Soil aggregates are among crucial factors for determining both the quality and erosion resistance of soils. Biochar is a soil amendment that has seen increasing use to improve specific soil properties, mainly the physical structure and the preserving capacity of water and nutrients, as well as sequestration of soil organic carbon. In this study, we applied the rice husk biochar (RHB) and cattle manure compost (COM) in a sandy loam rural soil, which is widely distributed in southern Taiwan, to investigate the combined effects of the biochar and compost on soil aggregation and dynamic change of organic carbon fractions. Through our incubation experiment, both biochar and compost could promote the soil aggregation after eight weeks incubation. The total amounts of macroaggregates (MaAs, >2.0 mm) and mesoaggregates (MeAs, 0.25–2.0 mm) increased by 1.3–9%. During aggregation processes, a considerably greater amount of the soil organic carbon was found to enrich mainly in MaAs and MeAs in all treatments. The COM addition in the soil further promotes organic carbon enrichment in microaggregates (MiAs, <0.25 mm) + fine particles and MeAs after incubation. Increasing labile organic C (LOC) fractions were significantly found in MaAs and MeAs during aggregation processes, whereas decreasing LOC fractions were found in MiAs. The input of fresh organic matter (RHB and COM) initial acts as binding agents in MiAs, and then further enhances the formation of MeAs and MaAs gradually. In conclusion, RHB promotes the physical protection of organic C by increasing soil aggregation and is hence a management option to enhance the C sequestration potential.

The erosivity factor in the universal soil loss equation (USLE) provides an effective means of evaluating the erosivity power of rainfall. The present study proposes three regression models for estimating the erosivity factor based on daily, monthly, and annual precipitation data of rainfall station network, respectively. The validity of the proposed models is investigated using a dataset consisting of 16,560 storm events monitored by 55 rainfall stations in southern Taiwan. The results show that, for 49 of the 55 stations, a strong positive correlation $(r^2>0.5)$ exists between the annual rainfall amount and the annual rainfall erosivity factor. In other words, the estimation model based on the annual precipitation data provides a reliable means of predicting the long-term annual rainfall erosivity in southern Taiwan. Furthermore, the root mean square error (RMSE) and mean absolute percentage error (MAPE) analysis results show that the estimation models based on annual and monthly precipitation data have a more accurate prediction performance than that based on daily precipitation data.

Soil degradation is a crucial problem, particularly in tropical and subtropical areas. Prevention or reduction of soil erosion requires strategies based on thorough rapid vegetation cover (VC) and favorable soil quality in subtropical and tropical areas. This study applied wood biochar (WB) and rice husk biochar (RHB) in a mudstone soil, which is widely distributed in Southern Taiwan, to investigate the effects of biochar application on soil erosion and vegetation restoration. The standard erosion unit plots (22.13 m in length and 9% in slope gradient) were set up to determine the relationship among soil losses, VC, and natural rainfall characteristics with and without biochar application. The results indicated that biochar application increased the growth rate (identified by cover ratio) of Bahia grass (Paspalum notatum Flüggé) by 2–2.6 times within 40 days compared with control (without biochar application) and increased VC by 20% after 120 days of treatment. The biochar application could effectively reduce soil losses by 60% at least in the mudstone soil. A well-predicted regression function of soil loss with VC and rainfall kinetic energy was established (amount of soil lost = −0.435 × ln VC + 0.54 × RKE, r = 0.89, p < 0.01).

This paper proposes to apply a Markov chain random field conditioning method with a hybrid machine learning method to provide long-range precipitation predictions under increasingly extreme weather conditions. Existing precipitation models are limited in time-span, and long-range simulations cannot predict rainfall distribution for a specific year. This paper proposes a hybrid (ensemble) learning method to perform forecasting on a multi-scaled, conditioned functional time series over a sparse $l_1$ space. Therefore, on the basis of this method, a long-range prediction algorithm is developed for applications, such as agriculture or construction works. Our findings show that the conditioning method and multi-scale decomposition in the parse space $l_1$ are proved useful in resisting statistical variation due to increasingly extreme weather conditions. Because the predictions are year-specific, we verify our prediction accuracy for the year we are interested in, but not for other years.

Invasive white leadtrees ( Leucaena leucocephala (Lam.) de Wit) are contributing to the destruction of coastal ecosystems and biodiversity in humid subtropical and tropical regions. Sudden clearcutting of invasive tress can lead to substantial soil loss, which subsequently reduces the efficiency when indigenous vegetation is restored. Soil samples ([ ]) were collected using the grid method to measure soil properties and to spatially analyze the soil erosion potential in the studied area after clearcutting of the invasive trees. Results indicated that the soil erosion rates ranged from 21 to 249 tons ha⁻¹ yr⁻¹, and the eroded soil depths ranged from 0.30 to 3.40 cm yr⁻¹after clearcutting. The network measurements for determining the mortality ratio of the restored indigenous vegetation demonstrated that Ficus benjamina L., Macaranga tanarius, and Scaevola taccada (Gaertner) Roxb were likely to be susceptible to damage from the substantial amount of soil loss and low SOC content. During the period between clearcutting and establishing the next stand of plants for reforestation, the network measurements for determining the soil properties and estimating the spatial heterogeneity of soil erosion potential are regarded as a major issue of concern regarding the efficiency of subsequent restoration efforts.