Annual measurements of the diameter growth and litter fall of trees began in 1998 using a 1.0 ha permanent plot beneath a flux tower at the Takayama flux site, central Japan. This opened up an opportunity for studies that compare the interannual variability in tree growth with eddy covariance-based net ecosystem production (NEP). A possible link between multiyear biometric-based net primary production (NPP) and eddy covariance-based NEP was investigated to determine the contribution of autotrophic production and heterotrophic respiration (HR) to the interannual variability of NEP in deciduous forest ecosystems. We also defined the NEP* as the measurable organic matter stored in an ecosystem during the interval in which soil respiration (SR) measurements were taken. The difference of biometric-based NEP* from eddy covariance-based NEP within a given year varied between 55% and 105%. Woody tissue NPP (stems and coarse roots) varied markedly from 0.88 to 1.96 Mg C ha−1 yr−1 during the 8-year study period (1999–2006). Annual woody tissue NPP was positively correlated with eddy covariance-based NEP (r2=0.52, P<0.05). However, neither foliage NPP (r2=0.03) nor HR (r2=0.06) were correlated with eddy covariance-based NEP. Therefore, it was hypothesized that interannual variability in the ecosystem carbon exchange was directly responsible for much of the interannual variation in autotrophic production, especially carbon accumulation in the woody components of the ecosystem. Moreover, similar interannual variations of biometric-based NEP* and eddy covariance-based NEP with small variations in SR and foliage NPP suggest a constant net accumulation of carbon in nonliving pools at the Takayama site.