So, how many moose are there really?

By Charlotte Westing

Moose –

Some people love to hunt moose; others put in for the moose draw to increase their family’s chance of getting a permit and begrudgingly load up in the boat when it’s time to fill the freezer. We completed moose surveys after an early-February dump of snow fell and didn’t melt or blow away, allowing us good tracking and spotting.

Our goal was to survey the population every three to five years but budget woes, weather, and a global pandemic had us scrambling to get it done. It had been five years since the last survey and I’m excited to provide an update on what we know about moose in our area. 

Survey efforts this year focused on the “Martin River” (Unit 6B, east of the Copper River including the Martin River Valley) and in the “West Delta” (Unit 6C, west of the Copper River). Mike Collins and Jared Kennedy piloted our survey planes, observers were Chloe Bee, Alex Arduser (volunteer), Milo Burcham (volunteer), and me. We complete these surveys with funding from the Alaska Department of Fish and Game (ADFG) and the US Forest Service (USFS).

The first step of the survey involves classifying habitat as having either a high or low likelihood of having moose, a process referred to as stratification. We did this using data from the last three surveys, which showed little variability from year to year.  


After stratifying the area, we divided our sampling effort so that we spend most of our time (75% to 85%) in portions of the survey area expected to have moose (we call these “highs”) and the rest in areas where we don’t expect them (though we inevitably find a couple, we call these “lows”). Sample units are about 6.5 square miles. We spend about 30 to 60 minutes flying circles in each one looking for moose and taking waypoints on the animals we find. Looking at just these raw numbers of moose seen, we observed 464 moose in Unit 6C and 173 moose in Unit 6B.

What we haven’t accounted for at this point is moose that we missed. Ideally, we assess this within the survey with more intense surveying. This basically means we’ll sample half the unit with twice the intensity and see if we can find any additional moose.

Additional moose seen allows you to calculate a Sightability Correction Factor (SCF). This year, we were worried about completing the survey before the weather came in. We opted to stick with standard intensity sampling and use the SCF from other surveys. Survey-specific conditions can really affect the SCF, such as snow or light conditions, so we went with the middle number of the last five surveys.

For the West Delta, the SCF was 1.11. That means for every 100 moose we saw, we probably missed 11. For the Martin River, the SCF was a little higher at 1.16, so for every 100 moose we saw, we probably missed 16. 

We take the number of moose we see in all the sampled units (51% of the total area) and we estimate the whole area. This year was different because we did better at predicting good habitat (we had few highs with no moose and few lows with many moose) and then we sampled nearly all of those “high” units. That leaves us mainly estimating moose in the areas where we don’t expect to find very many and then applying the SCF. Using this approach, we can estimate the precision of the survey (usually presented as a range of numbers).

Using estimates, we more easily track changes in the population and can set more liberal harvest quotas because estimates account for the difference between what is seen and what is actually present. Some wonder why we don’t survey the whole area like we used to until 1990. That “minimum count” approach provided only the minimum number of moose out there.

The technique was appropriate when the population was growing and expanding. However, our moose populations are now “settled out” and there are several reasons the technique is no longer ideal. Left out in the old-style minimum counts was the number of animals missed due to conditions, the way surveys were flown, who was flying, who was observing, or just how much time was spent in an area. Sampling helps to understand the influence of these factors. Instead of covering the whole area with less effort and getting a minimum number, we fly a smaller area with a much higher intensity and end up with “measures of precision.”

Our point estimate for the West Delta was 503 moose with a 90% confidence interval of 497 to 510 (this is that range of possible estimates). This is lower than the last three point estimates (601 in 2012, 609 in 2014, and 677 in 2018) however, it is within the range of possibilities for the last two surveys.   The point estimate is the mid-point between the high and low end of the confidence interval. It is no more likely to be correct than either extreme, but it’s the number we use for management.

In recent years, hunters and pilots have expressed concerns that the moose population has declined.  Both ADFG and USFS administer hunts on the West Delta, so we collaborate to set quotas. In 2020, we talked about local concerns with the Copper River/Prince William Sound Fish and Game Advisory Committee. We discussed possible population scenarios and outcomes with different levels of harvest. They chose to reduce the West Delta bull moose quota in 2020 in response to these concerns.

In 2021 and 2022, we reduced the cow quota and then the bull quota respectively. Now that we have population data, we know we have at least 464 moose (the number of moose we laid eyes on) on the West Delta, and the population is most likely about 500 moose (estimated number). The objective for this population is 600 to 800 moose.

To grow the population to within this range probably won’t be hard. We aggressively harvest cows relative to other populations. You probably noticed fewer cow permits were issued this year (20 instead of 35 where it’s been for eight of the last nine years). We also reduced the bull harvest in response to the smaller population size. This survey, we saw 22% calves, which is a good number of calves and normal for this area.

In the Martin River area (Unit 6B, east of the Copper River), we estimated 260 moose with a 90% confidence interval of 211 to 309. This is below the 2018 estimate of 420 moose, and the confidence interval for this survey does not overlap with the last. This probably means the population has declined. The upper end of the confidence interval still falls within the population objective (300 to 350) for this area but the point estimate does not.

The last high count was foreshadowed by fall surveys that documented exceptional calf recruitment. No fall surveys have been conducted since that time that could explain this result. Since 2018, harvest quotas have been set using a point estimate of 350 (instead of the point estimate of 420). It is most likely that apparent declines in abundance are related to years that were harder on calves. Hunters can expect next year’s Martin River hunt quota to return to 25 bulls.

Next year, I will continue to prioritize fall surveys so we can understand bull abundance and calf numbers on both sides of the Copper River. Spring surveys will focus on the area between the Ragged Mountains and Icy Bay, including the Bering River (Unit 6A). This was last surveyed in 2020.

We are also interested in how many calf moose are surviving the winter. The last survey showed continued decline and we reduced harvest. We’re hoping to see improvement when we look again!

Charlotte Westing is a wildlife biologist for the Alaska Department of Fish and Game Prince William Sound Area in Cordova.