Pushing precipitation to the extremes in distributed experiments: recommendations for simulating wet and dry years

TitlePushing precipitation to the extremes in distributed experiments: recommendations for simulating wet and dry years
Publication TypeJournal Article
Year of Publication2017
AuthorsKnapp AK, Avolio ML, Beier C, Carroll CJ, Collins SL, Dukes JS, Fraser LH, Griffin-Nolan RJ, Hoover DL, Jentsch A, Loik ME, Phillips RP, Post AK, Sala OE, Slette IJ, Yahdjian L, Smith MD
JournalGlobal Change Biology
Start Page1774
Date Published05/2017
Accession NumberSEV.772

Intensification of the global hydrological cycle, ranging from larger individual precipitation events to more extreme
multiyear droughts, has the potential to cause widespread alterations in ecosystem structure and function. With evidence
that the incidence of extreme precipitation years (defined statistically from historical precipitation records) is
increasing, there is a clear need to identify ecosystems that are most vulnerable to these changes and understand why
some ecosystems are more sensitive to extremes than others. To date, opportunistic studies of naturally occurring
extreme precipitation years, combined with results from a relatively small number of experiments, have provided
limited mechanistic understanding of differences in ecosystem sensitivity, suggesting that new approaches are
needed. Coordinated distributed experiments (CDEs) arrayed across multiple ecosystem types and focused on water
can enhance our understanding of differential ecosystem sensitivity to precipitation extremes, but there are many
design challenges to overcome (e.g., cost, comparability, standardization). Here, we evaluate contemporary experimental
approaches for manipulating precipitation under field conditions to inform the design of ‘Drought-Net’, a relatively
low-cost CDE that simulates extreme precipitation years. A common method for imposing both dry and wet
years is to alter each ambient precipitation event. We endorse this approach for imposing extreme precipitation years
because it simultaneously alters other precipitation characteristics (i.e., event size) consistent with natural precipitation
patterns. However, we do not advocate applying identical treatment levels at all sites – a common approach to
standardization in CDEs. This is because precipitation variability varies >fivefold globally resulting in a wide range
of ecosystem-specific thresholds for defining extreme precipitation years. For CDEs focused on precipitation
extremes, treatments should be based on each site’s past climatic characteristics. This approach, though not often
used by ecologists, allows ecological responses to be directly compared across disparate ecosystems and climates,
facilitating process-level understanding of ecosystem sensitivity to precipitation extremes.