When Yatabe and colleagues analysed data collected in the first stage of a lice surveillance program in Chile, they considered the location of farms whereby cages are nested within farms, farms are located within subzones, and zones are nested within productive-environmental zones. Not surprisingly, productive-environmental zones had a significant effect on lice numbers, consistent with farmers observations of spatially clustered infestations. Zones located within the Los Lagos region were associated with higher infestations, whereas those in Aysen region were associated with lower levels of infestation. This observed difference could be a result of farming density differences between the two regions or the oceanographic factors that have been considered to be important in L. salmonis including water depth, tidal range, patterns of water circulation and flow rate.
Taking into account the three most commonly farmed salmonids, it has been proposed that Coho salmon are the most resistant to L. salmonis, followed by rainbow trout and then Atlantic salmon. In the case of C. rogercresseyi, Coho salmon also appear to be the most resistant, with Atlantic salmon and rainbow trout being more susceptible. In relation to L. salmonis, it has been proposed that resistance in Coho salmon results from a heightened response to the parasite, including increased epithelial hyperplasia and inflammation. The same could be true for the resistance of Coho salmon to C. rogercresseyi.
Emamectin benzoate is considered one of the more effective agents for controlling sea lice because it targets both the juvenile and adult stages: it was the only drug permitted for use against sea lice in Chile between 2000 and 2007. Hydrogen peroxide baths are also used, but the treatment is only active on adult stages, resulting in constant regeneration. Previous work involving L. salmonis suggests that lice numbers are reduced by 7 days post-treatment with emamectin benzoate and that levels remain low for up to 64 days post-treatment. A field trial in Chile involving Caligus species showed that lice numbers can remain low for as long as 96 days post treatment. However it was found that colleagues found that only treatments applied in the last month before sampling were significantly associated with lower lice burdens (both emamectin bezoate and hydrogen peroxide bath); treatments applied two and three months before sampling were not statistically associated with lower burdens. This is easily explained in hydrogen peroxide baths by its selective action against adults only. In emamectin benzoate, this could be associated with resistance arising from a failure to rotate drugs (because no additional drugs were registered for use against lice between 2000–2007 in Chile). Farmers started to report a reduction in effectiveness of the drug in 2005 and this was supported by an in vitro study of C .rogercresseyi’s sensitivity to emamectin benzoate in 2008. Other suggested causes of the loss of sensitivity include the use of generic products and the method of delivery through medicated feeds. Another explanation is that treatment tends to only be applied to cages with high parasitic burdens, whilst those with low loads are not treated.
It was found that stocking density was significantly associated with higher mean parasite counts. This is unsurprising given that high stocking densities have been associated with reduced welfare leading to stress, which is likely to make animals more vulnerable to infestation and disease. It has been suggested that stocking densities over a threshold of 22kg/m3 are associated with a decrease in welfare and hence this might provide some indication of appropriate stocking densities to reduce lice burdens. Interestingly, a study of 40 salmon sites in Scotland over four years found no significant link between stocking density and burdens of L. salmonis.
There is an apparent association between fish weight and parasite burden, with larger fish harbouring greater burdens. This association is likely to be related to the time of exposure, with larger fish more likely to have spent a longer time in the sea. A link between age and burden of L. salmonis has been described in Atlantic salmon. Another possible explanation is that heavier fish have a larger body surface area in which parasites can attach. A laboratory study showed bigger fish acquire higher burdens of L. salmonis but when this is expressed as number of parasites per unit of surface area, smaller fish are more intensely infested.
Larval stages of sea lice are susceptible to low salinity such that they begin to die at a salinity of 20% or less. In addition, sea lice have a preference for saline environments and fish in freshwater lose lice burdens. It has also been suggested that water salinity could affect the settling rate and development rate of some lice. For these reasons, it is not surprising that low salinity has been associated with reduced parasite burdens.
It was found that infection with C. rogercresseyi significantly increased mortality in fish infected with Piscirickettsia salmonis. How P. salmonis impacts susceptibility to sea lice is not described.
Many studies have suggested a link between water temperature and sea lice abundance, with particular reference to the faster development of lice at higher temperatures in vivo. However, in many cases analysis of data has failed to demonstrate a statistically significant link between lice burden and water temperature in field conditions. However, this may be linked the narrow range of temperatures recorded in some of these studies.
Infection in neighbouring farms
The role of neighbouring farms as a source of sea lice is becoming a greater concern as the aquaculture industry grows. Multiple studies have identified a link between sea lice burden on farms and the abundance of fish in the surrounding area. Interestingly, have been recently suggested that infection pressure from neighbouring farms was actually greater than the infection pressure from within a farm. In these studies the ‘neighbour effect’ was detected on average up to 30km away. Estimated external infection pressure has been found to be the main predictor of salmon lice populations in newly stocked pens.
Other risk factors identified in the northern hemisphere include cage volume, level of treatment, current speed, loch flushing time, depth of cages and sea lice levels in the proceeding months.