8.1 Fish
The information presented in Section 6.0 provides a review of experimental results and theoretical methods which describe delta P thresholds for the major signs of GBT in fresh water fish. The evidence suggests the existence of three thresholds which are dependent on environmental variables such as water delta P, depth, and pO2. However, biological variables such as fish size, species, and behaviour also play important roles in the effects of DGS on fish. It is the goal of this section to use this information, along with information on habitat and other environmental variables, to derive water quality guidelines for DGS.
8.1.1 Factors Affecting Guideline Derivation and Application
There are a variety of factors which must be considered in both the derivation and application of water quality guidelines for DGS. The threshold information described earlier, available water depth, fish species, fish age class, available habitat, habitat usage, the presence of suspended or deposited particulate matter, and in the case of swim bladder over-inflation, the role of the swim bladder under normal conditions, are all important in establishing appropriate guidelines.
8.1.1.1 GBT Thresholds
The first threshold for signs of GBT occurs at low levels of delta P and is related to over-inflation of the swim bladder. As pointed out in Section 6.0 this sign of GBT depends, in part, on the size or age class of the fish. In the case of small rainbow trout (under 50 mm), swim bladder over-inflation can occur at delta P levels of about 25 mm Hg (sea level TGP% about 103% at zero water depth). Swim bladder rupture may occur as water delta P levels approach 76 mm Hg. At delta P levels less than 76 mm, the problem encountered by these animals is primarily one of over-buoyancy. At depths less than the compensation depth for the existing level of DGS, the animal must swim continuously in a head down orientation to maintain its position in the water column. The swimming requirement may lead to elevated stress in the animal and increased mortality, or perhaps make it more prone to predation (White et al. 1991). Where sufficient depth is available, small fish may use that depth to compensate for the excess buoyancy. This not only solves the problem of over-buoyancy but also protects the animal from other signs of GBT which occur at higher delta P levels. Swim bladder over-inflation is generally not a problem for fish over about 85 mm in length.
The second GBT threshold occurs at delta P levels of about 76 mm Hg and is associated with extra-corporeal inter-lamella bubbles and sub-dermal emphysema of external skin surfaces. These signs can affect both juvenile and adult fish. Because adult fish do not experience the over-buoyancy problems encountered by juvenile fish, they may not use depth as a compensating mechanism. As a result, they can be exposed to these signs of GBT while their juvenile counterparts are not.
The third threshold for signs of GBT occurs when delta P levels rise above 115 mm Hg. At these delta P levels, bubble growth in the vascular system begins (Figure 19). Mortality is generally rapid with the time to mortality decreasing as delta P levels increase. As with the second threshold for GBT, juvenile fish may be protected from this sign if they are at or below the depth required to compensate for swim bladder over-inflation.
8.1.1.2 Water Depth
As indicated by Equations 4, 5, and 6 and by Figure 19, water depth plays an important role in delta P thresholds for GBT signs. If sufficient depth is available to fish, and they make use of it, GBT signs may not appear at all. As described earlier, Shrimpton et al. (1990a and b) found that small fish would seek depth to avoid the problems of swim bladder over-inflation. It is not clear if larger fish, encountering the other signs of GBT (i.e., sub-dermal emphysema and cardiovascular bubbles), would move to deeper water to avoid the signs. It may be that fish stressed by signs of GBT move to depth as a normal response to stress. If so, and there is sufficient depth available, the signs of GBT may again be avoided.
8.1.1.3 Available Habitat and Habitat Usage
The importance of habitat and its usage cannot be overemphasized in the development and application of DGS guidelines. This is best illustrated by comparing two river systems which have similar biological structure and comparable levels of DGS, but very different depth regimes and very different levels of GBT in fish.
The Columbia River below the Hugh Keenleyside Dam in southern British Columbia has levels of DGS approaching a delta P of 360 mm Hg at various times of the year (Clark 1977, Maxwell 1985, Hildebrand 1991). Yet it is a very deep river with depths of ten metres being quite common. Five metres of depth would compensate for the highest delta P levels which have been reported for the river. The Bighorn River in Montana, below the Yellowtail Afterbay Dam, also has DGS levels approaching a delta P of 350 mm Hg for similar periods throughout the year (White et al. 1991). In this river, water depth is only about one metre for nearly 20 km downstream of the dam. This depth, even if used by fish, is insufficient to compensate for delta P levels above 74 mm Hg. Numerous fisheries surveys on the Bighorn River have found that at times of high DGS ( delta P = 350), up to 90% of all fish captured exhibited severe signs of GBT and high levels of mortality were common (White et al. 1991, also Figures 2 through 5). These surveys involved several thousand animals over a four-year period. On the Columbia River, surveys which also involved several thousand fish (both adults and juveniles) conducted over a similar time period found a maximum of 3% of the animals captured show mild signs of GBT (Hildebrand 1991). The only sign observed was sub-dermal bubble formation on the external skin surfaces of the affected animals. In these surveys, no mortalities were found which could be attributed to GBT. Thus, it appears that fish of the Columbia River are taking advantage of the available water depth to avoid the signs of GBT. It should be noted that when delta P levels in the Bighorn River were less than 76 mm Hg, no signs of GBT were observed in fish (White et al. 1991).
Depth plays yet another role in terms of specific habitat usage. For example, some species of fish may prefer shallow habitats for spawning and rearing. If water depths over these habitats do not adequately compensate for the prevailing levels of DGS, they may be avoided by spawning fish or the fry may have to move into deeper water to avoid the signs of GBT. The move to deeper water may or may not present adverse conditions to the young fish depending on predators, available cover, and a variety of other factors. Alternatively, if a river is supersaturated with dissolved gases (perhaps as a result of a dam), there may be tributaries flowing into the river which are not supersaturated. Fish may move into the tributaries for refuge from the effects of swim bladder over-inflation and other signs of DGS. Clearly, this would involve a displacement of fish from their normal habitat.
In many rivers and lakes, there can be periods when water levels are high and there is adequate depth to compensate for the signs of GBT. At other times water levels may be low and there is insufficient depth anywhere on the river or lake to provide protection from the signs of GBT.
In other situations, the levels of DGS in a river or lake may vary dramatically throughout the year (Hildebrand 1991, White et al. 1991). Thus, the timing of the high levels of DGS, water levels, and habitat use by various species for spawning and rearing can be critical in terms of the appearance of GBT signs.
Finally, there may be streams in which there is no habitat available for fish. Many of the smaller streams in British Columbia which could be used for small hydro-electric projects are inaccessible to fish. Water falls, high water velocities, or other barriers to fish passage isolate the stream or sections of the stream from fish use. Although higher levels of DGS could be tolerated in these streams, the higher delta P levels could affect fish in other streams into which these smaller streams discharge. An additional consideration is that, although a stream may not have fisheries habitat, there will probably be invertebrate and plant communities present. In some situations, it may be desirable to protect these.
8.1.1.4 Fish Species
As described in Section 6.0, the bulk of experimental data which describe the effects of DGS on fresh water fish species is for rainbow trout, cutthroat trout, and three Pacific salmon species. However, the other data which were reviewed in Section 6.2 indicate that species such as carp (Cyprinus carpio), black bullhead (Ictalurus melas), channel catfish (Ictalurus punctatus), mountain whitefish (Prosopium williamsoni), largescale sucker (Catostomus machrocheilus), torrent sculpin (Cottus rhotheus), and northern squawfish (Ptychocheilus oregonensis) are not more sensitive to DGS than trout and pacific salmon. Consequently, criteria which are developed from data on trout and Pacific salmon should be protective of these other species as well.
8.1.1.5 The Role of the Swim Bladder
The GBT sign of swim bladder over-inflation has been identified as the sign which occurs at the lowest delta P threshold. At low values of water delta P, it is a chronic sign affecting small or juvenile fish. Because the swim bladder is an organ which is used by fish for buoyancy control, it is important to consider its function independent of DGS and then re-examine this function in the presence of DGS. As Harvey (1963) has shown, sockeye salmon have a tissue density of 1.0634 g/ml. This density is probably characteristic of most species of Pacific salmon and trout and results in a tendency for these fish to sink in the water column. However, the swim bladder can be filled with gas and used to offset the high density and achieve neutral buoyancy. The swim bladders of physostome fish, such as trout and salmon, are filled with air at the water surface. The fish apparently gulps air into its mouth and forces it into the swim bladder by way of the pneumatic duct (Harvey 1963).
An analysis of the volume of gas in the swim bladder required to keep a sockeye salmon at neutral buoyancy shows that the gas volume should be a constant 6% of the total volume of the fish, regardless of water depth (Appendix E). Although the 6% volume is independent of water depth, the pressure within the swim bladder is not and would vary with water depth (i.e., hydrostatic pressure ). The actual volume of the swim bladder would also vary with water depth. As a fish goes deeper in the water column, hydrostatic pressure would cause the volume of the swim bladder to decrease. Thus, the deeper a fish is in the water column, the greater must be the volume of the swim bladder at the water surface in order to achieve neutral buoyancy at depth. Figure 22 shows the volume of the swim bladder at the water surface (as a percent of the total fish volume) required to achieve neutral buoyancy, at various water depths.
Considered from another perspective, a fish having a swim bladder volume equal to 6% of its total volume at a depth of two metres would have a larger volume at one metre and an even larger volume at the surface. The reduced hydrostatic pressures at the shallower depths would cause the swim bladder to expand. Thus, at depths less than two metres, the fish would be over-buoyant, with the amount of over-buoyancy increasing as water depth decreases. In order to maintain their position in the water column at the shallower depths, the fish would have to swim in a head down position (Harvey 1963, Shrimpton et al. 1990b). Clearly, over-buoyancy is a condition which fish must face in natural environments. On the other hand, if the fish drops below the two metre level it would be under-buoyant and would have to swim in a head up position in order to maintain its position in the water column.
Harvey (1963) has shown that physostome fish, such as sockeye salmon, will vent gas from the swim bladder when frightened in order to change their buoyancy. When this happens the fish must return to the surface to refill the bladder in order to adjust its buoyancy to other depths. Harvey (1963) also found that fish will gradually loose gas from the swim bladder as a result of diffusion into the blood and subsequently to the water by way of the gills. Again, the fish must return to the water surface to refill the swim bladder in order to maintain neutral buoyancy at depth.
When conditions of DGS appear, fish may have to adopt different strategies for dealing with over-buoyancy which occurs as a result of swim bladder over-inflation. They may spend more time at depths which they might otherwise avoid. Depending on the circumstances (i.e., available cover, the presence of predators, etc.), this may lead to reduced levels of survival. Conversely, the presence of DGS may allow the swim bladder to remain filled with no gas loss to the water, thus avoiding a trip to the surface to fill the swim bladder. In this situation, the fish may also avoid predators which it would normally encounter on its trip to and from the water surface. In either case, the strategies which fish adopt in these situations have not been examined to date. As a result, it is not possible to assess the net effects of DGS on swim bladder function in terms of the potential for increased or reduced mortality.
8.1.1.6 Water pO2
It has been shown by Rucker (1975b), Fidler (1988), and Shrimpton et al. (1990a and b) that water pO2 plays an important role in establishing the thresholds for the signs of GBT. In hyperoxic environments, fish are more protected from the effects of DGS. However, in hypoxic environments they may be more exposed than would be suggested by Equations 4 through 6 alone. For example, if water in a river is low in oxygen (as might be the case where water from the hypolimnion of a reservoir is discharged through a dam), Equations 4 through 6 might indicate safe conditions based on pO2 levels and available water depth. However, the depth factor may not be relevant in this case.
Figure 22: Volume of Swim Bladder for Neutral Buoyancy
This is because the low water pO2 may override a fish's normal response to compensate for over-buoyancy. As a result, fish may stay near the water surface where dissolved oxygen levels are higher as a result of re-aeration. In doing so, they become exposed not only to the problems associated with over-buoyancy, but to the other signs of GBT which occur at higher levels of delta P.
8.1.1.7 Hatchery Versus River and Lake Environments
Hatcheries present unique environments which are seldom encountered in rivers or lakes. High densities of fish, shallow water depths, long exposure periods, predominantly surface feeding, and high risk of disease all compound the effects of DGS. The shallow water environments and long exposure durations are particularly significant in relation to the results reported by Wright and McLean (1985) for chinook salmon in a hatchery environment. These results, which involved a water depth of 0.5 metres and an exposure period of 122 days were the principal evidence which supported the chronic nature of low levels of DGS and its effects on small fresh water fish (i.e., swim bladder over-inflation). However, it is highly unlikely that small fish in rivers and lakes would be confined to a water depth of 0.5 metres or less for up to 122 days. Thus, the development of water quality guidelines for DGS may require criteria for hatcheries which are different than those for rivers and lakes.
8.1.1.8 Water Temperature
As pointed out in an earlier section, Jensen et al. (1986) found that temperature has an effect on time to mortality resulting from exposure to DGS. However, as pointed out in Section 6.1, there is no evidence that temperature affects the thresholds of DGS for the signs of GBT. As will be described, water quality guidelines should be based on avoiding the appearance of the signs of GBT and not on an acceptable level of mortality. Consequently, based on the available evidence, temperature should not be a factor in the derivation of DGS water quality guidelines.
8.1.1.9 Altitude
Altitude becomes a factor in the derivation of DGS water quality guidelines only when water dissolved gas tensions are expressed in terms other than delta P. If TGP or TGP% are used, corrections must be made for altitude. In situations involving rapid changes in barometric pressure as a result of aircraft transport of fish, the guidelines should apply. However, in the case of over-inflation of the swim bladder, there are two components which must be considered in calculating the delta P. The first is the delta P resulting from the change in altitude. The second component is the delta P resulting from any DGS existing before the change in altitude.
8.1.1.10 Background Levels
The final consideration in the derivation and application of water quality guidelines for DGS is the prevailing background level of DGS. As pointed out in earlier sections, many river systems and lakes have naturally occurring DGS at various times of the year. This may be caused by solar heating, primary production, or upstream water falls. It is clear that water quality guidelines must recognize the existence of this form of DGS. It must also be recognized that there may be some levels of naturally occurring DGS which may lead to adverse conditions for fish and perhaps mortalities. Thus, it is not adequate to argue that DGS created by man-made activities at one location are acceptable simply because of naturally occurring levels elsewhere. Clearly guidelines must incorporate natural conditions in some way which can be rationally applied to man-made forms of DGS.
8.1.2 Rationale
Given the wide range of environmental and biological variables which can influence the impact of DGS on fish populations, a single value numerical guideline appears to be impractical. Such a guideline may be too restrictive in some situations and not restrictive enough in others. The first requirement of a guideline is to protect young fish from the chronic effects of swim bladder over-inflation at low levels of DGS. As such, water depth and pO2 levels as defined by Equation 4, combined with the experimental results of Shrimpton et al. (1990a and b), become the central criteria for the derivation of DGS water quality guidelines.
In addition to protecting small fish, it is necessary to protect fish of all sizes from the acute signs of GBT involving sub-dermal emphysema, the blockage of the gill respiratory water flow by extra-corporeal bubbles, and the development of bubbles in the cardiovascular system. Without the need for over-buoyancy compensation, larger fish will be especially prone to these signs at the water surface. In Section 6.1.2.1 it was shown that the lowest delta P from the data of Table C3 for acute mortality was about 76 mm Hg. Thus, a delta P threshold of 76 mm Hg becomes the second criteria for the derivation of a DGS water quality guideline. This is close to the threshold for extra-corporeal bubble growth and sub-dermal emphysema predicted by Equation 5.
The combination of Equation 4, the experimental results of Shrimpton et al. (1990a and b), and the delta P threshold of 76 mm Hg will form the basis for a guideline. It should be noted that 76 mm Hg is very close to the hydrostatic compensation pressure corresponding to one metre of water depth (i.e., 73.89 mm Hg). Thus, one meter of water depth can serve as a convenient division (or cross over point) between Equation 4 and the delta P threshold of 76 mm Hg. Additional restrictions will be required for hatchery environments where shallow water depth, crowding, and added exposure to diseases increases stress beyond that encountered in natural environments.
8.1.3 Guideline Derivation
The implementation of the derivation rationale is as follows.
For Water Depths Greater than One Metre: Where available water depths exceed one metre, the maximum delta P should not exceed 76 mm Hg regardless of water pO2 levels. For sea level conditions, this corresponds to a TGP% of about 110%. This guideline is consistent with observations on the Columbia River in both Canada and the United States and on the Bighorn River in Montana. At delta P levels below 76 mm Hg, no GBT signs involving sub-dermal emphysema or fish mortality have been observed in these systems. This portion of the guideline is also consistent with the US EPA guideline for DGS.
For Water Depths Less than One Metre: For water depths less than one metre, the guideline should be based on Equation 4 which describes the threshold for swim bladder over-inflation as a function of water depth and pO2 levels. For example, at a water depth of zero metres and a pO2 of 157 mm Hg, the delta P must not exceed 24 mm Hg. For sea level conditions, this corresponds to a TGP% of about 103%. As noted earlier, Equation 4 is consistent with the experimental results of Shrimpton et al. (1990a and b).
Figure 23 shows the results of combining the two guideline criteria. As shown in the figure, for water
Figure 23: Water Quality Guidelines for Dissolved Gases
depths greater than one metre, delta P levels greater that 76 mm Hg (i.e., to the right of the vertical line at delta P = 76 mm Hg) are unsafe for fish regardless of water pO2 or depth. For water depths less than one metre, the maximum allowable delta P depends on water pO2. The region to the right of the diagonal line, corresponding to a given water pO2, is unsafe for fish while the region to the left is considered safe for fish. This guideline recognizes that young fish may seek depth to avoid swim bladder over-inflation or to compensate for over-buoyancy, and that one metre is not an uncommon depth for fish to use for rearing habitat.
In rivers or lakes where there is a natural background level of DGS, the same guidelines should apply to any man-made alterations to the dissolved gas regime. That is, any changes to the dissolved gas levels, in combination with natural background levels, should not exceed the above guidelines. If natural levels are higher than the recommended guidelines it must be recognized that these levels may also be harmful to fish. Therefore, there is no justification for introducing comparable DGS levels of man-made origin to a river or lake.
In hatchery environments Equation 4 may not apply. With the much higher fish densities in hatcheries, accompanied by declines in pO2 along the rearing facility and surface feeding, fish may spend more time near the water surface and become subjected to higher stress levels than in natural environments.
For Hatchery Environments: It is recommended that the DGS guideline for hatcheries be set at a maximum delta P of 24 mm Hg (i.e., the threshold for swim bladder over-inflation under sea level normoxic conditions and zero water depth). This corresponds to a sea level TGP% of 103%. If pO2 levels in the hatchery drop to 100 mm Hg, the guideline should be a maximum delta P of 0 mm Hg.
8.1.4 Guideline Application
The application of the above DGS water quality guidelines to man-made alterations of aquatic environments must focus on fisheries habitat. The first step in applying the guideline is to assess the habitat which is available for use by the various fish species of a river or lake. This includes assessment of habitats for spawning, rearing, and adult holding along with information on the temporal usage of these habitats. These data, along with information on water depth (which may vary over the year) and pO2 levels (which may also vary over the year as well as diurnally), provide the necessary information for application of the guideline criteria. The conditions described in Section 8.1.1 will also have to be considered in establishing guideline compliance.
8.1.5 Guideline Application Examples
To provide water managers with guidance in the application of the guideline for dissolved gas supersaturation, two examples will be described. The first involves a small creek in east central British Columbia while the second involves the Columbia River in the southern part of the province.
8.1.5.1 Camp Creek
Camp Creek is a tributary of the Canoe River which flows into the northern reach of Kinbasket Reservoir near Valemount, BC The creek provides important spawning and rearing habitat for many of the sportsfish species of Kinbasket Reservoir (Triton Environmental Consultants Ltd. 1992, Slaney et al. 1993, Fidler 1994). The headwaters of Camp Creek lie near Mt. Lulu in the Cariboo Mountains of British Columbia. From its headwaters, Camp Creek flows 12 km east through a steep mountain valley. It then turns north and meanders through a wide flat valley for 18 km where it joins the Canoe River about 7 km south of Valemount, BC Only the lower 18 km of Camp Creek are accessible to migratory fish species. This portion of the creek is relatively low gradient with run-riffle sections separated by slower meandering glide-pool sections (Triton Environmental Consultants Ltd. 1992, Slaney et al. 1993). Creek substrate consists of large cobbles with considerable amounts of fine sand and silt deposited in the low velocity sections. Although there are a few deep pools in this section of the creek, they comprise a very small percent of the total habitat. The elevation of this section of the creek is approximately 925 m above sea level. The mean water depth for most of the year is about 0.5 m, with levels dropping to about 0.3 m in the fall and winter months.
Because the upper reaches of the creek (beyond the lower 18 km) are very steep, there is excellent potential for the development of a small hydroelectric facility on the creek. In any hydroelectric facility there are periods when the turbines cannot handle all of the water flow and some of the flow must be spilled. If this is done with a dam sluice-way system, there exists the potential for the creek below the dam to become supersaturated with dissolved gases.
Application of the guideline for dissolved gas supersaturation to this creek must consider the predominantly shallow water of the lower 18 km of the creek and the altitude. At an altitude of 925 m, the atmospheric pressure is approximately 690 mm Hg (US Standard Atmosphere 1976) and the partial pressure for oxygen (dry air) is approximately 145 mm Hg. Assuming that the creek is in equilibrium with the atmosphere, the partial pressure of dissolved oxygen would also be approximately 145 mm Hg. As noted above, most of the creek is at a depth of 0.3 m during the fall and winter months when some fish species will be spawning (e.g., Kokanee salmon, burbot, and mountain whitefish) and juvenile fish species will be rearing in the creek. Using this information in conjunction with Figure 23, one would obtain a guideline value of approximately 38 mm Hg. That is, if one moved left along a line parallel to, but between, the 100 mm Hg and 157 mm Hg pO2 lines of the figure to a point corresponding to a depth of 0.3 m, the guideline delta P would be read from the top axis as 38 mm Hg.
Although this would be the suggested guideline, other factors must be considered. For example, this guideline value would protect juvenile fish if they spent most of their time on the creek bed. Because there are so few areas of deeper water in the creek during this time of year, additional protection should be considered. The situation just described would be very similar to that in a hatchery and a guideline of 20 mm Hg would be more appropriate. This would allow fish to use the full 0.3 m of the water column as needed. In another creek where there may be a more even balance of shallow and deep water, the 38 mm Hg delta P guideline might be more appropriate.
8.1.5.2 Columbia River
The Columbia River, as described earlier, is a large, deep river supporting a wide range of sports and other fish species (Hildebrand 1991). Although the Columbia River itself is presently highly supersaturated with dissolved gases (Tables 2 and 4), there are many tributaries running into the river which are not supersaturated. The one exception is the Kootenay River which flows into the Columbia River near Castlegar, BC At times, this river is also highly supersaturated (Table 30). The DGS of the main stem Columbia River is caused primarily by spilling water at the Hugh Keenleyside Dam and at the Brilliant Dam on the Kootenay River. At times, the main stem dissolved gas levels are also influenced by high background levels of DGS in Lower Arrow Reservoir (Table 2).
The tributaries as well as the main stem of the Columbia River provide important spawning and rearing habitat for most fish species of the river. An exception is the white sturgeon which use the main stem of the river for spawning, rearing, and adult holding. For the Columbia River, the most important considerations in applying the guideline for dissolved gas supersaturation are the large variations in water depth which occur throughout the year and the high background levels of DGS associated with Lower Arrow Reservoir. The variations in water depth are the result of demands placed on river water flow by the Columbia River Treaty. The background DGS of Lower Arrow Reservoir varies throughout the year and appears to have a strong component related to solar heating (Ash et al. 1993). As a result of these factors, the guideline for the river may not be a set value for the entire river, but may have to be adjusted at various times of the year. For example, the maximum delta P allowed for any river or lake is 76 mm Hg. At times, the background levels in Lower Arrow Reservoir exceed this value (Table 2). Although this level is permitted in the guideline (i.e., as a result of natural processes), no anthropogenic structures or processes can increase delta P levels beyond this level.
In other situations when water levels are high and there is abundant spawning and rearing habitat available with water depths of 1.0 m or more, the maximum delta P would be 76 mm Hg. If water levels subsequently drop over these areas, dissolved gas levels may have to be reduced to avoid forcing fish into other habitats. For example, if water depth over rearing areas which are in use drops from 1.0 m to less than 0.5 m, the allowable delta P should be reduced to 38 mm Hg. This would afford approximately the same protection as the original 1.0 m of depth.
The application of the guideline in a variable mode, as just described, would permit a more flexible mode of operation for individuals or companies (e.g., power generating companies and agriculture operations) which alter river or lake natural dissolved gas regimes. However, this mode of guideline application would require that river fisheries activities, water depths, and dissolved gas levels be monitored on a regular or even continuous basis. This requirement could be incorporated into the water license for the particular operation. In cases where the individual or company did not wish to take advantage of a variable guideline, the guideline should be set at a conservative value which is protective to fish under all possible operations. In deep rivers (greater than 1 m deep), the most conservative situation will be when the criteria/guideline is expressed in terms of depth at which fish reside rather than water depth, as stipulated in the above example.
8.2 Invertebrates
The information available which describes the effects of DGS on fresh water invertebrates was reviewed in Section 6.3. The data indicate that these organisms are susceptible to signs of GBT at water delta P levels in excess of 76 mm Hg.
8.2.1 Factors Affecting Guideline Derivation and Application
With the exception of those factors involving the swim bladder, most of the other considerations described for fish (Section 8.1.1) would apply to guidelines for aquatic invertebrates. Because invertebrates are rarely raised in a hatchery environment, the shallow depths and crowding associated with fish hatcheries should not impose any added restrictions to the water quality guidelines for aquatic invertebrates.
8.2.2 Rationale
Fish, as a result of swim bladder over-inflation, exhibit a higher degree of sensitivity to the effects of DGS than do aquatic invertebrates. That is, GBT signs in fish appear at lower delta P levels than have been reported for aquatic invertebrates. Thus, water quality guidelines derived for fish should also be protective of aquatic invertebrates.
8.2.3 Guideline Derivation
The guideline for aquatic invertebrates is the same as for aquatic fish species. Where available water depths are one metre or more, the maximum delta P should not exceed 76 mm Hg regardless of water pO2 levels. For water depths less of than one metre, the guideline should be based on Equation 4 (Section 6.1.3) which describes the threshold for swim bladder over-inflation in fish as a function of water depth and pO2 levels. Figure 23 shows the results of combining the two guideline criteria. As indicated in the figure, delta P levels greater that 76 mm Hg are unsafe for aquatic invertebrates regardless of water pO2 or water depth. For water depths less than one metre, the maximum allowable delta P is a function of water pO2.
In situations where there is a natural background level of DGS, the same guidelines should apply to any man-made alterations to the dissolved gas regime of a river or lake. That is, any changes to the dissolved gas environment, in combination with natural background levels, should not exceed the above guidelines. If natural levels are higher than the recommended guidelines it must be recognized that these levels may also be harmful to invertebrates. Therefore, there is no justification for introducing comparable DGS levels of man-made origin to a river or lake.
8.3 Plants and Algae
To date there are no data which can be used to evaluate the effects of DGS on aquatic plants and algae. As a result, guidelines cannot be derived. However, based on the discussion of Section 6.5, it is not anticipated that aquatic plants and algae would be any more sensitive to DGS than fish.