Search

The Chaoyang profile represents a rare multi-period, continuous and complete sequence of aeolian paleo-deposits with a stable sedimentary origin and multi-stage paleoclimatic cycles. Benchmark profiles including soil types at different pedogenic stages can be used for the recognition and classification of paleosols and paleoclimate reconstruction. The loess–paleosol sequence benchmark profile (LBP) is also helpful in comparing the results of paleoenvironment reconstruction from different ecological regions. In this study, a loess–paleosol profile derived from thick loess in Chaoyang city of Liaoning province, Northeast China, was investigated as a well-preserved LBP that included various paleosol types. To determine the nature and origin of the Chaoyang profile, the geographic, stratigraphic and morphological characteristics were described in the field. Bulk samples from 42 horizons were collected for chemical and physical analysis, and sub-sampling of 946 samples at 2 cm intervals from the surface to the bottom were taken to measure grain size distributions and magnetic susceptibility. Results showed that the 19.85 m thick loess–paleosol profile had been continuously deposited since 423 ka BP. The upper part (0–195 cm), or UPP, was predominantly of aeolian loess deposition origin but was mixed with water-reworked materials from a nearby secondary loess source. The middle part (195–228 cm), or MIP, was also indirectly affected by the water-reworking process through the leaching of materials from the overlying UPP. The lower part (228–1985 cm), or LOP, was characterized by four reddish stratigraphic layers interbedded with five yellowish ones, indicating several types of paleosols developed under different ecological environments. The multi-stage paleoclimatic cycles as evidenced by morphological and physical characteristics as well as age dating and magnetic susceptibility correlated well with the Lingtai section and LR04 benthic δ18O. Because of these attributes, the Chaoyang profile can be deemed as a benchmark loess–paleosol profile for the recognition and classification of paleosols and paleoclimate reconstruction in Northeast China. The differences in morphological and physical properties between paleosols and loess suggest different soil fertility and agronomic properties and need further studies to assess their functionality with climate fluctuation. [ABSTRACT FROM AUTHOR]

[Display omitted] • A deep-learning model was developed for identifying and delineating soil horizons. • Data augmentation was used to increase the diversity of the profile image dataset. • Results of the proposed model compared well with manually delineated horizons. • An Android smartphone application was developed as a digital soil description tool. Soil horizons are the manifestation of pedogenesis and contain the basic morphologic indicators of soil formation. Accurate and quantitative tools for delineating soil horizons in-situ can assist soil scientists towards rapid and dynamic soil survey information. The objective of this study was to develop a deep-learning-based, soil-profile-imaging method for identifying and delineating soil master horizons to assist in digital soil descriptions. A total of 160 soil profile images from four soil orders (Alfisols, Entisols, Inceptisols, and Mollisols) were collected from north China in the Inner Mongolia and Liaoning regions. The 160 profile images were amplified to 2400 individual images for model building using data-augmentation procedures. The augmented profile image dataset was divided into training (70%), validation (15%), and test (15%) datasets. The training and validation datasets were imported into a nested, U-net network for model building. The proposed deep-learning (DL) model classified soil profiles into A, B, and C horizons. The mean pixel accuracy of the DL model was 0.86 with the training dataset, 0.82 with the validation dataset, and 0.83 for the test dataset. Results showed that the DL model was judged to be accurate enough for identifying and delineating soil master horizons from profile images in practice. Based on the proposed DL model, a smartphone application was subsequently developed to digitalize the soil profile and assist field evaluation of soil profile horizonation. The smartphone application, for the Android operating system (version 5.0 and greater), featured a response time of < 5 s. This study demonstrated that the proposed DL model and smartphone application could be a simple, fast, and digital tool for quickly identifying and delineating in-situ soil horizons. These tools allow for rapid data collection, which can be used for future artificial intelligence development and application in soil science towards digital soil profile description and dynamic soil survey. [ABSTRACT FROM AUTHOR]