4.3 Marsh degradation alters soil phosphorus availability
Our results showed that P availability in alpine wetland soils decreased with soil depth owing to organic matter accumulation in shallow soils and ‘surface aggregation’ of soil nutrients, particularly surface soils (Ockenden et al., 2014; Wu et a., 2020). This result conforms with those of previous studies on freshwater and salt wetland soils (Negassa et al., 2020; Qu et al., 2019; Wang et al., 2008; Xu et al., 2012). Soil P availability hardly had significant differences in the vertical variations among differently degraded marsh soils. However, it generally increased along the marsh degradation gradient from the RPM, reaching a maximum in the LDM or MDM soil and subsequently decreasing to the lowest in HDM (Figure 5a–b). This pattern can be primarily attributed to the changes in factors affecting soil P availability, such as the hydrological regime, vegetation presence, and external disturbance (Duhamel et al., 2017; Ganjegunte et al., 2018; Kröger et al., 2012; Qu et al., 2018; Wang et al., 2021a, b). As RPM evolved into marsh meadow, the drying-rewetting and changes in vegetation community promoted the dissolution of slowly cycling P, decomposition of extracellular polymers and dead microorganisms containing labile monoester and diester P and plant debris containing phytate P, and the release of colloid-bound P through the breakdown of soil aggregates (Stevens and Tullos, 2011; Wang et al., 2021b). Thus, soil P availability was significantly higher in LDM than in RPM (Figure 5a–b), although the released soluble P might have been slightly lost because of inundation (Kröger et al., 2012; Zhou et al., 2019) in wet season. This was similar to the results previously observed in the water-level fluctuation zones of the Three Gorges Reservoir area, China (Wang et al., 2021b), estuarine wetlands with different vegetation in the Yellow River Delta, China (Qu et al., 2021, 2018; Xu et al., 2012), and managed wetland cells of Mississippi, USA (Kröger et al., 2012).
As marsh meadow continuously degraded into the meadow, the comprehensive effects of desiccation and vegetation resulted in a significant increase in soil P availability (Figure 5a–b) through the decrease in labile P loss, according to Kröger et al. (2012), and transformation from slow inorganic P and organic P to bioavailable P owing to the activation of organic acids from microbes and plant roots and debris (Hallama et al., 2019; Qualls and Richardson, 2000; Schelfhout et al., 2021; Wang et al., 2021a). However, when the marsh was heavily degraded, soil microbes significantly decreased by 39.3%–94.1% compared with the other three types of wetlands (Pu et al., 2022), and the activities of phosphomonoesterase, phosphodiesterase, and phytase were lower than those in the other wetlands by 50.5%–55.9%, 44.3%–52.6%, and 34.2%–70.5% (data from unpublished results), respectively. These results indicate that the mobilisation of soil in slowly cycling P was inhibited by the low activities of microbes and phosphatase through overgrazing and rodent damage stress. Therefore, P availability in the HDM soils decreased compared with that in RPM (Figure 5a–b), and even P activation coefficient in surface soil was lower than the threshold of soil P deficiency (Figure 5b). This decrease indicated a risk of P limitation via plant biomass removal in HDM soils. Similarly, Shang et al. (2016), Teng et al. (2020) and Zhai et al. (2022) revealed that P limited plant growth in degraded alpine meadow and saline wetland ecosystems.