| Due to the complexity of tree species composition, topography and soil parent material in forests, the soil heterogeneity was universal in forests. Research on soil nutrients heterogeneity which can accurately to estimate soil C storage, and to reflect the theoretical basis in mechanism of species coexistence and different plants which were affected by nutrient niche separation. The main objective of the study was to compare spatial heterogeneity of soil nutrient between Choerospondias axillaris deciduous broadleaved forest and Lithocarpus glaber-Cylobalanopsis glauca evergreen broadleaved forest which had different dominant tree species. We established long-term observation sample area with 100 m×100 m, and then to divide one hundred of 10m×10m sampling subplots. Soil samples were retrieved from the 0-10cm,10-20cm and 20-30cm soil layer in each subplot before being transported back to the laboratory which were determinated soil organic carbon (SOC) and total nitrogen (TN) contents. Spatial heterogeneity of SOC and TN contents were determined by geostatistical analysis. Topography, soil, and stand factors were used as explanatory variables which were performed for SOC and TN contents. The results showed as follows:(1) SOC content was 18.61g kg-1 and TN content was 1.63g kg-1 in 0-30cm of the L. glaber-C. glauca forest, which was higher that SOC content was 16.68 g kg-1 and TN content was 1.44g kg-1 in the 0-30cm of the C. axillaris forest. A coefficient of variation (CV) in two forests were moderate. The CV of SOC was between 27.07% and 43.75%, the CV of TN content was between 20.89% and 25.35%, which were increased by soil layer deeper in the C. axillaris forest. The CV of SOC was between 35.4% and 33.76% and TN content was between 33.68% and 38.4%, which was only SOC decreased by soil deeper in the L. glaber-C. glauca forest.(2) SOC fitted well with the Exponential mode of semivariance function model, and the spatial autocorrelation range between 12.300m and 28.800m; TN fitted well with the Exponential and Spherical model, and the spatial autocorrelation range between 16.8m and 27.9m in the C. axillaris forest. In the L. glaber-C. glauca forest, SOC fitted well with the Exponential and Gaussian model of semivariance function model, and the spatial autocorrelation range between 102.191m and 232.200m; and TN fitted well with the Spherical model and the spatial autocorrelation range between 94m and 105.2m. The fractal dimensions in the C. axillaris forest were bigger that L. glaber-C. glauca forest, reflecting a more complicated partly spatial pattern in C. axillaris forest. SOC and TN showed strong spatial autocorrelation in the C. axillaris forest, mainly caused by structural factors. SOC and TN showed moderate spatial autocorrelation in the L. glaber-C. glauca forest, mainly caused by structural factors and random factors. Kriging showed that SOC revealed plaque form in two forests, which had more cracked plaque form. TN revealed belt form in two forests which has gradient significantly than SOC in two forests. SOC and TN contents were higher being observed in low-lying land; and were lower being observed in the ridge.(3) The spatial heterogeneity of SOC and TN contents were impacted by topography, soil and stand factors. In the study, topography was able to explain SOC in 0-10cm,10-20cm and 20-30cm with 4.6%,12.15% and 10%, to explain TN in 0-10cm,10-20cm and 20-30cm with 11.44%,13.61% and 14.19% respectively in the C. axillaris forest. The interpretation ratios were higher than in the L. glaber-C. glauca forest. Soil factors were able to explain SOC in 0-10cm,10-20cm and 20-30cm with 7.13% ,6.47% and 0.00%, to explain TN in 0-10cm,10-20cm and 20-30cm with 21.00%,8.26% and 10.98% respectively in the C. axillaris forest. The interpretation ratio of soil factors were able to explain SOC in 10-20cm and to explain TN in 0-10cm,10-20cm,20-30cm which were higher than in the L. glaber-C. glauca forest, in other soil layers were lower than in the L. glaber-C. glauca forest.(4) Be different with topography and soil factors, stand factors were able to explain SOC in 0-10cm,10-20cm,20-30cm with 20.83%,11.66% and 14%, and to explain TN with 10.41%,14.77% and 29.56% respectively in the L. glaber-C. glauca forest, which were higher interpretation ratios that in the C. axillaris forest. |
PDF Full Text Request