Table 2.

Correlation coefficients and stepwise regression model between county-level rate of SOC change (SOC_rate) and climate (mean annual temperature, MAT; and mean annual precipitation, MAP), initial soil properties (sand, silt, and clay content; soil pH in 1980, pH₁₉₈₀; and SOC stock in 1980, SOC₁₉₈₀), and management practices (cumulative N fertilizer input, Nf_Ninput; and average crop residue carbon input, Residue_Cinput) (n = 58)

Factors	Climate		Initial soil properties					Management practices
	MAT	MAP	Sand	Silt	Clay	SOC1980	pH1980	NfNinput	ResidueCinput
Pearson R	0.32*	0.08	0.03	0.03	−0.11	−0.58***	0.01	0.28*	0.45***
Partial R	−0.1	−0.05	−0.07	0.07	0.29*	−0.57***	−0.2	−0.06	0.39***

Stepwise regression model. SOC_rate(Mg C ha⁻¹ y⁻¹) = 0.13 − 0.009 × SOC₁₉₈₀ (Mg C ha⁻¹) + 0.005 × Clay(%) + 0.16 × Residue_Cinput(Mg C ha⁻¹ y⁻¹). Adjusted R = 0.49, P < 0.001. *P < 0.05; ***P < 0.001.