Responses Of Ecosystem Carbon Cycle To Asymmetric Spring And Autumn Warming In An Old-field Grassland

Resulted from human activities(e.g.,fossil fuel combustion and land use change),atmospheric CO2 concentration has increased by 40% since the industrial revolution,consequently leading to climate warming at the global scale.Climate warming has elevated Earth's land surface temperature by 0.85 oC over the past 130 years(1880-2012)and is expected to cause the extra 1.4~2.6 and 2.6~4.8 oC increments by the middle and late 21 st century,respectively.There is a seasonal asymmetry in climate warming.For example,the magnitude of warming is higher in spring and autumn than summer.The results of model simulations have revealed that spring warming can increase net ecosystem CO2 exchange(NEE)whereas autumn warming decreases NEE.Manipulative experiments are crucial in understanding the responses of ecosystems carbon exchange to asymmetric spring and autumn warming and the potential response mechanisms.In 2012-2015,a manipulative experiment using infrared radiators was conducted to examine the responses of ecosystem carbon cycle to asymmetric spring and autumn warming in an old-field grassland of North China Plain.The results showed that:(1)Across the three years(2012-2015),spring and autumn warming significantly increased soil temperature by 0.28 and 0.45 oC,respectively.In spring(March-May),spring warming substantially stimulated soil temperature by 1.22 oC.Autumn warming significantly elevated soil temperature by 1.93 oC in autumn(September-Novmber).In addition,either spring or autumn warming did not affect soil moisture at the depth of 10 cm and 20 cm.However,autumn warming substantially decreased soil moisture at the depth of 30 cm.No interactions between spring and autumn warming on soil temperature or soil moisture at the depth of 10,20,and 30 cm were found.(2)Across 2012-2015,spring warming increased gross primary productivity(GEP)and net ecosystem CO2 exchange(NEE)by 7.76% and 12.80%,respectively,but did not affect ecosystem respiration(ER).However,autumn warming had no effect on NEE or its two components,GEP and ER.In addition,spring and autumn warming did not interact to affect ecosystem carbon exchange(GEP,NEE,and ER).(3)Across the three years,there were no effects of spring warming on soil respiration(Rs),soil heterotrophic respiration(Rh),or the ratio of Rh/Rs.However,autumn warming significantly suppressed Rs by 12.11% and marginally decreased Rh by 13.57%,but did not affect the ratio of Rh/Rs.There were no interactions between spring and autumn warming on Rs,Rh,or the ratio of Rh/Rs.(4)Spring warming increased aboveground net primary productivity(ANPP)by 21.10% whereas autumn warming did not impact ANPP.There were no effects of spring or autumn warming on belowground net primary productivity(BNPP).In addition,spring warming marginal increased NDVI by 7.67%,but autumn warming did not affect NDVI.No interactions between spring and autumn warming on ANPP,BNPP,or NDVI were detected.(5)The simulations in GEP and NEE could have been attributed to the increases in NDVI and ANPP.However,the decreases in soil moisture at the depth of 30 cm under autumn warming have resulted in the suppressions in Rs and Rh.Our findings have revealed that asymmetric spring and autumn warming can substantially stimulate ecosystem carbon uptake of old-field grassland,but decrease the potential of carbon release from soils under future climate change scenarios.Therefore,asymmetric spring and autumn warming can increase terrestrial ecosystem carbon sequestration,and consequently suppress atmospheric CO2 concentration.Our observations will facilitate model parameterization and calibration in projecting ecosystem carbon cycle in response to asymmetric seasonal warming.