10K outplant growth analysis

Outplant Growth and Survival Analysis - 10K Seed Hardening Project

Author: AS Huffmyer
Date: October 16, 2025
Assessment Date: August 20, 2025

Project Summary

The 10K Seed Hardening Project investigates the effects of different hardening treatments on Pacific oyster (Crassostrea gigas) seed resilience and performance. This project aims to evaluate whether pre-conditioning juvenile oysters through various stress exposures (temperature, salinity, immune challenge) can improve their survival and growth when outplanted to field conditions.

Experimental Design

Approximately 10,000 Pacific oyster seed were divided into five treatment groups:

  1. Control - No hardening treatment
  2. 35C - Temperature hardening (35°C exposure)
  3. FW - Fresh water (salinity stress) hardening
  4. FW 35C - Combined fresh water and temperature hardening
  5. Immune - Immune challenge hardening with PolyIC exposure

Each treatment group consisted of 6 replicate bags, with 150 oysters initially placed in each bag. Oysters were outplanted to field conditions and monitored for growth and survival over time.

Growth Analysis

Methods

Growth measurements were conducted on August 20, 2025. For each bag, multiple oysters were measured for:

  • Length (mm)
  • Width (mm)

Since depth measurements were not available, volume was estimated using a polynomial regression model trained on data from the same source population at Goose Point, where length, width, and depth were all measured. The volume calculation used the ellipsoid formula:

Volume = (4/3) × π × (length/2) × (width/2) × (depth/2)

The polynomial model was:

volume ~ poly(length.mm, 2) + poly(width.mm, 2) + length.mm:width.mm

Data Quality Control

Several quality control steps were implemented:

  • Bag 12 was identified as an outlier (much larger than all others) and removed from analysis
  • Observations resulting in negative predicted volumes were removed (n<10)
  • Upper outliers (volume > 125,000 mm³) were filtered out
  • Standardized residuals were checked for outliers (threshold = 3 standard deviations)

Statistical Analysis

A linear mixed effects model was used to test for treatment effects on growth:

model <- lmer(sqrt(Predicted_Volume_Poly) ~ treatment + (1|purple.tag:treatment), data=data)

The model included:

  • Fixed effect: Treatment group
  • Random effect: Bag nested within treatment (purple.tag:treatment)
  • Transformation: Square root transformation of predicted volume to improve normality

Results

Growth by Bag Figure 1: Predicted oyster volume (mm³) by bag, colored by treatment. Each point represents an individual oyster measurement.

Growth by Treatment Figure 2: Predicted oyster volume (mm³) by treatment group. No significant differences were observed among treatments.

Key Finding: There were no significant differences in growth among treatment groups (p > 0.05). All hardening treatments resulted in similar final sizes when outplanted to field conditions.

The lack of treatment effects suggests that:

  1. The hardening treatments did not negatively impact growth capacity
  2. Field conditions may override any growth advantages conferred by pre-conditioning
  3. All oysters, regardless of treatment, were able to grow similarly under field conditions

Interpretation and Conclusions

Growth Performance

The absence of treatment effects on growth is both informative and encouraging:

  1. No negative impacts: None of the hardening treatments resulted in reduced growth, suggesting that the stress exposures did not cause lasting growth depression or metabolic costs that persisted into the field phase.

  2. Field conditions dominate: Once deployed to the field, environmental conditions (food availability, water quality, temperature) appear to be the primary drivers of growth, superseding any metabolic or physiological differences induced by pre-conditioning.

  3. Phenotypic plasticity: Pacific oysters demonstrated growth plasticity, with all groups converging on similar growth trajectories regardless of their conditioning history.

Next steps

  • Normalize size to initial outplant sizes (pending AH image analysis)
  • Conduct winter/spring assessments in 2026

Data and Code Availability

All data and code are available on our repo.

Written on October 16, 2025