Consequences of herbivore extinction for plant population and community dynamics

人类活动正以前所未有的速度导致物种灭绝。虽然我们经常对其原因有详细的了解，但关于灭绝对生态系统的后果的信息却很缺乏。我和我的研究小组致力于了解大型食草动物的灭绝是如何影响非洲大草原上的植物种群的。我们在肯尼亚莱基皮亚的研究地点构成了研究这种灭绝后果的理想系统。Laikipia由10000平方公里的私人财产组成，其中每个财产都与三种情况中的一种有关，反映了脊椎动物群落典型的渐进灭绝，规模偏大：1)大象（~3000公斤）+黑斑羚（~70公斤）+ dik-dik（~5公斤）；2)黑斑羚+ dik-dik；或者只吃丁丁。我和我的研究小组利用一系列半透性的带电围栏，模拟了上述灭绝情景，以解决大型食草哺乳动物如何影响金合欢树的丰度和分布，以及热带草原火灾频率、营养循环和生物多样性的关键决定因素。具体来说，我们正在努力回答以下问题：1)大型食草哺乳动物的灭绝如何改变热带草原树木的授粉？2)大型食草哺乳动物对热带草原树木种子产生、萌发和补充的影响如何影响树木种群的生长？3)相对于非生物因素（如降雨）的变化，大型食草哺乳动物在推动树木种群增长方面有多重要？为了解决前两个问题，我和我的研究小组正在将实地实验与空间统计和矩阵模型相结合。为了解决第三个问题，我们对肯尼亚中部降雨量的五倍梯度的矩阵模型进行了“地面真实”预测。因此，我和我的研究小组将阐明对生态系统灭绝后果的理解。此外，通过实地实验、长期观察和统计建模相结合，我们将开发和完善预测植物种群和群落灭绝后果的工具。

Human activities are causing extinctions to occur at unprecedented rates. Although we often possess detailed knowledge about their causes, a dearth of information exists regarding the consequences of extinction for ecological systems. My research group and I work to understand how extinctions of large herbivores affect plant populations in African savannas. Our study sites in Laikipia, Kenya comprise an ideal system in which to study the ramifications of such extinctions. Laikipia consists of a 10,000 square km mosaic of private properties, where each property is associated with one of three scenarios reflecting the progressive, size-biased extinction typical of vertebrate communities: 1) elephant (~3000 kg) + impala (~70 kg) + dik-dik (~5 kg); 2) impala + dik-dik; or 3) dik-dik only. Using a series of semi-permeable, electrified fences, my research group and I are simulating the above extinction scenarios to address how large mammalian herbivores influence the abundance and distribution of Acacia trees, critical determinants of savanna fire frequency, nutrient cycles, and biodiversity. Specifically, we are working to answer the following questions: 1) How do extinctions of large mammalian herbivores change pollination of savanna trees?; 2) How do impacts of large mammalian herbivores on seed production, germination, and recruitment of savanna trees influence population growth of trees?; and 3) How important are large mammalian herbivores in driving population growth of trees, relative to variation in abiotic factors (e.g., rainfall)? To address the first two questions, my research group and I are combining field experiments with spatial statistics and matrix models. To address the third question, we are "ground truthing" projections from our matrix models across a five-fold gradient in rainfall in central Kenya. Thus, my research group and I will shed light on understanding the ramifications of extinctions for ecosystems. Moreover, through a combination of field experiments, long-term observations, and statistical modeling, we will develop and refine tools that predict the consequences of extinction for plant populations and communities.