植树造林
环境科学
生物量(生态学)
农学
草原
土壤碳
生产力
农林复合经营
土壤水分
土壤科学
生物
宏观经济学
经济
作者
Murray R. Davis,A. H. Nordmeyer,David Henley,Michael S. Watt
标识
DOI:10.1111/j.1365-2486.2007.01372.x
摘要
Abstract Biomass and soil carbon (C) and nitrogen (N) were measured in a replicated trial after afforestation of a New Zealand upland subhumid low‐productivity grassland with four tree stockings, including an unplanted control, of Pinus nigra . Total biomass accumulation of P. nigra in the cool, dry and N‐limited environment was low, ranging from 10 to 20 Mg ha −1 dry weight at age 10. Carbon and N accumulation in above‐ and belowground tree biomass ranged between 5–10 and 0.03–0.07 Mg ha −1 , respectively. Soil C, N and bulk density were measured 5 and 10 years after the trees were planted. Soil samples taken at year 5 from between tree rows spaced 5 m apart were considered to be representative of grassland not affected by afforestation. Co‐variance analysis showed that, at year 10, soil C and N concentrations, and soil bulk density and C and N mass were not significantly affected by afforestation. The results are at variance with paired site studies in more humid environments that show soil C declines following afforestation, but confirm other studies and model predictions that show soil C decline in the early stages after afforestation in low‐productivity environments is limited. Afforestation did not affect root biomass of herbaceous species and this may have contributed to the lack of effect on soil C. Although afforestation by itself did not significantly affect soil C and N, over the measurement period soil C concentrations increased, while soil N declined by 450 kg ha −1 . The decline in soil N was confined to lower soil layers and could not be accounted for by uptake in vegetation. The observed decline in soil N is consistent with results of other work in grazed, depleted grassland in the region that indicates losses of soil N occur that cannot be accounted for by pathways directly associated with grazing.
科研通智能强力驱动
Strongly Powered by AbleSci AI