Data from: Quantifying risk of overharvest when implementation is uncertain

1. Sustainable harvest management implies an ability to control harvest rates. This is challenging in systems that have limited control of resources and resource users, which is often the case in small game harvest management. The difference between management strategies and actual harvest bag size (i.e. implementation uncertainty) may be substantial, but few studies have explored this. 2. We investigated how different management strategies and ecosystem variables affected realised harvest of willow ptarmigan (Lagopus lagopus L.) among nine independently-managed, state-owned hunting areas in Central and South Norway during 2008-2015. First, we focused our empirical analysis around three response variables of interest: hunting bag (scaled by area), hunting effort (number of hunting days scaled by area) and hunter efficiency (shot birds per hunting day). Akaike’s Information Criteria (AIC) guided model selection among candidate GLMMs. Then, we used model-averaged parameter estimating from the statistical models in numerical simulations to explore risk of overharvest due to implementation uncertainty. 3. The most parsimonious model explaining hunting bag included total allowable catch (TAC) and willow ptarmigan density. Hunting effort was explained by number of permits sold and type of quota (daily vs. weekly quota). The most parsimonious model describing hunter efficiency only included the effect of willow ptarmigan density. 4. Our results show that managers have only partial control over harvest rates in this system, and that hunters were relatively more efficient and harvest rates higher at low densities. This effect was present for all management strategy scenarios, including when managers adjusted TAC according to population estimates from monitoring programs. 5. Synthesis and applications. Quantifying risk of unsustainable harvest rates under different scenarios enables managers to make informed decisions, when dealing with competing objectives of harvest opportunities and sustainability. The substantial risk of high harvest rates at low densities reported here should encourage frequent use of threshold strategies. This study is one of the first approaches for quantifying implementation uncertainty in small game harvest, and shows how estimates from empirical analyses could be used to quantify risk of overharvest.1. 可持续狩猎管理意味着对狩猎强度的管控能力。在资源与资源使用者管控难度较大的系统中，这一目标极具挑战性——小型猎物狩猎管理场景往往正是此类系统。不同管理策略与实际狩猎收获量（即实施不确定性）之间的差异可能十分显著，但目前鲜有研究对此展开探讨。2. 本研究于2008-2015年间，针对挪威中南部9处独立管理的国有狩猎区域，探究了不同管理策略与生态系统变量对柳雷鸟（willow ptarmigan, Lagopus lagopus L.）实际狩猎收获量的影响。首先，我们将实证分析聚焦于三项核心响应变量：单位面积狩猎收获量、按面积标准化的狩猎天数（狩猎投入）以及猎手狩猎效率（单日狩猎命中鸟类数量）。研究通过赤池信息准则（Akaike’s Information Criteria, AIC）在候选广义线性混合模型（Generalized Linear Mixed Models, GLMMs）中筛选最优模型。随后，我们利用统计模型得到的模型平均参数估计值开展数值模拟，以探究由实施不确定性引发的过度狩猎风险。3. 解释狩猎收获量的最优简约模型包含总允许捕获量（Total Allowable Catch, TAC）与柳雷鸟种群密度两个变量。狩猎投入可由售卖的狩猎许可数量与配额类型（单日配额 vs 周度配额）进行解释。而解释猎手狩猎效率的最优简约模型仅包含柳雷鸟种群密度的影响效应。4. 研究结果表明，在该研究系统中，管理者仅能部分管控狩猎强度；且当种群密度较低时，猎手狩猎效率相对更高，狩猎强度也随之提升。这一效应在所有管理策略场景中均存在，包括管理者依据监测项目得到的种群估计值调整总允许捕获量的场景。5. 综合与应用：量化不同场景下不可持续狩猎强度的风险，可帮助管理者在狩猎机遇与可持续性的多重目标之间做出明智决策。本研究揭示的低密度下高狩猎强度的显著风险，应推动阈值策略的频繁应用。本研究是首批针对小型猎物狩猎实施不确定性开展量化的研究之一，同时展示了如何利用实证分析得到的估计值量化过度狩猎风险。