Data from: Seed predation has the potential to drive a rare plant to extinction

1. Pre-dispersal seed predation is sometimes considered unlikely to dramatically affect plant population growth because plants are generally expected to produce more seeds than there are safe sites for germination. Lupinus constancei is a rare herb of limited distribution, with fewer than 400 reproductive individuals restricted to a single square kilometre of north-western California, USA. In addition to the vulnerability resulting from its extremely small population size, L. constancei faces heavy seed predation by small mammals.
2. As a stop-gap measure to prevent population decline, managers began covering a large number of the reproductive plants with herbivory exclosures in 2003, but the population-level effects of seed predation and the effectiveness of this caging treatment were unknown. We used ten years of demographic data to compare the population dynamics of plants inside herbivory exclosures with those sustaining ambient rodent seed predation.
3. We found that the stochastic population growth rate would be robust without seed predation (λs = 1.17), but without continued human intervention (i.e. use of exclosures), the current rate of predation would result in a decline towards extinction (λs = 0.92).
4. After our study concluded, high mortality due to two extreme winter droughts followed by a wildland fire reduced the number of reproductive plants to ~103, making extinction of L. constancei more likely.
5. Synthesis and applications. The prevalence of consumer-driven population decline is largely unknown, but this study demonstrates that pre-dispersal seed predation by rodents can have powerful population-level effects, and represents one set of conditions under which consumer pressure has the potential to drive plant extinction. However, with continued management to limit the effects of seed predation in the short-term and investigation into the ultimate drivers of this high seed predation rate in the long-term, the Lassics lupine population could be restored to a robust rate of growth.

1. 种子扩散前捕食（pre-dispersal seed predation）有时被认为不太可能对植物种群增长产生显著影响，因为学界普遍认为植物产生的种子数量会多于可供萌发的安全位点。康斯延氏羽扇豆（Lupinus constancei）是一种分布受限的稀有草本植物，全球仅存不足400株繁殖个体，局限分布于美国加利福尼亚州西北部的1平方公里范围内。除种群规模极小带来的固有脆弱性外，该物种还面临小型哺乳动物对种子的严重捕食压力。
2. 作为防止种群衰退的应急干预措施，管理人员于2003年开始使用植食动物排除笼（herbivory exclosures）覆盖大量繁殖植株，但此前学界对种子捕食的种群水平效应以及该笼养措施的有效性尚不明确。本研究依托十年的种群统计数据，对比了植食动物排除笼内植株与暴露在正常啮齿类种子捕食压力下植株的种群动态差异。
3. 研究结果显示，若无种子捕食干扰，随机种群增长率（stochastic population growth rate）将保持稳健（λs = 1.17）；但如果不持续采取人工干预措施（即使用植食动物排除笼），当前的捕食压力将导致种群向灭绝方向衰退（λs = 0.92）。
4. 本研究结束后，两场极端冬季干旱引发的高死亡率以及随后的野外火灾，将繁殖植株数量缩减至约103株，使得康斯延氏羽扇豆的灭绝风险进一步升高。
5. 综合与应用。捕食者驱动的种群衰退现象的普遍性目前仍不清晰，但本研究证实，啮齿类造成的种子扩散前捕食可对种群水平产生强烈影响，且代表了一类捕食者压力有可能推动植物物种灭绝的情形。不过，若短期持续开展管理以限制种子捕食的影响，并长期探究该高种子捕食率的根本驱动因素，拉塞克斯羽扇豆（Lassics lupine）种群有望恢复至稳健的增长速率。