For many Michigan walleye anglers, there's an ongoing debate about the relative merits of two Great Lakes fisheries -- Lake Erie and Saginaw Bay.

But some walleye fans say it's an incomplete discussion since they're leaving a third great Great Lakes walleye fishery off the list -- Green Bay, located on the west end of the Upper Peninsula's shore of Lake Michigan.

Although it's a long way from Michigan's largest population centers -- and therefore, gets less exposure -- Green Bay boasts an excellent walleye fishery.

In addition to an exceptional walleye fishery in the Menominee River, much of the action is concentrated in the two large bays at the head of Green Bay -- Big Bay de Noc and Little Bay de Noc.

Little Bay, especially, is a destination for many walleye anglers from the northern reaches of the Great Lakes region for one reason: Walleyes are numerous.

Just how numerous? Department of Natural Resources fisheries personnel are trying to find out.

Troy Zorn, a research biologist working out of the DNR's Marquette office, has a multi-faceted study in progress, using gill nets and boom shockers, oxytetracycline-marked hatchery fish and more, to get to the bottom of it.

Among his findings thus far, Zorn said there appears to be plenty of fish regardless of DNR stocking efforts. The DNR, he explained, has been planting hatchery fish marked with oxytetracycline-medicated feed since 2004.

"In Little Bay de Noc, we've seen natural reproduction every year and the two best year-classes we've seen (in terms of catch per unit effort) have been in the years we didn't stock Little Bay with fish," Zorn said. "Those year-classes were several times better.

"Even when we stock we see natural reproduction in those years," Zorn continued. "In 2004, for example, we stocked 570,000 walleyes in Little Bay de Noc and when we captured young walleyes in our surveys, 38 percent of them were naturally produced."

DNR personnel use boom shockers and gill nets with relatively small mesh sizes (one to two inches) to target walleyes up to 2 years old. They're finding naturally produced fish in both Little Bay and Big Bay.

"We've consistently seen many more juvenile walleyes in Little Bay de Noc than Big Bay de Noc," Zorn said. "Several times more fish are caught in Little Bay than in Big Bay."

Zorn said he thinks the differences among the tributaries to the bays may be the important factor.

"Little Bay de Noc has several nice spawning rivers with good rapids," he said. "The Whitefish River is a huge producer. The Escanaba has nice rapids. The Rapid River has good rapids. Big Bay de Noc just doesn't have those high-gradient spawning rivers.

"Most of the rivers flowing into Big Bay are sandy and they don't have the rapids habitat. The exception is the Sturgeon, which is sandy in the lower reaches and then turns into bedrock. Whatever habitat it has is well upriver. But we've been unable to effectively sample it."

Big Bay does have some reefs that could provide spawning habitat, Zorn said, but that habitat may be less ideal than spawning streams.

"Big Bay gets pretty rough and a lot of the shoreline area gets pounded by waves," he explained. "2007 was the only year that it produced a good number of fish. Perhaps the reefs do produce, but we just don't see the numbers of walleyes in Big Bay de Noc as we do in Little Bay de Noc. In the early years of the study, unless we stocked fish in Big Bay, we hardly saw any young fish.

"I've looked at some of the old literature about commercial fishing and it talks more about Little Bay de Noc as a walleye fishery than Big Bay de Noc."

Zorn said the Whitefish River (which flows into Little Bay) has a "fabulous estuary" with lots of shelter for young walleye.

"In Big Bay there are fewer places for young walleye to hide," he said.

Zorn is trying to develop models to assess the spawning runs. The Cedar River, west of Little Bay, produced a spawning run of about 8,500 fish in 2005. The Menominee River, farther west, had a run of about 58,000 fish in 2006.

"Good numbers of fish from the years that weren't stocked have shown up in the Menominee, so it's pretty well documented that there's decent reproduction there some years," Zorn said.

The Escanaba River had a run of about 8,400 fish in 2008. The Ford River has had a big run in the past, but fisheries personnel couldn't sample enough of it to estimate the run, Zorn said. "We're going to go back," he added.

Both the Whitefish and Rapid rivers had good runs in 2009, but Zorn hasn't finished crunching the numbers yet.

"One of the issues is there wasn't any information on these runs before this study," Zorn said. "So we're trying to figure out what the age structure of each population is and combine them with the stocking history to give us a feel for the level of natural reproduction."

Part of the problem is that past surveying efforts haven't been big enough to describe the population trends in the two bays, let alone the entire system.

"We've had a summer fishery assessment since the late 1980s in which we set two gill nets in both bays one day a month for four months," Zorn said. "Starting this year we're using a wider variety of larger gill nets -- including some with larger mesh sizes -- and we'll be fishing six nets a day for several weeks. That will get us a more representative sample."

One of the bonuses to the new sampling technique is fisheries managers will be collecting data on species other than yellow perch and walleyes, which have long been the bread-and-butter fish of the Green Bay system.

"Now we're getting data on everything -- smallmouth bass, pike, exotic species such as round goby and ruffe -- we've even captured a couple of sturgeon," Zorn said. "Bass fishing seems to be improving in the near-shore waters. We're getting a better picture of the overall fish community."