4.2 Marsh degradation influences the transformation of soil phosphorus forms
The most remarkable changes in P forms due to marsh degradation was an increase in organic P concentration and proportion, and organic P became the predominant P form in surface soils (Figures. 3a–d and 4a–b) in Zoige, which is consistent with the results of the Min River Estuary, China (Zhang et al., 2015). However, some studies have revealed that the concentration and proportion of organic P in wetland soils decreased because cultivation and drainage stimulated the transformation from organic P to inorganic P and increased organic P loss (Negassa et al., 2020; Schlichting et al., 2002; Wang et al., 2006). Negassa et al. (2020) demonstrated that the proportion of organic P was higher in long-term rewetted peatlands than in drained peatlands. This result was closely associated with marsh desiccation, combined with the changes in the plant community or overgrazing. The alpine marsh experienced light or moderate degradation, and the dense plant communities of hygrophytes and mesophytes, such as Carex , Blysmus sinocompressus, andKobresia tibetica (Table 2), increased plant coverage, aboveground biomass (Table 2), and biodiversity and richness owing to water table drawdown (Zeng et al., 2021) and drying-rewetting (Wang et al., 2021b). Consequently, plant roots might take up more bioavailable inorganic P and then return more organic P (Huang et al., 2015b) via a high litter input compared to RPM, particularly in surface soil (Table 2). Although livestock foraged some plant shoots during grazing, this effect was very minor and hardly altered the above increase in soil organic P because of the limitation of flooding or excess water during the outdoor grazing season (wet season) in LDM and MDM. Alpine marshes are heavily degraded, and more livestock excreta can compensate for the decrease in organic P owing to livestock foraging and an increase in the bioaccumulation of organic P in surface soils during overgrazing. Furthermore, the improvement in aerobic environments in degraded marshes can enhance the microbial activity and decomposition rate of litter and livestock excreta (Gyaneshwar et al., 2002; Jones et al., 2018; Prado and Airoldi, 1999; Qian et al., 2010), which can induce labile organic P migration from the upper to lower layers of natural soil profiles through soil porewater to further increase the organic P concentration in subsurface soils (Pan et al., 2021).
Ca-P was not only the dominant inorganic P form but also the majority of total P for the whole soil profile in this study (Figures 3a–d and 4a–b) because the soil was derived from the calcareous parent material of lacustrine sediment, where the concentration of calcium carbonate equivalent ranged from 61.3 to 240.6 g kg−1 (data from an investigation of soil series survey in Sichuan Province, China). For Ca-P forms, a remarkable change was observed in the concentration and proportion of available Ca2-P increased under marsh degradation, whereas those of unavailable Ca10-P decreased (Figures 3a–d and 4a–b). This is because marsh desiccation accompanied by the transition of the plant community or overgrazing may enhance the activity and/or quantity of phosphate-dissolving bacteria, such as Pseudomonas , Bacillus, and Erwini (Liu et al., 2015; Susilowati et al., 2019) and increase the organic acids from plants, microbes, and/or livestock excreta (Almeida et al., 2020; Gyaneshwar et al., 2002; Miao et al., 2013; Nuryana et al., 2019). Improved phosphate-dissolving bacteria and organic acids can further stimulate the transformation of Ca10-P to other labile inorganic P forms such as Ca2-P, Ca8-P, and Al-P (Almeida et al., 2020; Susilowati et al., 2019). Similarly, Jamil et al. (2016) revealed that Calcisols with greater biological activity under sugarcane crop cover had a high concentration of Ca2-P in the estuary plains of Pakistan. However, the input of external P fertiliser from surrounding pasture uplands resulted in an increasing trend of exchange P (primarily Ca2-P) with increasing hydroperiod in isolated Lake Okeechobee in Florida, USA (Cheesman et al. 2010), which was contrasting with our result, owing to a significant loss of soluble P (e.g. Ca2-P) and lack of external P in RPM with perennial flooding. Another remarkable change in inorganic P forms was that marsh degradation decreased the concentration and proportion of soil Fe-P but increased those of soil occluded P similar to the changes in Al-P or Fe-Al-P during soil drying (Bai et al., 2019; Zhang et al., 2015). This is primarily because soil Fe2+ is oxidised and transformed into colloidal ferric hydroxide and amorphous hydrated iron oxide (Ding et al., 2016; Wang et al., 2021b), which can adsorb and immobilise more phosphate during soil desiccation, resulting in the transformation from Fe-P to occluded P (Sah and Mikkelsen, 1986, 1989).
In addition, marsh degradation did not alter the vertical variation of all forms of P. However, the vertical variation differed among the different forms (Figures 3a–d and 4a–b). The concentration and proportion of Ca10-P evidently increased with increasing soil depth because of the downward trend of soil minerals from the surface to the bottom soil, whereas the concentrations of other inorganic P forms and organic P generally decreased with depth because of ‘surface aggregation’ from plants and microbes (Meyerson et al. 2000; Reddy et al. 1999; Wu et a., 2020). These results are supported by those of previous studies on wetland soils (Luo et al., 2021; Wang et al., 2008; Yuan et al., 2015; Zhang et al., 2020).
In summary, the differences in P forms among marsh soils with different degrees of degradation likely indicates that marsh degradation primarily induced the transformation from recalcitrant Ca10-P to organic P and labile Ca2-P via litter return or livestock excreta combined with marsh desiccation.