6.1. Water Quality

In addition to some of the micro- and macroenvironmental parameters discussed above, animals that live in an aquatic environment have requirements particular to their liquid medium. Fish have species-specific and sometime even life stage–specific optimal ranges for each water quality parameter; when parameters fall outside the acceptable range, fish become stressed and more susceptible to disease.

Standard (i.e., control) and experimental water quality parameters (e.g., temperature, ammonia, nitrite, nitrate, pH, dissolved oxygen, carbon dioxide, hardness, alkalinity, supersaturation, salinity, chlorine, chloramine, suspended solids, and heavy metals such as copper, zinc, and cadmium) are to be documented as thoroughly as possible so that the study can be properly assessed or replicated. Most of these parameters can directly or indirectly affect the behavior, physiology, metabolism, reproduction, and immunology of fish (Haywood 1983; Kroupova et al. 2008; Lewis and Morris 1986; Randall and Tsui 2002; Tomasso 1994).

Many water quality parameters are affected by others. For instance, the temperature of the water directly affects the amount of dissolved oxygen in the water—as the water temperature increases, oxygen levels decline. The pH of the water affects the amount of the relatively more toxic un-ionized ammonia in the water versus the amount of ionized ammonia. It is therefore important to document as many water quality parameters as possible to reduce variability in experimental outcomes.

Without adequate filtration, nitrogenous wastes and other excretory products accumulate in an aquatic system (Burrows 1964). The exchange rate and water velocity may also affect the behavior and growth of fish in both flow-through and closed recirculating systems (d’Orbcastel et al. 2009). Thus it is usually relevant to describe the type of mechanical and biological filtration used, including any supplementary equipment (e.g., mechanisms that use UV, ozone, or oxygen).

6.2. Diet

As with terrestrial animals, the source, type, form, quantity, and nutrient and caloric content of the diet can affect aquatic animals and study results. If the food is presented in pellet form, the pellet size is relevant information to provide as certain fish ingest only certain size ranges of food. The number of feedings per day can influence the growth of many fish species (Lambert and Dutil 2001) and uningested food can compromise water quality.

Reports of unintended exposure of aquatic animals used in research to contaminants (e.g., endocrine disruptors, dioxin, melamine) in commercial diets have highlighted the importance of documenting the source and components of diets (Andersen et al. 2008; Fiedler et al. 1998; Rappe et al. 1998; Yan et al. 2009). These compounds can cause changes to genetics and metabolism, with resulting pathologies in the reproductive, immune, and neurological systems (Andersen et al. 2003; Fenske et al. 2005; Länge et al. 2001; Örn et al. 2003), thus potentially confounding research results.

6.3. Housing

An adequate description of housing for aquatic animals used in research will include the type of system (e.g., raceway, tank, aquarium, cage), including the material of which the system is constructed (e.g., concrete, fiberglass, polyethylene, glass) (Arndt et al. 2001) and lighting (e.g., intensity, hours, and circadian cycle) (Bayarri et al. 2002; Downing and Litvak 2001; Head and Malison 2000; Hossain et al. 1998; Karakatsouli et al. 2008). The color of the inside of a tank can also be relevant as it may compromise research results by affecting the behavior, physiology, and stress level of fish (Barcellos et al. 2009; Papoutsoglou et al. 2000; Rotllant et al. 2003; Strand et al. 2007).

Although the presence of structures in a tank or aquarium reduces the ability to observe and monitor the animals, most aquatic animals prefer refugia to avoid tank mates or perceived predators. Aquatic animals maintained in glass or plastic tanks can also become stressed by cohorts in adjacent tanks or by activities in the room. For these reasons bare tanks are neither scientifically nor behaviorally, socially, or environmentally advisable.

As international welfare principles increasingly include fish and other aquatic animals, it is appropriate for study reports to fully characterize approaches to environmental enrichment, such as adjustments in tank size, the provision of substrate or structures, water movement, artificial vs. natural light, conspecifics and sex ratio, artificial or real plants, and varied diet.

6.4. Animal Numbers

Information about stocking density and male:female ratio is a basic requirement in all study publications. Although many species of fish prefer to exist in schools, others are more solitary. As with mammals, maintaining fish species and other aquatic animals according to their behavioral preference will minimize stress in individuals.

Stocking density and sex ratio are known to have a profound influence on feed intake, growth, performance, behavior, and survival of aquatic animals (Correa and Cerqueira 2007; Di Marco et al. 2008; Hecht and Uys 1997; van de Nieuwegiessen et al. 2009). In general, greater stocking density leads to decreased performance and increased aggression in most aquatic animals. However, in the larvae and fingerlings of some fish species increased stocking density has been associated with greater feed intake, more swimming activity, and less aggression (van de Nieuwegiessen et al. 2009).

Subtle effects of confinement, such as changes in social behavior, breeding dynamics, and genetic integrity, are becoming increasingly recognized in aquatic animals (Saxby et al. 2010). Dominance hierarchies have also been documented in aquatic animals maintained in captivity (Paull et al. 2010; Pullium et al. 1999).