Spring may come earlier to North American forests

Trees play an impor­tant role in tak­ing up car­bon diox­ide from the atmos­phere, so researchers led by David Med­vigy, assis­tant pro­fes­sor in Princeton’s depart­ment of geo­sciences, wanted to eval­u­ate pre­dic­tions of spring bud­burst — when decid­u­ous trees push out new growth after months of win­ter dor­mancy — from mod­els that pre­dict how car­bon emis­sions will impact global temperatures.

The date of bud­burst affects how much car­bon diox­ide is taken up each year, yet most cli­mate mod­els have used overly sim­plis­tic schemes for rep­re­sent­ing spring bud­burst, mod­el­ing for exam­ple a sin­gle species of tree to rep­re­sent all the trees in a geo­graphic region.

In 2012, the Prince­ton team pub­lished a new model that relied on warm­ing tem­per­a­tures and the wan­ing num­ber of cold days to pre­dict spring bud­burst. The model, which was pub­lished in the Jour­nal of Geo­phys­i­cal Research, proved accu­rate when com­pared to data on actual bud­burst in the north­east­ern United States.

In the cur­rent paper pub­lished online in Geo­phys­i­cal Research Let­ters, Med­vigy and his col­leagues tested the model against a broader set of obser­va­tions col­lected by the USA National Phe­nol­ogy Net­work, a nation-wide tree ecol­ogy mon­i­tor­ing net­work con­sist­ing of fed­eral agen­cies, edu­ca­tional insti­tu­tions and cit­i­zen sci­en­tists. The team incor­po­rated the 2012 model into pre­dic­tions of future bud­burst based on four pos­si­ble cli­mate sce­nar­ios used in plan­ning exer­cises by the Inter­gov­ern­men­tal Panel on Cli­mate Change.

The researchers included Su-Jong Jeong, a post­doc­toral research asso­ciate in Geo­sciences, along with Elena Shevli­akova, a senior cli­mate mod­eler, and Sergey Maly­shev, a pro­fes­sional spe­cial­ist, both in the Depart­ment of Ecol­ogy and Evo­lu­tion­ary Biol­ogy and asso­ci­ated with the U.S. National Oceanic and Atmos­pheric Administration’s Geo­phys­i­cal Fluid Dynam­ics Laboratory.

The team esti­mated that, com­pared to the late 20th cen­tury, red maple bud­burst will occur 8 to 40 days ear­lier, depend­ing on the part of the coun­try, by the year 2100. They found that the north­ern parts of the United States will have more pro­nounced changes than the south­ern parts, with the largest changes occur­ring in Maine, New York, Michi­gan, and Wisconsin.

The researchers also eval­u­ated how warm­ing tem­per­a­tures could affect the bud­burst date of dif­fer­ent species of tree. They found that bud­burst shifted to ear­lier in the year in both early-budding trees such as com­mon aspen (Pop­u­lus tremu­loides) and late-budding trees such as red maple (Acer rubrum), but that the effect was greater in the late-budding trees and that over time the dif­fer­ences in bud­ding dates narrowed.

The researchers noted that early bud­burst may give decid­u­ous trees, such as oaks and maples, a com­pet­i­tive advan­tage over ever­green trees such as pines and hem­locks. With decid­u­ous trees grow­ing for longer peri­ods of the year, they may begin to out­strip growth of ever­greens, lead­ing to last­ing changes in for­est make-up.

The researchers fur­ther pre­dicted that warm­ing will trig­ger a speed-up of the spring “green­wave,” or bud­burst that moves from south to north across the con­ti­nent dur­ing the spring.

The find­ing is also inter­est­ing from the stand­point of future changes in spring­time weather, said Med­vigy, because bud­burst causes an abrupt change in how quickly energy, water and pol­lu­tants are exchanged between the land and the atmos­phere. Once the leaves come out, energy from the sun is increas­ingly used to evap­o­rate water from the leaves rather than to heat up the sur­face. This can lead to changes in daily tem­per­a­ture ranges, sur­face humid­ity, stream­flow, and even nutri­ent loss from ecosys­tems, accord­ing to Medvigy.

Cita­tion:

Jeong, Su-Jong, David Med­vigy, Elena Shevli­akova, and Sergey Maly­shev. 2013. Pre­dict­ing changes in tem­per­ate for­est bud­burst using continental-scale obser­va­tions and mod­els. Geo­phys­i­cal Research Let­ters. Arti­cle first pub­lished online: Jan. 25, 2013. DOI: 10.1029/2012GL054431

This research was sup­ported by award NA08OAR4320752 from the National Oceanic and Atmos­pheric Admin­is­tra­tion, U.S. Depart­ment of Commerce.