No abstract

Land uses and cultivation are important factors controlling SOC storage on the Loess Plateau. These factors may also affect the relative importance of different mechanisms for the stabilization of organic matter in the soil. Easily oxidizable organic carbon (EOC), aggregation and aggregate C fractions in the soil were measured under different land uses. Aggregates were fractionated using a wet-sieving procedure to obtain the distribution of water-stable aggregates. The fractions of aggregates, aggregate SOC and aggregate EOC in grassland and forestland were generally higher than those in farmland. Furthermore, because conventional cultivation destroyed aggregates, the dominant aggregate size fractions were b 0.5 mm for farmland and N0.5 mm for other land uses. Compared to the corresponding values in farmland, the mean weight diameter (MWD) in forestland and grassland increased by 808%-417%, and the stability ratio of water-stable aggregate (WSAR) increased by 920%-553%. Aggregate formation and its dominant size fraction were associated closely with its carbon fractions. SOC and EOC in farmland tended to be concentrated in smaller-sized aggregates, whereas SOC and EOC under other land uses tended to concentrate in larger-sized aggregates. EOC tended to concentrate in larger aggregates than SOC. The small fractions of the aggregates formed large fractions by combining with fresh organic matter. So converting slope farmland to forestland and grassland could improve the storage and quality of SOC, and the tendency of SOC transfer.

The performance of three machine learning methods (support vector regression, random forests and artificial neural network) for estimating the LAI of paddy rice was evaluated in this study. Traditional univariate regression models involving narrowband NDVI with optimized band combinations as well as linear multivariate calibration partial least squares regression models were also evaluated for comparison. A four year field-collected dataset was used to test the robustness of LAI estimation models against temporal variation. The partial least squares regression and three machine learning methods were built on the raw hyperspectral reflectance and the first derivative separately. Two different rules were used to determine the models’ key parameters. The results showed that the combination of the red edge and NIR bands (766 nm and 830 nm) as well as the combination of SWIR bands (1114 nm and 1190 nm) were optimal for producing the narrowband NDVI. The models built on the first derivative spectra yielded more accurate results than the corresponding models built on the raw spectra. Properly selected model parameters resulted in comparable accuracy and robustness with the empirical optimal parameter and significantly reduced the model complexity. The machine learning methods were more accurate and robust than the VI methods and partial least squares regression. When validating the calibrated models against the standalone validation dataset, the VI method yielded a validation RMSE value of 1.17 for NDVI(766,830) and 1.01 for NDVI(1114,1190), while the best models for the partial least squares, support vector machine and artificial neural network methods yielded validation RMSE values of 0.84, 0.82, 0.67 and 0.84, respectively. The RF models built on the first derivative spectra with mtry = 10 showed the highest potential for estimating the LAI of paddy rice.

No abstract

No abstract