Mule Deer Inherit Spring Migration Distance More Strongly Than Departure Dates

IInherited Routes

Distance and Route Speed Had Far Higher Heritability Than Spring Dates

Bonar and coauthors matched GPS collar tracks to genomic data from 143 adult does in the Piceance Basin of northwestern Colorado, covering 151 spring migrations by does collared from 2012 to 2016. To the authors' knowledge, this was the first empirical evidence of heritable migration behavior in ungulates.

Our rule is to map the places migrating mule deer cross your ground in spring and plan fences, gates and scouting around those crossings in the springs ahead. Choose your watching days from that spring's snow and green-up and never from the previous year's dates.

IISpring Departures

Piceance Does Departed Winter Range From March 8 to June 1

From October to April Piceance deer use one winter range and then migrate south or east to two summer ranges. May 3 was the average start date, though individual starts ranged from March 8 to June 1. Trips averaged 7 days and the longest lasted 67. Does averaged 327.1 meters per hour along the route compared with 71.2 meters per hour on winter range in March.

Migration distance was measured in a straight line between the route's first and last GPS points. Does headed to the eastern summer range averaged 39.8 km (about 24.7 miles). Those headed to the southern range averaged 31.1 km (about 19.3 miles). The authors point to earlier work showing that longer migrations can put mule deer in the path of more highways and fences while shorter migrations can leave them more exposed to hunters.

Using blood samples the team built a genomic relatedness matrix that estimates heritability without a pedigree. They removed deer related above 0.1 to keep a few close relatives from driving the result, leaving 109 does in the reported models. The models all adjusted for age and year.

IIIReturn Trips

Distance and Route Speed Showed Moderate Heritability and High Repeatability

Repeatability is the share of variation that stays with an individual doe from year to year. It followed the same split. Distance had a repeatability of 0.61 and route speed 0.68, both of which the authors call highly repeatable. The start date came in at 0.15 and the end date at 0.21. The authors point to a meta-analysis that found average repeatability of animal behavior was 0.37.

Trip duration broke from that pattern. The abstract puts duration among the moderately heritable traits, yet the results show duration heritability of 0.02 on the observed scale and repeatability of 0.43.

Those models also estimated heritability of 0.27 for hind leg length and 0.12 for ingesta-free body fat in these does. Our read is that migration distance in this herd was about as heritable as leg length in a doe. The 0.61 repeatability for distance divides almost evenly into a genetic share of 0.31 and a non-genetic individual share of 0.30. The authors link high repeatability to strong fidelity to ranges and routes, citing earlier work that found more than 80% fidelity to mule deer migration routes.

IVTiming Drivers

Age, Summer Range and Year Explained Most Variation in Dates

Age, summer range and year explained 77% of start date variation and 69% of end date variation. Additive genes and the non-genetic individual share each explained between 8% and 10%. Snow and plants were not measured directly in the study, so year is the closest stand-in for each spring's conditions in these models. The authors cite earlier work by Lendrum and colleagues in the same basin linking year-to-year differences in spring migration to snow depth and plant phenology. The authors also say low heritability can result from selection wearing down genetic variation or from noise added by a changing environment.

Autumn remains a separate question beyond what this study covered. FDR's Montana migration guide describes a similar weather pull on autumn departures in Montana and found that colder nights, wetter weeks and longer routes increased the odds that collared does left summer range.

VRoute Loss

Inherited Route Traits Would Change Only Over Hundreds of Generations

The authors conclude that wild mule deer could respond to selection from environmental change through microevolutionary shifts in migration. The team also measured evolvability as how far a trait could shift per generation under selection and found it low. Distance evolvability was 0.33% while timing evolvability was centered on zero. According to the paper, rates like these could still lead to noticeable change over hundreds of generations. The authors also suggest that moderately heritable route speed leaves some room for faster migration, which may pay off as conditions become less predictable.

Our read is that a herd losing a route to fences or development will not evolve a replacement within any time scale that matters to a hunter or a manager. The authors cite work proposing that mule deer depend on memory to find their way and this may make migration less flexible, though the authors add that social learning may allow some change in routes or destinations.

VIStudy Reach

These Estimates Describe One Colorado Herd of Does Migrating in Spring

The study looked only at spring migration in adult does and gives no direct answer about bucks or autumn migration, with no resident deer sampled. The intervals were wide. Distance heritability ranged from 0.01 to 0.79 across its interval. Route speed had an equally wide interval, from below 0.01 to 0.87. The authors note that genomic estimates based on mostly unrelated animals can understate heritability. Daughters tend to settle where their mothers live so related does may share habitat that can blur genetic effects with environmental ones, and removing close relatives was meant to reduce that bias. Everything here was observational.