, Volume 295, Issue 1-2, pp 137-150
Date: 10 May 2007

Spatial heterogeneity of soil nitrogen in a subtropical forest in China

Rent the article at a discount

Rent now

* Final gross prices may vary according to local VAT.

Get Access

Abstract

Spatial variability of soil total nitrogen (N), available N (KCl extractable NH 4 + and NO 3 ), and spatial patterns of N mineralization and nitrification at a stand scale were characterized with geostatistical and univariate analysis. Two extensive soil spatial samplings were conducted in an evergreen broadleaf forest in Sichuan province, southwestern China in June and August 2000. In a study area of 90 × 105 m2, three soil samples were collected from each 5 × 5 m2 plot (n = 378) in June and August, and were analyzed for total N and available N contents. Net N mineralization and nitrification were measured by in situ core incubation and the rates were estimated based on the difference of NH 4 + and NO 3 contents between the two sampling dates. Total N, NH 4 + , and NO 3 were all spatially structured with different semivariogram ranges (from high to low: NH 4 + , NO 3 , and total N). The semivariograms of mineralization and nitrification were not as spatially structured as available N. NH 4 + was the dominant soil inorganic N form in the system. Both NH 4 + and NO 3 affected spatial patterns of soil available N, but their relative importance switched in August, probably due to high nitrification as indicated by greatly increased soil NO 3 content. High spatial auto-correlations (>0.7) were found between available N and NH4+, available N and NO 3 on both sampling dates, as well as total N measurements between both sampling dates. Although significant, the spatial auto-correlation between NH 4 + and NO 3 were generally low. Topography had significant but low correlations with mineralization (r = −0.16) and nitrification (r = −0.14), while soil moisture did not. The large nugget values of the calculated semivariograms and high-semivariance values, particularly for mineralization and nitrification, indicate that some fine scale (<5 m) variability may lie below the threshold for detection in this study.