1. Introduction
Soil organic carbon (SOC) can reflect soil health, and plays an extremely important role in increasing soil carbon storage, improving soil fertility, and promoting plant growth (Sollins et al., 2007; Li et al., 2018). However, the SOC group consists of sub-groups with variable turnover rates, each with a different sensitivity to environmental changes (Guo, Wang, Wang, Wu, & Cao, 2018). Soil active organic carbon typically includes microbial biomass carbon (MBC), easily oxidized carbon (EOC) and particulate organic carbon (POC). While the proportion of soil active organic carbon to soil total organic carbon is low, this ration can reflect the changes in soil carbon groups due to soil management measures and environmental changes (Jha et al., 2012; Sahoo, Singh, Gogoi, Kenye, & Sahoo, 2019). The soil active organic carbon is directly involved in the biological and chemical conversion process of soil (Sun et al., 2014), plays a vital role in the cycling of soil nutrient, and stores soil nutrients (Simard, Fyles, D, & Nguyen, 2001). Furthermore, soil active organic carbon is easily affected by plants and microorganisms in a significant way (Chen, Zhou, & Xiao, 2010; Kimura, Murase, & Lu, 2004). However, variation in soil active organic carbon contents across different vegetation types is poorly understood.
Soil enzyme activities are involved in the biochemical processes of the soil system and are linked to “plant-soil enzymes-soil nutrients” (Araújo et al., 2013; da Silva et al., 2012; Lino et al., 2015; Nannipieri et al., 2012). In particularly, enzyme activities (i.e., amylase, catalase, urease and sucrase) related to the soil carbon cycle and serve as important indicators of soil fertility. Amylase and sucrase are involved in the conversion of soil carbohydrates and can hydrolyze organic matter into glucose and sucrose for plant growth and microbial activity (Xie et al., 2017; Ge et al., 2011). Urease acts on carbon-nitrogen bonds in organic matter, and produces carbon dioxide and water by hydrolyzing ammonia or amino salts, while catalase is related to the redox ability of the soil (Baddam et al., 2016; Nowak et al., 2004). These enzyme activities have an important influence on the carbon cycles in soil ecosystems (Bergstrom et al., 1999; Burns et al., 2013). Previous studies have shown that plants can not only directly influence soil enzyme activities by secreting exogenous enzymes, but also affect the composition and diversity microbial species by releasing exudate and oxygen into the rhizosphere, which indirectly affect enzyme activity (Singh & Kumar, 2008). Moreover, plants also indirectly mediate enzyme activities in the soil by controlling the volume of aboveground litter (Caravaca, Alguacil, Torres, & Roldán, 2005). Therefore, these enzyme activities were often chosen to understand the variations in SOC and soil quality (Acosta-Martínez et al., 2007; Chen et al., 2016).
The Loess Plateau is located in the north-central China and has one of the highest concentration of loess on earth, with a total area of 64,000 square kilometers. It also has a high rate of soil erosion and is one of the most ecologically fragile environments in the world, meaning vegetation is important to enhance fertility levels and the soil’s ability to hold water (García, Hernández, & Costa, 1994). Over the past few decades, extensive efforts at restoring the environment have improved the fragile natural ecosystems on the Loess Plateau (Intergovernmental Panel on Climate Change, 2014). Vegetation restoration not only benefit for water preservation and reduction of soil erosion (Ran, Lu, & Xu, 2013), but significantly improve the properties and quality of soil (Zhang et al., 2019). Studies have shown that returning farmland to forests not only improves the SOC reserves and quality, but also improves the conversion trend of soil SOC-related fractions (Deng et al., 2019; Xun et al., 2010; Liu et al., 2014). However, due to a difference in vegetation types, environmental factors, and regional variation, there is a lack of information on the relationship between enzyme activities and soil carbon fractions across different vegetation types. Therefore, we selected vegetation from forests (Xanthoceras sorbifolia ), shrublands (Hippophae rhamnoides and Caragana korshinskii ), and grasslands on the Loess Plateau to study the distribution characteristics of soil active organic carbon components and soil enzyme activities under different vegetation types. This provided a reference value for the sustainable ecological restoration of the Loess Plateau and subsequent improvement of the soil. We hypothesized that both SOC components and enzyme activities in the forest were higher than that of shrublands and grasslands, and that the SOC components and enzyme activities in the surface layer of all vegetation types were higher than in the lower layer. This study provides additional insight into nutrient cycling processes in the ecological restoration of vegetation, and improves the sustainability of ecosystem restoration by identifying effective vegetation to plant.