Limits...
Modelling Infectious Hematopoietic Necrosis Virus Dispersion from Marine Salmon Farms in the Discovery Islands, British Columbia, Canada.

Foreman MG, Guo M, Garver KA, Stucchi D, Chandler P, Wan D, Morrison J, Tuele D - PLoS ONE (2015)

Bottom Line: Numerical particles released from infected farm fish in accordance with IHNV shedding rates estimated through laboratory experiments are dispersed by model oceanic flows.Results demonstrate that neighbouring naïve farms can become exposed to IHNV via water-borne transport from an IHNV diseased farm, with a higher risk in April than July, and that many events in the sequence of farm outbreaks in 2001-2002 are consistent with higher risks in our farm connectivity matrix.Applications to other diseases, transfers between farmed and wild fish, and the effect of vaccinations are also discussed.

View Article: PubMed Central - PubMed

Affiliation: Institute of Ocean Sciences, Fisheries and Oceans Canada, P.O. Box 6000, Sidney, B.C., V8L 4B2, Canada.

ABSTRACT
Finite volume ocean circulation and particle tracking models are used to simulate water-borne transmission of infectious hematopoietic necrosis virus (IHNV) among Atlantic salmon (Salmo salar) farms in the Discovery Islands region of British Columbia, Canada. Historical simulations for April and July 2010 are carried out to demonstrate the seasonal impact of river discharge, wind, ultra-violet (UV) radiation, and heat flux conditions on near-surface currents, viral dispersion and survival. Numerical particles released from infected farm fish in accordance with IHNV shedding rates estimated through laboratory experiments are dispersed by model oceanic flows. Viral particles are inactivated by ambient UV radiation levels and by the natural microbial community at rates derived through laboratory studies. Viral concentration maps showing temporal and spatial changes are produced and combined with lab-determined minimum infectious dosages to estimate the infective connectivity among farms. Results demonstrate that neighbouring naïve farms can become exposed to IHNV via water-borne transport from an IHNV diseased farm, with a higher risk in April than July, and that many events in the sequence of farm outbreaks in 2001-2002 are consistent with higher risks in our farm connectivity matrix. Applications to other diseases, transfers between farmed and wild fish, and the effect of vaccinations are also discussed.

No MeSH data available.


Related in: MedlinePlus

Average a) April and b) July mean surface elevations (cm) and flows (cm/s) at 10m depth in the Nodales Channel region.Numbered red dots denote farms, as in Fig 1, while the black bracketed numbers denote the number of days since the first IHNV outbreak was diagnosed at farm 5 in August 2001. (These numbers were taken from Table 1 in [7].)
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4482787&req=5

pone.0130951.g008: Average a) April and b) July mean surface elevations (cm) and flows (cm/s) at 10m depth in the Nodales Channel region.Numbered red dots denote farms, as in Fig 1, while the black bracketed numbers denote the number of days since the first IHNV outbreak was diagnosed at farm 5 in August 2001. (These numbers were taken from Table 1 in [7].)

Mentions: Approximately 2.5 months after these five IHNV outbreaks, further outbreaks occurred in the Frederick Arm and Nodales Channel region (Fig 1). As the mean flow fields shown in Fig 7 are generally westward, it is unlikely this second outbreak was caused by water-borne transmission from the previous five infected farms. This is substantiated by Fig 6 which does not show any connection from farms 1–5 to the others. The sequence of outbreaks for this second set is shown in Fig 8a, along with the monthly-average 10 m flow fields for April 2010. The initial diagnosis was at farm 13 and the infection then spread within 1 and 6 days to farms 12 and 18, respectively. The former is only 1 km away and though [7] notes its location to be upstream from farm 13, our average flow field shows a clockwise eddy that, in addition to the oscillating tidal currents, could easily connect the two farms. The relatively strong connection from 13 to 12 in Fig 6 substantiates this claim. As noted in [7] and seen in Fig 8, farm 18 is about 10 km downstream so water-borne transmission within 6 days is certainly feasible. But the relatively rapid virus inactivation with time (Fig 2) decreases the likelihood and Fig 6 does not show such a connection. In fact, as Fig 6 does not show any direct connection from either farm 12 or 13 to farm 18, it is much more likely that this outbreak was a result of poor biosecurity or from virus spill-over from a wild source, rather than water-borne transmission from an infected farm. 41, 42, 48, 55, and 70 days after the outbreak at farm 13, the next farms to be infected were 11, 17, 10, 15, and 16, respectively. This sequence of outbreaks is not consistent with the mean flow fields seen in Fig 8, though again as depicted in Fig 2 of [7], diagnosis of the disease does not necessarily correspond with the beginning of the epidemic. Fig 6 does show that i) farm 11 could be infected from farms 12 or 13, ii) 17 could be infected from 13, 15, or 16, iii) 10 and 15 have only weak connections to 11, 12, and 13, and iv) 16 has only stronger connections to 15. So although water-borne transmission may have played a minor role, poor biosecurity practices among workers moving between the farms or naturally infected wild fish are more likely transmission agents [7]. Nevertheless, as indicated in Fig 6, water-borne transmission cannot be discounted completely.


Modelling Infectious Hematopoietic Necrosis Virus Dispersion from Marine Salmon Farms in the Discovery Islands, British Columbia, Canada.

Foreman MG, Guo M, Garver KA, Stucchi D, Chandler P, Wan D, Morrison J, Tuele D - PLoS ONE (2015)

Average a) April and b) July mean surface elevations (cm) and flows (cm/s) at 10m depth in the Nodales Channel region.Numbered red dots denote farms, as in Fig 1, while the black bracketed numbers denote the number of days since the first IHNV outbreak was diagnosed at farm 5 in August 2001. (These numbers were taken from Table 1 in [7].)
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4482787&req=5

pone.0130951.g008: Average a) April and b) July mean surface elevations (cm) and flows (cm/s) at 10m depth in the Nodales Channel region.Numbered red dots denote farms, as in Fig 1, while the black bracketed numbers denote the number of days since the first IHNV outbreak was diagnosed at farm 5 in August 2001. (These numbers were taken from Table 1 in [7].)
Mentions: Approximately 2.5 months after these five IHNV outbreaks, further outbreaks occurred in the Frederick Arm and Nodales Channel region (Fig 1). As the mean flow fields shown in Fig 7 are generally westward, it is unlikely this second outbreak was caused by water-borne transmission from the previous five infected farms. This is substantiated by Fig 6 which does not show any connection from farms 1–5 to the others. The sequence of outbreaks for this second set is shown in Fig 8a, along with the monthly-average 10 m flow fields for April 2010. The initial diagnosis was at farm 13 and the infection then spread within 1 and 6 days to farms 12 and 18, respectively. The former is only 1 km away and though [7] notes its location to be upstream from farm 13, our average flow field shows a clockwise eddy that, in addition to the oscillating tidal currents, could easily connect the two farms. The relatively strong connection from 13 to 12 in Fig 6 substantiates this claim. As noted in [7] and seen in Fig 8, farm 18 is about 10 km downstream so water-borne transmission within 6 days is certainly feasible. But the relatively rapid virus inactivation with time (Fig 2) decreases the likelihood and Fig 6 does not show such a connection. In fact, as Fig 6 does not show any direct connection from either farm 12 or 13 to farm 18, it is much more likely that this outbreak was a result of poor biosecurity or from virus spill-over from a wild source, rather than water-borne transmission from an infected farm. 41, 42, 48, 55, and 70 days after the outbreak at farm 13, the next farms to be infected were 11, 17, 10, 15, and 16, respectively. This sequence of outbreaks is not consistent with the mean flow fields seen in Fig 8, though again as depicted in Fig 2 of [7], diagnosis of the disease does not necessarily correspond with the beginning of the epidemic. Fig 6 does show that i) farm 11 could be infected from farms 12 or 13, ii) 17 could be infected from 13, 15, or 16, iii) 10 and 15 have only weak connections to 11, 12, and 13, and iv) 16 has only stronger connections to 15. So although water-borne transmission may have played a minor role, poor biosecurity practices among workers moving between the farms or naturally infected wild fish are more likely transmission agents [7]. Nevertheless, as indicated in Fig 6, water-borne transmission cannot be discounted completely.

Bottom Line: Numerical particles released from infected farm fish in accordance with IHNV shedding rates estimated through laboratory experiments are dispersed by model oceanic flows.Results demonstrate that neighbouring naïve farms can become exposed to IHNV via water-borne transport from an IHNV diseased farm, with a higher risk in April than July, and that many events in the sequence of farm outbreaks in 2001-2002 are consistent with higher risks in our farm connectivity matrix.Applications to other diseases, transfers between farmed and wild fish, and the effect of vaccinations are also discussed.

View Article: PubMed Central - PubMed

Affiliation: Institute of Ocean Sciences, Fisheries and Oceans Canada, P.O. Box 6000, Sidney, B.C., V8L 4B2, Canada.

ABSTRACT
Finite volume ocean circulation and particle tracking models are used to simulate water-borne transmission of infectious hematopoietic necrosis virus (IHNV) among Atlantic salmon (Salmo salar) farms in the Discovery Islands region of British Columbia, Canada. Historical simulations for April and July 2010 are carried out to demonstrate the seasonal impact of river discharge, wind, ultra-violet (UV) radiation, and heat flux conditions on near-surface currents, viral dispersion and survival. Numerical particles released from infected farm fish in accordance with IHNV shedding rates estimated through laboratory experiments are dispersed by model oceanic flows. Viral particles are inactivated by ambient UV radiation levels and by the natural microbial community at rates derived through laboratory studies. Viral concentration maps showing temporal and spatial changes are produced and combined with lab-determined minimum infectious dosages to estimate the infective connectivity among farms. Results demonstrate that neighbouring naïve farms can become exposed to IHNV via water-borne transport from an IHNV diseased farm, with a higher risk in April than July, and that many events in the sequence of farm outbreaks in 2001-2002 are consistent with higher risks in our farm connectivity matrix. Applications to other diseases, transfers between farmed and wild fish, and the effect of vaccinations are also discussed.

No MeSH data available.


Related in: MedlinePlus