures during the eelgrass growing season to increased prevalence of wasting disease in the San Juan Islands, Washington. Aoki et al ( 27 ) observed similar links between warm thermal anomalies and increased wasting disease prevalence across 23° latitude in the Northeast Pacific. Although experimental studies show that warmer temperatures enhance L. zosterae growth and abundance ( 28 ), we do not yet know how eelgrass microbiomes vary along temperature gradients and with fluctuating disease prevalence in natural meadows.…”

“…Warming is hypothesized to disrupt beneficial microbiomes and their functions within marine habitat forming species, like eelgrass, and may contribute to disease outbreaks (2,6,83). Indeed, warm thermal anomalies are linked to higher prevalence of wasting disease along our latitudinal gradient (27) and at our San Juan Island, WA sites (23). This may result from cumulative or interactive effects of warming on plant or pathogen physiology (59,83,84).…”

“…We similarly observed effects of SST on the day of sampling and 30-day maximum and minimum SSTs on seawater microbiomes. Thus, warm temperatures during the summer growing season (30 days prior to sampling) alter phyllosphere microbiomes, in addition to enhancing the prevalence of wasting disease lesions ( 27 ). However, the microbial taxa that are correlated with warmer 30-day maximum temperatures differ from those that are correlated with wasting disease prevalence, except for one family, Amoebophilaceae, that is negatively correlated with both warmer temperatures and higher disease prevalence.…”

“…We also quantified shoot density by placing quadrats at 4 m intervals from meter 4 to 16 on the transect. Further sampling details can be found in Aoki et al ( 27 ), including the use of an image classification machine learning tool, Eelgrass Lesion Image Segmentation Application (EeLISA [ 90 ]), to quantify the presence or absence of lesions and severity of lesions on leaves. Briefly, this entailed collection of the third-youngest leaf, careful removal of epiphytes, and high-resolution scanning to generate images of each leaf.…”

“…Third-youngest leaves occur outside the leaf sheath in OR, unlike other regions, and are covered with epiphytes, precluding our ability to determine lesion status in the field. Because 2nd- and 3rd-youngest leaves in OR did not differ in lesion coverage ( 27 ), we decided to include 2nd-youngest leaf samples from OR in our region-wide analyses.…”