I have "visited" your living room a time or two.

Sonar technology is constantly improving and as a result the potential for using it in places like the Skagit is becoming more practical though I suspect that it is still a long ways off.

While I have to admit to not tracking the use of sonar closely the last time I checked the latest promising technology has been what is called daul frequency identification sonar or DIDSON. This is the system currently be used successful in a number of locations in Alaska and I'm suspect that is where you have seen it in action (Anchor River?). While the DIDSON is a vast improvement over some of the split screen trechnology some of the same limitions would be a concern on the Skagit..

1) There remains some difficultites in detecting targets are longer ranges.

2) Finding a suitable site with a relatively flat bottom and even flows on a large river like the Skagit is difficult - the sonar needs to be able to "hit" the target and "bounce" back to the recievers.

Both of these could potentially overcome with placement of multiple units and innovating aiming strategies. However given the long migration period of steelhead in the Skagit (November to May) and the dramatic variability in flows (likley to see annual variations in flows from less than 10,000 cfs and more than 50,000 cfs). Those difference will change fish behavior and the ability to consistently detect them. As well as putting that expensive gear at risk during high flow events.

3. It appears that accuracy is better when there are active migrating fish (sometimes difficult to separate downstream migrating fish from non-fsih targets) and when there are significant numbers of fish moving there the monitoring area - numbers like several 100/hour.

4) species Identication continues to be a problem. In some of the Alaska rivers fish like sockeye and Chinook can be separated by target size. In other cases differences in behavior between species of interest or a single species make the id question easier.

Both of these factors when appleid to the Skagit steelhead question are likely to be sources of significant errors that might limit the practical application of the technology. The "patchy" nature of the steelhead migration and the relatively low numbers of fish moving through the area are likely to place a wide confidence intervals on any estimates.

In addition early in the migration steelhead numbers would be completely "masked" by chum and coho salmon - similar sized fish that would be much more numerous. During the spring period id would be complicated by spring Chinook migrating at the same time.

Assuming for a minute that an esitmate of the run could be made using this DIDSON technology (and I think that is far from a sure thing) I'm not sure that it would be very applicable for in-seasoan management. It might provide an alternate estimate of the steelhead escapement/run to the current redd counting method. However given the wide error confidence intervals on in-season estimate I would think that there would not be much confidnence for in-season use uintil well into the run.

The type of precise folks seem to what for in-management decisions would likely not be available until well nto the spring - late April/May. Not sure what the utility of information at the time would be for decisions such as we are discussing here.

Believe it or notthre are always some fisheries management folks are actually interested in gettng the best information upon which to make management decisions as well as monitoring various populations. However the application of new technology will also be dependent on it being reliable, economically feasible, and reasonably practical. As sonar methods improve I can see applications here in Washington. Some of themost likley places to first see such methods might be in places like the Cedar river sockeye. They also might work well on spawning runs of summer steelhead in places like the upper Columbia but with those generously supplied concret counting fences provided by BPA and others not sure why other methods are needed.

One of the last places in Washington were I would expect to see the application of sonar methods would be basins like the Skagit. The overlapping species complexes, the diverse behaviors within the various species and dynamic nature of the flows and other habitat features in the Skagit all operate together to make it a difficult "nut" to crack.

I'm sure that others may have a different spin on the potential use of sonar on the Skagit. However the above is one inquiring mind's take on the issue.

Tight lines
Curt