| Forest ecosystem plays an irreplaceable role in the global carbon cycle.Forest net primary productivity(NPP)is an important parameter for assessing the material and energy cycle of forest ecosystems,and it also influences the carbon sink capacity of forest ecosystems.The temporal variation and uncertainty of NPP in forest ecosystem at the local scale have rarely been reported,which limits our understanding of the control mechanisms in NPP and its interannual fluctuations at the local scale and increases the difficulty in assessment of the uncertainty of forest NPP estimation at regional and global scales.In this study,the leaf and woody tissues of the forest ecosystem were monitored continuously from 2008 to 2019,and the interannual variation in herb and fine root NPP were monitored for a few years in the Maoershan natural secondary forest in Northeast China.The absolute and relative uncertainties of NPP were estimated according to the error propagation.The main results indicated that:(1)During 2008-2019,the mean(±standard deviation)annual total litterfall was 4.93 ±0.36 t hm-2 yr-1.The total and leaf litterfall showed significant increasing trends of 1.42%yr-1 and 1.17%yr-1,respectively.The coefficients of interannual variation in total(7.3%)and leaf(5.6%)litterfall were substantially lower than those in the stand basal area increment(21.0%)and gross primary productivity(10.2%),which indicated that leaf productivity was much more stable than woody tissue productivity.Annual production of canopy litterfall was jointly controlled by climatic factors(i.e.,air temperature,thermal time,precipitation,and radiation)and biotic(i.e.,basal area and species richness)factors.Differences in the responses of leaf litterfall to climate perturbation among tree species and plots partially offset one another and resulted in low interannual fluctuations in total and leaf litterfall for the entire forest canopy.Nevertheless,significant climate legacy effects were detected in all components of the litterfall,with the strongest one occurring in the woody tissue component.Multiple regression based on variance decomposition showed that the most important factor driving the variability of the total litterfall and its non-leaf components was climate legacy effect(67.2%to 86.6%),while those for the leaf litterfall were basal area(40.4%)and species richness(38.3%).(2)During 2008-2019,the annual total woody tissue(branch,stem,belowground)increment(WI)ranged between 3.18 t hm-2 yr-1(2008)and 6.50 t hm-2 yr-1(2013),with no significant increasing trend and a mean of 5.10±1.00 t hm-2 yr-1.The interannual variation of WI(19.70%)was smaller than that of spatial(inter-plot)variation(29.13%).Climate was an important factor affecting the interannual variability of WI.Obvious climate legacy effect was detected in woody tissue,but annual mean temperature and precipitation in the concurrent year had not affected on WI.This result further proved the buffering effect of long-lived tissues and organs with a slow turnover rate on climate concurrent responses.Multivariate regression based on variance decomposition showed that the increment of total woody tissue,stem and branch were mainly affected by basal area(61.4%to 75.0%),while belowground WI was mainly affected by climate legacy effect(50.0%)and species richness(49.4%).The interannual synchronization of WI and litterfall was weak,and the temporal stability of WI was lower,which indicated that canopy had a higher priority of carbon allocation than woody tissues.The temporal variation of WI in each tree species was also much higher than that of total WI,which was consistent with the results on canopy litterfall,indicating that the responses of different tree species to climate factors offset each other and enhanced the stability of NPP in this ecosystem.(3)The mean NPP of total litterfall and total woody tissues were very close in the natural secondary forest during the 12 years,they were 2.34±0.43 t C hm-2 yr-1 and 2.33 ±0.48 t C hm-2 yr-1,respectively.The NPP of shrub calculated by the age of 10 years was 0.08±0.06 t C hm-2 yr-1.The total NPP of herb was 0.31±0.09 t C hm-2 yr-1,of which the NPP of annual and perennial herbs were 0.22±0.09 t C hm-2 yr-1,while the NPP of early-spring plants was only 0.09±0.03 t C hm-2 yr-1.The fine root NPP estimated by the decision matrix method(3.75±1.30 t C hm-2 yr-1)was much higher than that by ingrowth core method(1.18 ±0.41 t C hm-2 yr-1),with a mean of 2.47±0.96 t C hm-2 yr-1.The total NPP of the ecosystem was 7.53±1.16 t C hm-2 yr-1 by summing the components.The NPP in canopy and woody tissues accounted for 31.08%and 30.94%,respectively;the NPP in shrub was the lowest(1.06%);the proportion of NPP in herb was only 4.06%;the proportion of fine root NPP was the largest(32.80%).(4)The absolute uncertainty of NPP in the secondary forest was 1.16 t C hm-2 yr-1.The absolute uncertainty of the NPP of each component was in the order of fine root>woody tissues>canopy litterfall>herb>shrub,while the relative uncertainty of NPP in each component was in the order of shrub>fine root>herb>canopy litterfall>woody tissues.The relative uncertainty of the canopy litterfall NPP was 18.38%,which was mainly from litterfall production(92.22～98.65%).The relative uncertainty of NPP in woody tissues caused by measurement error of diameter at breast height was 16.58%.Among the components of woody tissues,the relative uncertainty of NPP in branch(35.57%)was significantly higher than that in stem(23.97%)and belowground(24.59%).The relative uncertainty of shrub NPP was as high as 74.04%;the relative uncertainty of NPP of herb was 38.01%;and that of fine root NPP was 39.13%.It was concluded that the long-lived tissues and organs had a buffering effect of on concurrent climate responses,the carbon allocation priority of canopy(mainly leaves)was higher than that of woody tissue,and using tree ring width as a parameter may overestimate the sensitivity of production of trees or forests to climate change.The temporal stability in NPP of the ecosystem was enhanced by increasing species diversity.The NPP of the Maoershan natural secondary forest was 7.53 ±1.16 t C hm2 yr-1,and the uncertainty of NPP could be effectively reduced by repeated measurement for many years.This study helps deepen our understanding of the interannual variation and uncertainty of NPP in forests at local scale,and also provides theoretical basis and methodology support for accurate assessment of the ecosystem carbon budget. |
PDF Full Text Request