J Plant Ecol ›› 2020, Vol. 13 ›› Issue (5): 621-632.DOI: 10.1093/jpe/rtaa049

• Research Articles • Previous Articles     Next Articles

Riparian plant species differ in sensitivity to both the mean and variance in groundwater stores

Kelly A. Steinberg1,2, *, Kim D. Eichhorst1 and Jennifer A. Rudgers1   

  1. 1 Department of Biology, University of New Mexico, Albuquerque, NM 87131, USA, 2 Bosque School, Bosque Ecosystem Monitoring Program, 4000 Bosque School Road NW, Albuquerque, NM 87120, USA

    *Corresponding author. E-mail: kellyanns09@gmail.com
  • Received:2020-03-07 Revised:2020-05-24 Accepted:2020-08-04 Online:2020-08-10 Published:2020-10-01



Determining the ecological consequences of interactions between slow changes in long-term climate means and amplified variability in climate is an important research frontier in plant ecology. We combined the recent approach of climate sensitivity functions with a revised hydrological ‘bucket model’ to improve predictions on how plant species will respond to changes in the mean and variance of groundwater resources.


We leveraged spatiotemporal variation in long-term datasets of riparian vegetation cover and groundwater levels to build the first groundwater sensitivity functions for common plant species of dryland riparian corridors. Our results demonstrate the value of this approach to identifying which plant species will thrive (or fail) in an increasingly variable climate layered with declining groundwater stores.

Important Findings

Riparian plant species differed in sensitivity to both the mean and variance in groundwater levels. Rio Grande cottonwood (Populus deltoides ssp. wislizenii) cover was predicted to decline with greater inter-annual groundwater variance, while coyote willow (Salix exigua) and other native wetland species were predicted to benefit from greater year-to-year variance. No non-native species were sensitive to groundwater variance, but patterns for Russian olive (Elaeagnus angustifolia) predict declines under deeper mean groundwater tables. Warm air temperatures modulated groundwater sensitivity for cottonwood, which was more sensitive to variability in groundwater in years/sites with warmer maximum temperatures than in cool sites/periods. Cottonwood cover declined most with greater intra-annual coefficients of variation (CV) in groundwater, but was not significantly correlated with inter-annual CV, perhaps due to the short time series (16 years) relative to cottonwood lifespan. In contrast, non-native tamarisk (Tamarix chinensis) cover increased with both intra- and inter-annual CV in groundwater. Altogether, our results predict that changes in groundwater variability and mean will affect riparian plant communities through the differential sensitivities of individual plant species to mean versus variance in groundwater stores.

Key words: climate change, climate variability, riparian, Rio Grande, New Mexico, cottonwood, willow, tamarisk

确定气候的长期缓慢变化与放大的气候变异性之间相互作用的生态后果是植物生态学的一个重要前沿研究。我们将最近的气候敏感性函数方法与修订的水文“水桶模型”相结合,以改进对植物物种如何响应地下水资源平均值和方差变化的预测。利用河岸植被覆盖和地下水位长期数据集的时空变化,构建了旱地河岸廊道常见植物种类的首个地下水敏感性函数。我们的研究结果展示了这种方法的重要性,它可以识别在日益变化的气候层,随着地下水储量的下降,哪些植物物种能够繁茂的生长(或衰落)。研究结果表明,河岸带植物物种对地下水位平均值和方差的敏感性各不相同。杨树(Populus deltoides ssp. wislizenii)随着地下水年际方差的增大,植被覆盖度预计将下降,而狼柳树 (Salix exigua)和其他原生湿地物种预计将受益于更大的年际方差。非本地物种对地下水变化不敏感,但俄罗斯橄榄(Elaeagnus angustifolia)在更深的平均地下水位条件下敏感性将降低。暖空气温度调节了杨树对地下水的敏感性,在最高温度更高的年份/地点,杨树对地下水的变化比在低温地点/时期更敏感。地下水年内变异系数越大,杨树覆盖度下降幅度越大,但与年际变异系数相关性不显著,这可能因为相对于杨树的寿命,这个时间(16 a)太短。与此相反,非本地柳树(Tamarix chinensis)覆盖度随地下水年内和年际变异系数的增加而增加。总之,我们的研究结果预测,地下水变异性和平均值的变化将通过单个植物物种对地下水储量中的平均值和方差的不同敏感性影响河岸带植物群落。

关键词: 气候变化, 气候变异, 河岸, 格兰德河, 新墨西哥, 杨树, 柳树