Micromorphological Features of Diagnostic Horizons in Several soils in Southwest China: Implication for Soil Taxonomic Classification

The comparative studies on micromorphological features in diagnostic horizons of Stagnic Anthrosols, Ustic Ferrosols and Ustic Vertosols in southwestern China were conducted to underpin the rationale for Chinese Soil Taxonomy. The following findings were explored: (1) Stagnic Anthrosols had the specific micromorphological features, e.g., the humic formation in anthrostagnic epipedon, the platy structures in plow subhorizon, the secondary formation of ferromanganese and the weakly optical-orientation clay domains in hydragric horizon, etc.: (2) The groundmasses of ferric horizon in Ustic Ferrosols appeared in hue of 2.5YR or redder, and had pellicular grain structure; (3) Ustic Vertosols had a crust horizon (Acr), and crack structure dominated in Acr and angular blocky structure in disturbed horizon; (4) Because of the distinct differences in micromorphological features among these three soils, the specific micromorphological features might be employed as diagnostic horizons to differentiate soils while the quantifiable micromorphological features might potentially be selected as diagnostic indices for Chinese soil taxonomic classification.

THEORY AND METHOD FOR WETLAND BOUNDARY DELINEATION

Based on the analysis of the subjectivity of wetland boundary criteria and their causes at present, this paper suggested that, under the condition that the mechanism of wetland formation process has not been understood, "black box" method of System Theory can be used to delineate wetland boundaries scientifically. After analyzing the difference of system construction among aquatic habitats, wetlands and uplands, the lower limit of rooted plants was chosen as the lower boundary criterion of wetlands. Because soil diagnostic horizon is the result of the long-term interaction among all environments, and it is less responsive than vegetation to short-term change, soil diagnostic horizon was chosen as the indicator to delineate wetland upper boundary, which lies at the thinning-out point of soil diagnostic horizon. Case study indicated that it was feasible using the lower limit of rooted plants and the thinning-out point of soil diagnostic horizon as criteria to delineate the lower and upper boundaries of wetland. In the study area, the thinning-out line of albic horizon was coincident with the 55.74m contour line, the maximum horizon error was less than 1m, and the maximum vertical error less than 0.04m. The problem on wetland definition always arises on the boundaries. Having delineated wetland boundaries, wetlands can be defined as follows: wetlands are the transitional zones between uplands and deepwater habitats, they are a kind of azonal complex that are inundated or saturated by surface or ground water, with the lower boundary lying at the lower limit of rooted plants, and the upper boundary at the thinning-out line of upland soil diagnostic horizon.

An Insight into Machine Learning Algorithms to Map the Occurrence of the Soil Mattic Horizon in the Northeastern Qinghai-Tibetan Plateau

Soil diagnostic horizons, which each have a set of quantified properties, play a key role in soil classification. However, they are difficult to predict, and few attempts have been made to map their spatial occurrence. We evaluated and compared four machine learning algorithms, namely, the classification and regression tree(CART), random forest(RF), boosted regression trees(BRT), and support vector machine(SVM), to map the occurrence of the soil mattic horizon in the northeastern Qinghai-Tibetan Plateau using readily available ancillary data. The mechanisms of resampling and ensemble techniques significantly improved prediction accuracies(measured based on area under the receiver operator characteristic curve score(AUC)) and produced more stable results for the BRT(AUC of 0.921 ± 0.012, mean ± standard deviation) and RF(0.908 ± 0.013) algorithms compared to the CART algorithm(0.784 ± 0.012), which is the most commonly used machine learning method. Although the SVM algorithm yielded a comparable AUC value(0.906 ± 0.006) to the RF and BRT algorithms, it is sensitive to parameter settings, which are extremely time-consuming.Therefore, we consider it inadequate for occurrence-distribution modeling. Considering the obvious advantages of high prediction accuracy, robustness to parameter settings, the ability to estimate uncertainty in prediction, and easy interpretation of predictor variables, BRT seems to be the most desirable method. These results provide an insight into the use of machine learning algorithms to map the mattic horizon and potentially other soil diagnostic horizons.

Chinese Soil Taxonomy:A Milestone of Soil Classification in China

1 Background Taxonomy is the branch of science dedicated to discovering,characterizing,naming,and classifying objects or organisms so as to understand relationships between them and the factors of their formation.The aims of classification are to identify and understand the objects for establishing an orderly system for the grouping objects.