In Southern China, the hilly red soil region accounts for 2.0 × 10⁶ km² and is an important production base for agriculture, forestry and animal husbandry. However, because of the increasing demand for firewood, timber and food in recent decades, many types of vegetation have been destroyed, which has resulted in local soil degradation and soil fertility decline, giving the region the title "red desert". This soil degradation significantly limits sustainable economic development in this region and so restoring vegetation and improving soil fertility are vital. Natural vegetation restoration, by preventing human disturbance, is an important method for improving fertility by redistributing nutrients in degraded red soil in subtropical hilly regions. However, improving soil fertility by natural vegetation restoration is a long-term process, and the effect of different plants in improving soil fertility is not well understood. In this study, the distribution regularities of nutrients in the soil profile under natural vegetation restoration compared with that in artificial vegetation plots and bare plots were investigated and the effect of natural vegetation restoration on improving soil fertility in subtropical hilly regions was assessed.
A long-term natural vegetation restoration experiment (-2 hectares) was initiated in 1981 at the Red Soil Experiment Station of the Chinese Academy of Agricultural Sciences, Qiyang County, Hunan Province. The effect of natural vegetation on soil fertility was assessed based on six 31-year-old natural vegetation restoration zones, including three natural arbor plots (Cinnamomum camphora, Liquidambar formosana, and Catalpa ovate), two natural shrub plots (Loropetalum chinensis and Sinobambusa tootsik), and one natural herbage plot (Imperata cylindrica). Two artificial arbor plots (Pinus elliottii and Castanea mollissima) and one adjacent bare plot were also used. Soil samples were collected at 0-20, 20-40, 40-60, 60-80, and 80-100 cm depths in April 2012. Soil organic matter (SOM), labile organic matter (LOM), total nitrogen (TN), hydrolysable nitrogen (HN), total phosphorus (TP), available phosphorus (AP), total potassium (TK), available potassium (AK) and pH were determined.
The different vegetation types had significant effects on soil fertility: (1) In Imperata cylindrica and Sinobambusa tootsik plots, the average pH values at different depths were 5.13 and 4.96, respectively, which was significantly higher than that in artificial vegetation and bare plots (average pH 4.56-4.62). Obvious acidification was found in Liquidambar formosana and Loropetalum chinensis plots (average pH 4.22 and 4.38, respectively) compared with that in the bare plot. (2) For all plots, soil nutrient concentrations except TK and TP significantly decreased with soil depth. The highest SOM, LOM, TP, and AP were observed in arbor plots, followed by the herbage plot, and the lowest values were found in shrub plots. Compared with artificial arbor, natural arbor increased SOM by 31.78%-113.19%. HN, TK and AK decreased in the following order: shrub plots > arbor plots > herbage plot. For TN, the decreasing order was arbor plots > shrub plots > herbage plot. (3) There was a significant positive correlation between SOM, LOM, TN and HN (P < 0.01). (4) The integrated soil fertility decreased in the following order: Liquidambar formosana > Catalpa ovate > Loropetalum chinensis > Cinnamomum camphora, Sinobambusa tootsik > Imperata cylindrica > Pinus elliottii > Castanea mollissima > bare. In summary, the results illustrated that natural vegetation restoration is better for improving soil fertility than artificial vegetation.