Within-field demography of weeds exhibiting wind-mediated long distance seed movement can be largely governed by extra-field seed-source populations. Thus, for these species, a clear understanding of the temporal and spatial patterns of seed dispersal can benefit the development of effective management options. The spatial distribution of the seeds released from the onset of fruiting, in early summer, to the onset of the rainy season, in early autumn, was studied during 2 years at a Mediterranean-climate locality in Erigeron bonariensis L. (Hairy fleabane), a wind-dispersed invasive weed inhabiting ruderal environments and crop fields. Each year, a small source population was established in an open field in SW Spain and seed traps were arranged in the eight cardinal directions at distances up to 100 m (year 1), or in the NE and NW directions at distances up to 65 m (year 2). Counts of trapped seeds were carried out at 4–6 day intervals and the number of seeds released by the source population was estimated each year in most census dates. Four empirical dispersal models based on either thin-tailed or fat-tailed density kernels were tested using year 1 data for their ability to represent the spatial distribution of seeds. To test for anisotropic dispersal, model parameters were allowed to vary according to the wind pattern in each cardinal direction. Based on information criteria, a model including a fat-tailed, Log-hyperbolic secant kernel showing parameter response to the wind pattern, highlighting striking anisotropic dispersal, was selected and evaluated using year 2 data. Distance percentiles 50 and 80 attained by the seed crop released in year 1 season were modeled at 530 m and 10,498 m, respectively. The opposite quadrants encompassing the dominant downwind (N-NE) and upwind directions (S-SW) received 52.5% and 10.8% of seeds. The year 1 population, consisting of 85 plants, generated a modeled seed rain of at least 10 seeds m⁻² up to 200 m downwind. Implications of results for management of this herbicide resistance-prone species are discussed.

表现出风介导的长距离种子运动的杂草的田间人口统计学在很大程度上可以由田外种子源种群控制。因此，对于这些物种，清楚了解种子传播的时空模式有助于制定有效的管理方案。在地中海气候地区Erigeron bonariensis 的2 年内，研究了从初夏结果到雨季开始到初秋释放的种子的空间分布L. (Hairy fleabane)，一种栖息于荒凉环境和农田的随风传播的侵入性杂草。每年，在西班牙西南部的开阔地中建立一个小源种群，并在 8 个主要方向上距离达 100 m（第 1 年），或在 NE 和 NW 方向上布置达 65 m 的种子陷阱（第 2 年）。被困种子的计数每隔 4-6 天进行一次，并且在大多数普查日期中每年估计源种群释放的种子数量。使用第 1 年的数据测试了基于细尾或肥尾密度内核的四种经验扩散模型表示种子空间分布的能力。为了测试各向异性扩散，允许模型参数根据每个基本方向的风模式而变化。根据信息标准，使用第 2 年的数据选择并评估了一个模型，该模型包括一个显示对风模式的参数响应的肥尾对数双曲正割核，突出了显着的各向异性扩散。第 1 年释放的种子作物达到的距离百分位数 50 和 80 分别模拟为 530 m 和 10,498 m。包含主要顺风 (N-NE) 和逆风方向 (S-SW) 的相反象限收到了 52.5% 和 10.8% 的种子。由 85 株植物组成的第 1 年种群产生了至少 10 粒种子的模拟种子雨 第 1 年释放的种子作物达到的距离百分位数 50 和 80 分别模拟为 530 m 和 10,498 m。包含主要顺风 (N-NE) 和逆风方向 (S-SW) 的相反象限收到了 52.5% 和 10.8% 的种子。由 85 株植物组成的第 1 年种群产生了至少 10 粒种子的模拟种子雨 第 1 年释放的种子作物达到的距离百分位数 50 和 80 分别模拟为 530 m 和 10,498 m。包含主要顺风 (N-NE) 和逆风方向 (S-SW) 的相反象限收到了 52.5% 和 10.8% 的种子。由 85 株植物组成的第 1 年种群产生了至少 10 粒种子的模拟种子雨⁻²最多 200 m 顺风。讨论了结果对这种易产生除草剂抗性物种的管理的影响。