Variations in leaf area index in northern and eastern North America over the past 21,000 years: a data-model comparison

摘要: Over the past 10 years, there has been a growing debate about relative importance of late-Quaternary variations in climate and CO2 as drivers changes vegetation structure. In this study, we compare new datasets pollen-reconstructed leaf area index values (LAIs) for northern eastern North America with BIOME4 simulations regional histories order to identify sources data-model disagreements assess influence Of on LAI. BIOME4, an equilibrium biogeochemistry-biogeography model, was forced by from Hadley Centre Unified GCM simulate LAI at 1000-year intervals 21,000 years six regions America. included sensitivity experiments investigate effects Pollen-reconstructed LAIs were based modern analog technique cross-referencing surface pollen samples against recent observations MODIS sensor. biased towards too high due lower-than-present atmospheric CO2; suggested that bias ranged 13 +/- 7% during Lateglacial period (21-12 ka) 7 4% Holocene (11-0 ka).:Simulated reconstructed trends agreed well western Canada, southeastern Canada/northern US, America, but significant discrepancies observed Midwestern LIS, Alaska, US. Causes varied region (1) inaccurate simulation prairie-forest ecotone US too-wet simulations; (2) combination too-high simulated tundra too-low Alaskan lacked analogs; (3) scrublands Sensitivity tests indicated driven primarily climate, secondary influence. which allowed vary while constant better than did constant. Our work emphasizes need models incorporate both dynamics We show is viable paleo-vegetation proxy can be used benchmark rigorous, quantitative comparisons focusing history, particularly densities fossil data.