Regional demography of Icelandic rock ptarmigan and its implications for harvest management

The rock ptarmigan is a popular game bird in Iceland, but management of hunting has tended to be controversial.  We were interested in whether regional differences in ptarmigan demography exist and, if so, their implications for harvest management.

We fit integrated population models (IPMs) to monitoring data from six hunting regions of Iceland during 2005 – 2023, and then examined equilibrium and non-equilibrium harvests strategies based on an objective to maximize sustainable harvest.

Survival and reproductive rates tended to be similar among regions, except in the demographically important Northeast, where survival was lower and productivity was higher.  There was a negative relationship between post-breeding age ratio and spring abundance in all regions, although the strength of density dependence varied.  Spring abundance was stable in all regions, although harvest rates tended to decline, and adult survival rates tended to increase.  There was a tendency for temporal patterns of demography to be positively correlated among regions.  Evidence for cyclical patterns in abundance was weak, but this may be an artifact of the relatively short time series.

Sustainable harvest potential varied among regions and was greatest in the East and Northeast.  It appears that harvests during 2005 – 2022 were somewhat less than the sustainable maximum.  Non-equilibrium harvest strategies, in which the allowable harvests depend on spring population size and anticipated productivity, are so-called “bang-bang” strategies, meaning that no or very little harvest is optimal until the population is at or above its maximum level of net production.

Synthesis and applications:  Regional differences in ptarmigan demography warrant different harvest strategies.  Yet the hunting season in Iceland historically has been regulated on a country-wide basis, with population abundance and dynamics in the Northeast playing a key role.  To implement regional harvest management, decision makers must first agree on harvest-management objectives that satisfy most stakeholders, who tend to hold very diverse values.  Moreover, the historical relationship between season length and harvest rate is tenuous at best, yet season length is the primary mechanism traditionally used to control harvest.  The uncertainty in this relationship could be addressed in an adaptive-management framework.

Methods
We relied on ptarmigan monitoring data collected and compiled by the Icelandic Institute of Natural History and made publicly available (https://www.ni.is/en/research/monitoring-and-research/voktun-rjupnastofnsins).   We partitioned data into six regions used for setting hunting seasons in Iceland: E – East, NE – Northeast, NW – Northwest, S – South, W – West, and WF - Westfjords (Fig. 1).  For each region, we compiled transect-based estimates of spring abundance, hunting-season age ratios, spring age ratios (where and when available), total harvests, and the number of days afield by hunters.  For each region, we used the same basic structure for population dynamics of the spring breeding population, expressing change in population size as a function of seasonal survival of juveniles and adults, post-breeding age ratio, and harvest rate. To estimate model parameters, we used Markov Chain Monte Carlo simulation, using JAGS 4.3.0 run in the R computing language with the package jagsUI.