Data from: Comparative water use by maize, perennial crops, restored prairie, and poplar trees in the US Midwest

Water use by plant communities across years of varying water availability
indicates how terrestrial water balances will respond to climate change
and variability as well as to land cover change. Perennial biofuel crops,
likely grown mainly on marginal lands of limited water availability,
provide an example of a potentially extensive future land cover
conversion. We measured growing-season evapotranspiration (ET) based on
daily changes in soil profile water contents in five perennial
systems—switchgrass, miscanthus, native grasses, restored prairie, and
hybrid poplar—and in annual maize (corn) in a temperate humid climate
(Michigan, USA). Three study years (2010, 2011 and 2013) had normal
growing-season rainfall (480–610 mm) whereas 2012 was a drought year (210
mm). Over all four years, mean (±SEM) growing-season ET for perennial
systems did not greatly differ from corn (496 ± 21 mm), averaging 559
(±14), 458 (±31), 573 (±37), 519 (±30), and 492 (±58) mm for switchgrass,
miscanthus, native grasses, prairie, and poplar, respectively. Differences
in biomass production largely determined variation in water use efficiency
(WUE). Miscanthus had the highest WUE in both normal and drought years
(52–67 and 43 kg dry biomass ha−1 mm−1, respectively), followed by maize
(40–59 and 29 kg ha−1 mm−1); the native grasses and prairie were lower and
poplar was intermediate. That measured water use by perennial systems was
similar to maize across normal and drought years contrasts with earlier
modeling studies and suggests that rain-fed perennial biomass crops in
this climate have little impact on landscape water balances, whether
replacing rain-fed maize on arable lands or successional vegetation on
marginal lands. Results also suggest that crop ET rates, and thus
groundwater recharge, streamflow, and lake levels, may be less sensitive
to climate change than has been assumed.