Biochar Addition Alters C: N: P Stoichiometry in Moss Crust-Soil Continuum in Gurbantünggüt Desert

Abstract

The biogeochemical cycling of soil elements in ecosystems has changed under global changes, including nutrients essential for plant growth. The application of biochar can improve the utilization of soil nutrients by plants and change the stoichiometry of carbon (C), nitrogen (N), and phosphorus (P) in plants and soil. However, the response of ecological stoichiometry in a moss crust-soil continuum to local plant biochar addition in a desert ecosystem has not been comprehensively explored. Here, we conducted a four-level Seriphidium terrae-albae biochar addition experiment (CK, 0 t ha−1; T1, 3.185 t ha−1; T2, 6.37 t ha−1; T3, 12.74 t ha−1) to elucidate the influence of biochar input on C: N: P stoichiometry in moss crusts (surface) and their underlying soil (subsurface). The results showed that biochar addition significantly affected the C, N, and P both of moss crusts and their underlying soil (p < 0.001). Biochar addition increased soil C, N, and P concentrations, and the soil N content showed a monthly trend in T3. The C, N, and P concentrations of moss crusts increased with the addition levels of biochar, and the moss crust P concentrations showed an overall increasing trend by the month. Moreover, the soil and moss crust C: P and N: P ratios both increased. There was a significant correlation between moss crust C, N, and P and soil C and N. Additionally, nitrate nitrogen (NO3−N), N: P, C: P, EC, pH, soil moisture content (SMC), and N have significant effects on the C, N, and P of moss crusts in turn. This study revealed the contribution of biochar to the nutrient cycle of desert system plants and their underlying soil from the perspective of stoichiometric characteristics, which is a supplement to the theory of plant soil nutrition in desert ecosystems.

Keywords: desert ecosystems; ecological stoichiometry; local plant biochar; moss crusts; nutrient cycling.

Figure 1

Schematic diagram of moss crust and soil sampling.

Figure 2

Changes of C, N, and P contents in moss crusts underlying soil (A–C) and moss crusts (D–F) with different levels of biochar addition. The C, N, and P contents are total forms, CK:0 t ha⁻¹; T1: 3.185 t ha⁻¹; T2: 6.37 t ha⁻¹; T3: 12.74 t ha⁻¹). Different capital letters indicate a significant difference among five sampling months with the same treatment; different lowercase letters indicate a significant difference among four biochar treatments in the same sampling period. Vertical bars show the standard error (SE) (n = 3).

Figure 3

Linear regression indicating relationships between OC (C), TN (N), and TP (P) in the soil and C, N, and P in the moss crusts under different biochar treatments. The solid blank line indicates a significant correlation (p < 0.05). Notes: BC = moss crust C, BN = moss crust N, BP = moss crust P, SC = soil C, SN = soil N, SP = soil P.

Figure 4

Redundancy analysis (RDA) of the moss crusts stoichiometric characteristics and its underlying soil physicochemical characteristics. In the biplot, the red lines in the figure represent the soil physicochemical characteristics, and the blue lines represent the moss crust C: N: P stoichiometry, respectively. Notes: * p < 0.05, ** p < 0.01, and *** p < 0.001, SMC = soil moisture content, EC = electrical conductivity.

Figure 5

Structural equation model (SEMs) illustrating the effects of addition biochar on C, N, P of moss crusts and its underlying soil. Adjacent numbers that are labeled in the same direction as the arrow represents path coefficients, and the width of the arrow is in proportion to the degree of path coefficients. Blank and red arrows indicate negative and positive relationships, two ways arrows indicate mutual influence, continuous and dashed arrows represent the important and not important influence relationships, respectively. Significance levels are denoted with * p < 0.05, ** p < 0.01, and *** p < 0.001, The low chi-square (CMINDF), non-significant probability level (p > 0.05), high goodness-of-fit index (GFI > 0.90), and low root-mean-square errors of approximation (RMSEA) listed below the SEMs indicate that our data matches the hypothetical models. SMC = soil moisture content, R² values indicate the proportion of variance explained by each variable.