Using a mechanistic framework to model the density of an aquatic parasite Ceratonova shasta

Abstract

Ceratonova shasta is a myxozoan parasite endemic to the Pacific Northwest of North America that is linked to low survival rates of juvenile salmonids in some watersheds such as the Klamath River basin. The density of C. shasta actinospores in the water column is typically highest in the spring (March-June), and directly influences infection rates for outmigrating juvenile salmonids. Current management approaches require quantities of C. shasta density to assess disease risk and estimate survival of juvenile salmonids. Therefore, we developed a model to simulate the density of waterborne C. shasta actinospores using a mechanistic framework based on abiotic drivers and informed by empirical data. The model quantified factors that describe the key features of parasite abundance during the period of juvenile salmon outmigration, including the week of initial detection (onset), seasonal pattern of spore density, and peak density of C. shasta. Spore onset was simulated by a bio-physical degree-day model using the timing of adult salmon spawning and accumulation of thermal units for parasite development. Normalized spore density was simulated by a quadratic regression model based on a parabolic thermal response with river water temperature. Peak spore density was simulated based on retained explanatory variables in a generalized linear model that included the prevalence of infection in hatchery-origin Chinook juveniles the previous year and the occurrence of flushing flows (≥171 m³/s). The final model performed well, closely matched the initial detections (onset) of spores, and explained inter-annual variations for most water years. Our C. shasta model has direct applications as a management tool to assess the impact of proposed flow regimes on the parasite, and it can be used for projecting the effects of alternative water management scenarios on disease-induced mortality of juvenile salmonids such as with an altered water temperature regime or with dam removal.

Keywords: Actinospore; Disease; Klamath River; Management; Myxozoan; Predictive model; Salmonids.

Figure 1. Map of the Klamath River basin in the United States of America.

The Klamath River (bold line) runs westward from Upper Klamath Lake to the Pacific Ocean. The map indicates the location of six hydroelectric dams (bars), the infectious zone (shading), the Beaver Creek monitoring site (KBC), and three annelid monitoring sites (circle markers).

Figure 2. The lifecycle of Ceratonova shasta alternates between two spore stages and two hosts, cycling through each phase during different seasons of the year.

Actinospore stages infect vertebrate, salmonid hosts (Oncorhynchus spp.) and myxospore stages infect invertebrate, fabricid annelid hosts (Manayunkia occidentalis). Directional orbits of the lifecycle are represented by gray lines and arrows. In the spring (year = i), mature actinospores are released into the water column and encounter outmigrating juvenile salmonid hosts (left loop). Infected juveniles release myxospores that are consumed by annelid hosts. Infected annelids release actinospores in the fall that encounter adult salmonids returning to the river to spawn (right loop). After swimming upriver to spawn, the fish die and release myxospores that can infect annelids. Actinospore development in annelids takes more time in the colder water temperatures of the winter than during the summer. Infected annelids release mature actinospores in spring of the next year (year = i+1), and the cycle continues.

Figure 3. Characteristics of Ceratonova shasta densities observed in spring.

Spore densities (spores/L) in the spring include (A) the onset of spore release from annelid hosts, (B) the general seasonal pattern of spore density (gray line), and (C) the peak magnitude of spore density. To highlight these features, the black markers shown here are observed values of spore density measured at the Beaver Creek monitoring site in 2014.

Figure 4. Schematic showing monitoring data included in the final spore model for Ceratonova shasta that predicts spore density.

Modules predict the onset of spore release from annelid hosts (A), spore density increase with water temperature (B), and peak magnitude of spore density (C). Inputs for the modules are biological monitoring data (adult Chinook Salmon returning to spawn, i.e., escapement, density of spores in water samples, the prevalence of infection of hatchery-origin juvenile Chinook from the previous spring [POI_i−1]) and environmental monitoring data (river discharge and water temperature) data shown in the top gray shaded bars. The bottom-left graph (from Fig. 3) shows an example of the three components of spore density that this model predicts (monitoring data from 2014). Predictions generated by each module are validated against observations of spore densities from monitoring data (Figs. 5–7). Arrow direction indicates the order of calculations and analyses.

Figure 5. Goodness-of-fit for each module and the final spore model that simulate characteristics of Ceratonova shasta spore density.

On each panel A–D, the solid line represents a 1-to-1 relationship and the Pearson correlation coefficient (r) is reported in the bottom right. (A) The week of spore release simulated for each year by module A compared to observations of initial spore detection (spores/L ≥ 1). The observed week of spore release was not available for 2007 and 2009 as spore monitoring started when spore density was already above 1 spore/L (post-onset). (B) The normalized spore density values simulated by module B compared to observed spore density (normalized by dividing by the maximum annual spore density to range from 0 to 1). Only years excluded from developing the quadratic thermal response were used in this assessment of model fit. (C) The annual spring peak of spore density (natural log of spores/L) simulated by module C compared to observed peak spore density for each year. The selected model simulates the seasonal maximum of spores based on the prevalence of infection of hatchery-origin Chinook juveniles from the previous year (HPOI) and the occurrence of a flushing flow (FF; see Table 3). (D) Spore density (natural log spores/L) simulated by the final spore model compared to observed spore density for 2007–2018.

Figure 6. The relationship between observed Ceratonova shasta spore density (spores/L) and water temperature (°C) during spring months (February–June) fit with a quadratic curve.

Spore density values in each year were normalized by dividing by the maximum annual spore density to range from 0 to 1. The selected years (2008, 2010, 2014, and 2015) met the criteria of having spore densities >1 spore/L, and no flushing flow ≥171 m³/s for 72 h.

Figure 7. Model simulating spore densities of Ceratonova shasta for weeks 1–30 in years 2007–2018.

Simulated spore densities (ln spores/L; orange line) were generated by the spore model that combined modules A, B, and C. Observed spore densities are displayed for model assessment. The range of observed spore density values (black lines; min–max) and mean (black squares) are based on replicate water samples measured at the Beaver Creek monitoring site.