Legacy effects of a regional drought on aboveground net primary production in six central US grasslands

TitleLegacy effects of a regional drought on aboveground net primary production in six central US grasslands
Publication TypeJournal Article
Year of Publication2018
AuthorsGriffin-Nolan RJ, Carroll CJW, Denton EM, Johnston MK, Collins SL, Smith MD, Knapp AK
JournalPlant Ecology
Volume219
Start Page505
Pagination505-515
Date Published05/2018
Accession NumberSEV.798
Abstract

Global climate models predict increases in
the frequency and severity of drought worldwide,
directly affecting most ecosystem types. Consequently,
drought legacy effects (drought-induced
alterations in ecosystem function postdrought) are
expected to become more common in ecosystems
varying from deserts to grasslands to forests. Drought
legacies in grasslands are usually negative and reduce
ecosystem function, particularly after extended
drought. Moreover, ecosystems that respond strongly
to drought (high sensitivity) might be expected to
exhibit the largest legacy effects the next year, but this
relationship has not been established. We quantified
legacy effects of a severe regional drought in 2012 on
postdrought (2013) aboveground net primary productivity
(ANPP) in six central US grasslands. We
predicted that (1) the magnitude of drought legacy
effects measured in 2013 would be positively related
to the sensitivity of ANPP to the 2012 drought, and (2)
drought legacy effects would be negative (reducing
2013 ANPP relative to that expected given normal
precipitation amounts). The magnitude of legacy
effects measured in 2013 was strongly related
(r2 = 0.88) to the sensitivity of ANPP to the 2012
drought across these six grasslands. However, contrary
to expectations, positive legacy effects (greater
than expected ANPP) were more commonly observed
than negative legacy effects. Thus, while the sensitivity
of ANPP to drought may be a useful predictor of the
magnitude of legacy effects, short-term (1-year)
severe droughts may cause legacy effects that are
more variable than those observed after multiyear
droughts.

DOI10.1007/s11258-018-0813-7