Data from: A characterization of autumn nocturnal migration detected by weather surveillance radars in the northeastern US

Billions of birds migrate at night over North America each year. However, few studies have described the phenology of these movements, such as magnitudes, directions, and speeds, for more than one migration season and at regional scales. In this study, we characterize density, direction, and speed of nocturnally migrating birds using data from 13 weather surveillance radars in the autumns of 2010 and 2011 in the northeastern US. After screening radar data to remove precipitation, we applied a recently developed algorithm for characterizing velocity profiles with previously developed methods to document bird migration. Many hourly radar scans contained wind-borne "contamination," and these scans also exhibited generally low overall reflectivities. Hourly scans dominated by birds showed nightly and seasonal patterns that differed markedly from those of low reflectivity scans. Bird migration occurred during many nights, but a smaller number of nights with large movements of birds defined regional nocturnal migration. Densities varied by date, time, and location but peaked in the second and third deciles of night during the autumn period when the most birds were migrating. Migration track (the direction to which birds moved) shifted within nights from south southwesterly to southwesterly during the seasonal migration peaks; this shift was not consistent with a similar shift in wind direction. Migration speeds varied within nights, although not closely with wind speed. Airspeeds increased during the night; groundspeeds were highest between the second and third deciles of night, when the greatest density of birds was migrating. Airspeeds and groundspeeds increased during the fall season, although groundspeeds fluctuated considerably with prevailing winds. Significant positive correlations characterized relationships among bird densities at southern coastal radar stations and northern inland radar stations. The quantitative descriptions of broad-scale nocturnal migration patterns presented here will be essential for biological and conservation applications. These descriptions help to define migration phenology in time and space, fill knowledge gaps in avian annual cycles, and are useful for monitoring long-term population trends of migrants. Furthermore, these descriptions will aid in assessing potential risks to migrants, particularly from structures with which birds collide and artificial lighting that disorients migrants.每年，数十亿只鸟类于北美上空夜间迁徙。然而，鲜有研究能在多迁徙季与区域尺度上，阐明这类迁徙活动的物候特征，包括迁徙规模、方向与速度。本研究依托美国东北部2010与2011年秋季的13部天气监视雷达（weather surveillance radars）数据，解析夜间迁徙鸟类的种群密度、迁徙方向与飞行速度。在对雷达数据进行降水滤除预处理后，我们结合已有的鸟类迁徙记录方法，采用新近开发的速度剖面解析算法开展分析。大量逐时雷达扫描数据存在风致“污染”，且这类数据的整体反射率普遍偏低。以鸟类信号为主的逐时扫描数据所呈现的夜间与季节迁徙模式，与低反射率扫描数据存在显著差异。多数夜间均有鸟类迁徙，但仅少数夜间出现大规模鸟类移动，以此构成区域夜间迁徙事件。种群密度随日期、时段与区位发生变化，在秋季迁徙高峰时段的夜间第二至第三个十分位区间达到峰值，此时迁徙鸟类数量最多。在季节性迁徙峰值阶段，夜间迁徙轨迹（即鸟类飞行方向）会从南西南向逐步转为西南向，这一变化与风向的同类偏移并不一致。夜间迁徙速度存在波动，但与风速的相关性并不显著。夜间空速（airspeed）随时间推移逐渐升高；地速（groundspeed）则在夜间第二至第三个十分位区间达到峰值，此时鸟类迁徙密度最高。整个秋季迁徙期间，空速与地速均呈上升趋势，但地速会随盛行风出现显著波动。南部沿海雷达站与北部内陆雷达站记录的鸟类密度之间存在显著正相关关系。本研究针对大范围夜间迁徙模式所开展的定量描述，可为鸟类学研究与生物保护实践提供关键支撑。这些描述可明确迁徙活动在时间与空间维度上的物候特征，填补鸟类年度生活史研究中的知识空白，同时有助于监测迁徙鸟类种群的长期动态趋势。此外，本研究的定量结果还可助力评估迁徙鸟类面临的潜在威胁，尤其是鸟类撞击人工设施以及灯光干扰导致迷失方向所带来的风险。