Connecticut River Coordinator's Office
Northeast Region
Photo of a volunteer stocking fry in the Sawmill River - Photo credit:  Draper White
Photo of a volunteer stocking fry in the Sawmill River. Credit: Draper White

Restoration Program activities, to date, have successfully returned Atlantic salmon to the Connecticut River in annual runs that average in the hundreds. This is the first phase of salmon restoration. The second phase will involve activities to increase the rate and number of returning salmon. Though this goal is part of this Plan, it is difficult to estimate the exact amount of increase or to project the exact timing of such increase. A precise projection or estimate is difficult to develop because the size of the run is dependent upon many factors. Past projections that proved inaccurate were based on a variety of factors including historic run size, hatchery production and releases, and projected stream production. Problems with each of these factors affect the accuracy of any projections.

Future estimates, for example, are not safely based on historic run sizes. Loss of native stocks, loss and degradation of salmon habitat, construction of dams, and large-scale exploitation of salmon at sea and in freshwater are factors that may have permanently reduced the potential size of salmon runs in the Connecticut River, with respect to the magnitude of pre-colonial runs.

Additionally, it is difficult to accurately estimate instream smolt production and outmigration survival, complicating interpretation of rates of return for stocked fry. Existing data indicate that adult returns from stocked fry will be more variable than return rates observed in the past for hatchery stocked smolts, due to naturally variable instream mortality. Because rates for fry stocked fish survival are less predictable, projections are less reliable.

Finally, smolt productivity rates in streams in Canada and Europe cannot be used to estimate the levels of expected production in the Connecticut River basin. New England streams are not as pristine as those in Canada or Norway nor are they as biologically productive as those in Spain or the United Kingdom. Connecticut River projections must be based on data collected within the basin. However, Connecticut River data cannot yet be used to predict production rates with a high level of confidence because the amount of data is limited to only a few years.

The number of adult salmon returning to the river is determined by several key factors: the number of smolts leaving the river, the natural mortality of those smolts, and the commercial harvest at sea. Each of these factors is influenced by many other variables. For example, the number of smolts leaving each year depends on the number of fry stocked years earlier, the weather during the subsequent growing seasons, the impact of predation during the seaward migration, the number of smolts killed by hydroelectric turbines, and the river flow during the smolt migration. Even in a native salmon population, all of these environmental factors vary, resulting in naturally fluctuating adult salmon runs from one year to the next, regardless of human influences.

A mathematical model has been developed to provide a very simplified idea of the potential of the Restoration Program. This model uses only two variables: the smolt production of the habitat and the marine survival rate of the smolts. In the Connecticut River basin, the total amount of habitat available for instream salmon smolt production is estimated at 243,000 rearing units. Average smolt production is estimated to be two smolts per unit. If all available habitat in the basin were fully stocked, multiplying these two figures results in a projected annual smolt run of 486,000. Production naturally varies from year-to-year and between tributaries, so the basin's total smolt output can be expected to vary by at least 25%, or within an annual range of 364,500 to 607,500. The marine survival rate of smolts also varies widely by an estimated range of 0.25% to 2.5%. The lower end of this range has been observed in Connecticut River smolts and the upper end of the range has been observed in other U.S. salmon stocks. Table 7 provides the results of this simple model and displays the wide range of sizes of adult runs that may ultimately be possible in the future. Though it is unlikely the higher return figures will be experienced in the Connecticut River, the upper survival range (2.5%) is included to demonstrate the long-term potential for a river, given the described variables.

Figure 4. Observed and Retrospectively Predicted 2SW runs in the
Connecticut River under two scenarios of fishing and survival.
Observed and Retrospectively Predicted 2SW runs in the Connecticut River under two scenarios of fishing and survival.

Scenario A uses a conservative estimate of extant exploitation of non-maturing stocks under the hypothetical condition of no intercept fishery. Scenario B assumes the same no-fishing condition as Scenario A but is adjusted to reflect the higher range of the 1970s survival rates observed for the Connecticut River.

Table 7. Potential Adult Salmon Returns Based on Smolt Production at a Range of Smolt Survival Rates, Assuming Fully Stocked Habitat.
Potential Smolt
Production with Fully Stocked Habitat
Smolt-to-Adult Survival Rates
0.25% 2.5%
364,500 911 9,113
486,000 1,215 12,150
607,500 1,519 15,188

The Restoration Program is described in the Program Summary, section II of this Plan, as a multi-phase Program. Phase I has already established a Connecticut River stock of salmon and a small annual run of adult salmon. Phase II, described in section III, involves the building of the run size from the current average of a several hundred fish per year to over 1,000 fish, the lower range of projected runs in Table 7. If Phase II can be accomplished early in the 2000s, a clearer picture can be obtained about the full potential for the river in Phase III. Phase III will address further increases in returns (which may correspond to the upper range of returns shown in Table 7) as part of full restoration.

There are many uncertainties as to how quickly the current phase of restoration, Phase II, can be accomplished, including: funding levels, hatchery capabilities, downstream bypass performance, changes in climate, and the status of the Greenland fishery. However, there are two major factors that will determine the rate of restoration success.

Since salmon in the Connecticut River can be expected to adapt to the river ecosystem slowly, results will be slow in arriving. The current stock of salmon originated mostly from Penobscot River salmon. As the young salmon are stocked into the Connecticut River basin, natural selection takes over. The poorly adapted fish (in a genetic sense) perish prior to reproducing and their genes are not inherited. The better adapted salmon survive and pass on their genes. Slowly, population traits evolve that are necessary to survival, such as: a well-timed smolt migration, the best average date for adult migration, increased tolerance to warmer water temperatures, and the new development of defense mechanisms against non-native predators, such as rainbow trout, brown trout, largemouth bass, and smallmouth bass. Over time, as the new Connecticut River stock gradually develops, these necessary traits, survival rates in the ocean and production rates in freshwater will increase. Adaptation in salmon is always slow, but occurs at different rates in different tributaries. There are no data with which to predict the rate of adaptation in the Connecticut River because there are no other restoration programs of comparable scale.

Researchers have recently defined what constitutes one parameter of suitable Atlantic salmon habitat in the ocean: water at temperatures between 39 and 46o Fahrenheit. Examination of oceanographic data reveals that the amount of such habitat in the northwest Atlantic Ocean had steadily decreased during the 1980s and 1990s (when the major effort at restoration has taken place). This means that the Restoration Program has been subjected to inhospitable marine conditions throughout much of its history. Analyses of stocking data and survival rates by the National Marine Fisheries Service have concluded that if marine habitat and survival in the northwest Atlantic Ocean had remained at the levels observed during the 1970s, the Restoration Program would most likely have produced adult runs exceeding 1,000 fish for at least four years prior to 1996 (Figure 7). Recent evidence indicates that the condition of marine habitat in the northwest Atlantic Ocean may be cyclic and that minor improvements beginning in 1995 may foreshadow an upturn in the cycle. If the marine climate does indeed improve, better survival rates may be realized in the upcoming years, providing that this is a primary limiting factor to improved return rates.

The increase of fry stocking to all habitat in the basin, combined with improvements in downstream fish passage should result in increased numbers of adult salmon in future years. Stock development through natural selection and improvement in marine habitat conditions should further increase returns. Under these circumstances, adult runs between 900 and 1,500 fish should be achievable in the next 10 to 20 years with higher returns possible. Once Phase II is accomplished, managers will continue working with the population to maximize run size to the full potential of the basin.

Return to top


Last updated: September 13, 2010
Connecticut River Coordinator's Office
Northeast Fisheries Resources Home
Northeast Region Home

U.S. Fish and Wildlife Service Home Page | Department of the Interior  |  | About the U.S. Fish and Wildlife Service  | Accessibility  | Privacy  | Notices  | Disclaimer  | FOIA