ENSO,ENSO, DroughtDrought andand thethe ChangingChanging CarbonCarbon CycleCycle NingNing ZengZeng Dept.Dept. AtmosphericAtmospheric andand OceanicOceanic ScienceScience andand EarthEarth SystemSystem ScienceScience InterdisciplinaryInterdisciplinary CenterCenter UniversityUniversity ofof MarylandMaryland Contributors: M. Heimann, A. Mariotti, C. Roedenbeck, P Wetzel, E. Maier-Reimer H. Qian, R. Iacono, E. Munoz CO2 ∆CO2 from climate feedback Enhanced global warming from carbon-climate interaction: Land uptake ∆ Land uptake the C4MIP results --- UMD Earth System Model (CABO) Large differences in land response: Ocean uptake ∆ Ocean uptake interannual variability as a testbed Friedlingstein et al., J. Climate in press AtmosphericAtmospheric CO2CO2 VariabilityVariability 19581958--20002000 Lagged Correlations Corr = 0.6 5 months lag dCO2/SOI --- dCO2/dt Emission -SOI 3-6 months lag Hydrology/SOI th 20 Century Observed Forcing Physical Climate Coupled Carbon Model Transport Atmosphere Reanalysis Atmospheric CO2 Precipitation Precip, Temp Temperature Carbon Radiation (CRU, CMAP, GISS) Exchange wind,vapor Soil Moisture/temperature Vegetation+ Land Soil Carbon Circulation Temperature Salinity Ocean Physics Ocean Chemistry/Biology Reanalysis Goal: To understand the changing carbon cycle in the 20th Century The VEgetation-Global Atmosphere-Soil Model (VEGAS) Atmospheric CO2 4 Plant Functional Types: Broadleaf tree Photosynthesis Autotrophic Needleleaf tree respiration C3 Grass (cold) C4 Grass (warm) NPP 3 Vegetation carbon pools: Carbon Leaf allocation Heterotrophic Root respiration (Rh) Wood Turnover 3 Soil carbon pools: Fast Intermediate Slow VEGASVEGAS IIII Photosynthesis:Photosynthesis: LightLight (PAR,(PAR, LAI,LAI, Height),Height), soilsoil moisture,moisture, temperature,temperature, CO2CO2 Respiration:Respiration: Temperature,Temperature, soilsoil moisture,moisture, lowerlower soilsoil poolspools slowerslower decaydecay Competition:Competition: NetNet growth,growth, shadingshading =>=> fractionalfractional covercover Fire:Fire: moisture,moisture, fuelfuel load,load, PFTPFT dependentdependent resistanceresistance Wetland/CH4:Wetland/CH4: moisture,moisture, topographytopography gradientgradient MechanismsMechanisms ofof InterannualInterannual VariabilityVariability II Modeled land-atmo flux vs. MLO CO2 growth rate Terrestrial carbon model forced by observed climate variability LandLand vs.vs. oceanocean fluxesfluxes Land: VEGAS Land contributes to most of the interannual variability, Ocean: HAMOCC with significant contribution from ocean Modeling results supported by in-situ data and inversion El Nino 97/98 VEGAS Inversion Roedenbeck 2003 Spatial patterns from PC1 PC2 multi-variate EOF analysis 1901 1995 C flux Tropics during El Nino 1) Drier and warmer conditions coexist at tropical locations Precip 1+) Drier and warmer across much of tropical land during El Nino 2) Less precip => Less growth (lower NPP) Temp and more fire => Less C uptake Higher T => more respiration (higher Rh) => more C release Why CO2 correlates so well with ENSO: A ‘conspiracy’ theory Tropics during El Nino Precipitation Temperature decrease increase Out of phase NPP decrease Rh increase Additive Land-atmo flux (Rh-NPP) increase + Spatially coherent climate anomalies Large land-atmo C flux Fire carbon flux during 1997-98 El Nino CASA VEGAS Mean 1997-98 El Nino Anomalies Input: climate only Input: satellite fire counts, climate RecentRecent AnomalousAnomalous growthgrowth inin CO2CO2 1980 2003 Proposed explanations: 1. Fire in Siberia, North America, and other places 2. Accelerated carbon emission from China, India 3. Mid-latitude drought MidMid--latitudelatitude Drought:Drought: 19981998--20022002 RecentRecent AnomalousAnomalous growthgrowth inin CO2CO2 DryingDrying oror Warming?Warming? FireFire inin thethe US:US: NaturalNatural vsvs anthropogenicanthropogenic factorsfactors Model Observation WhatWhat happenedhappened inin AustraliaAustralia 19981998--20022002 ?? l Observation ConclusionsConclusions Drought is the major source of CO2 variability, whether it’s tropical (ENSO) or midlatitude. The high correlation between CO2 and ENSO is mainly due to a ‘conspiracy’ between climate anomalies and plant/soil physiology. – This conspiracy makes it difficult to narrow down the relative roles of NPP, Rh, and fire based on observations alone, yet models do not agree. Recent anomalously large CO2 growth can be explained by a (so far) unusual midlatitude drought, a possible glimpse into a warmer world Understanding the mechanisms and processes underlying such interactions provides crucial insight into the fate of anthropogenic CO2 and the degree of future climate change.