以南疆为例的区域暴恐袭击风险评估

为探寻暴恐袭击高风险区域,推进公共安全精细化管理,以南疆地区为例,采用网格化社会经济数据、遥感数据和兴趣点(POI)数据等多源地理空间数据,从暴恐分子出现可能性、暴恐袭击目标选择偏好和暴恐袭击后果3个方面建立暴恐袭击风险评估指标体系,根据层次分析法(AHP)-Entropy、聚类分区等确定地理空间指标权重,最终得到南疆地区30″×30″(约1 km×1 km)细粒度的暴恐袭击风险的空间分布情况。结果表明:南疆地区的暴恐袭击高风险以上网格主要位于喀什、和田及阿克苏地区的部分市区和县城,与暴恐袭击历史事件的分布较一致,证明风险评估指标体系的合理性和可行性。     

基于介度熵的机场飞行区关键冲突点识别*

为了探究机场航班流量增大和作业车辆增多而导致机场飞行区内频繁出现活动目标冲突情况，提出1种基于介度熵的机场飞行区关键冲突点识别方法。首先，以飞行区内活动目标为节点，活动目标之间的潜在冲突关系为连边，建立飞行区交通态势网络模型。然后，采用介度熵法评价交通态势网络中各活动目标的冲突指数，得到冲突指数最大的活动目标即关键冲突点。最后，将介度熵法与度中心性、介数中心性和邻接信息熵3种识别方法进行对比，并以西安咸阳国际机场为例进行验证。研究结果表明:识别飞行区关键冲突点时，介度熵法较其他3种方法更有效。通过建立飞行区交通态势网络，采用介度熵法识别飞行区关键冲突点并进行调配，一定程度上可以有效预防冲突发生，保障飞行区运行安全。     

Lessons from other disciplines for setting management thresholds for biodiversity conservation

Successful, state-dependent management, in which the goal of management is to maintain a system in a desired state, involves defining the boundaries between different states. Once these boundaries have been defined, managers require a strategic action plan with thresholds that initiate management interventions to either maintain or return the system to a desired state. This approach to management is widely used across diverse industries from agriculture, to medicine, to information technology, but it has only been adopted in conservation management relatively recently. Conservation practitioners have expressed a willingness to integrate this structured approach in their management systems, but they have also voiced concerns, including lack of a robust process for doing so. Given the widespread use of state-dependent management in other fields, we conducted an extensive review of the literature on threshold-based management to gain insight into how and where it is applied and identify potential lessons for conservation management. We identified 22 industries using 75 different methods for setting management thresholds in 843 studies. Methods spanned six broad approaches, including expert driven, statistical, predictive, optimization, experimental, and artificial intelligence methods. The objectives of each of these studies influenced the approaches used, including the methods for setting thresholds and selecting actions, and the number of thresholds set. The role of value judgments in setting thresholds was clear; studies across all industries frequently involved experts in setting thresholds, often accompanied by computational tools to simulate the consequences of proposed thresholds under different conditions. Of the 30 conservation studies examined, two-thirds used expert-driven methods, consistent with prior evidence that experience-based information often drives conservation management decisions. The methods we identified from other disciplines could help conservation decision makers set thresholds for management interventions in different contexts, linking monitoring to management actions and ensuring that conservation interventions are timely and effective.