Jingyi Huang

An error budget for soil salinity mapping using different ancillary data

Secondary soil salinisation occurs as a function of human interaction with the landscape. Increasing salinity is a major constraint to crop yield. The electrical conductivity of a saturated soil-paste extract (ECe, dS m–1) defines the level of salinity in soil. In order to manage salinity, farmers need to map its variation. However, ECe determination is time-consuming and expensive. Digital mapping of ECe is possible by using ancillary data such as easy-to-obtain digital elevation model, gamma-ray spectrometry and electromagnetic (EM) induction data. In this paper, we used these ancillary data and empirical best linear unbiased prediction (E-BLUP) to make a digital map of ECe. In this regard, we found that elevation, radioelement of thorium (Th) and logEM38-v were the most statistically useful ancillary data. We also developed an error-budget procedure to quantify the relative contributions that model, input (for all the ancillary datasets), and combined and individual covariate (for each of the ancillary datasets) error made to the prediction error of our map of ECe. The error-budget procedure used ordinary kriging, E-BLUP and conditional simulation to produce numerous realisations of the data and their underlying errors. Results show that the combined error of model error and input error was ~4.44 dS m–1. Compared with the standard deviation of observed soil ECe (3.61 dS m–1), the error was large. Of this error, most was attributable to the input error (1.38 dS m–1), which is larger than the model error (0.02 dS m–1). In terms of the input error, we determined that the larger standard deviation is attributable to the lack of ancillary data, namely the ECa in areas adjacent to the Darling River and on the aeolian dune where data collection was difficult owing to dense native vegetation.

Modeling Coastal Salinity in Quasi 2D and 3D Using a DUALEM-421 and Inversion Software

Rising sea levels, owing to climate change, are a threat to fresh water coastal aquifers. This is because saline intrusions are caused by increases and intensification of medium-large scale influences including sea level rise, wave climate, tidal cycles, and shifts in beach morphology. Methods are therefore required to understand the dynamics of these interactions. While traditional borehole and galvanic contact resistivity (GCR) techniques have been successful they are time-consuming. Alternatively, frequency-domain electromagnetic (FEM) induction is potentially useful as physical contact with the ground is not required. A DUALEM-421 and EM4Soil inversion software package are used to develop a quasi two- (2D) and quasi three-dimensional (3D) electromagnetic conductivity images (EMCI) across Long Reef Beach located north of Sydney Harbour, New South Wales, Australia. The quasi 2D models discern: the dry sand (<10 mS/m) associated with the incipient dune; sand with fresh water (10 to 20 mS/m); mixing of fresh and saline water (20 to 500 mS/m), and; saline sand of varying moisture (more than 500 mS/m). The quasi 3D EMCIs generated for low and high tides suggest that daily tidal cycles do not have a significant effect on local groundwater salinity. Instead, the saline intrusion is most likely influenced by medium-large scale drivers including local wave climate and morphology along this wave-dominated beach. Further research is required to elucidate the influence of spring-neap tidal cycles, contrasting beach morphological states and sea level rise.

Digital soil mapping of a coastal acid sulfate soil landscape

Coastal floodplains are commonly underlain by sulfidic sediments and coastal acid sulfate soils (CASS). Oxidation of sulfidic sediments leads to increases in acidity and mobilisation of trace metals, resulting in an increase in the concentrations of conducting ions in sediment and pore water. The distribution of these sediments on floodplains is highly heterogeneous. Accurately identifying the distribution ofCASS isessential for developing targeted management strategies. One approach is the use of digital soil mapping (DSM) using ancillary information. Proximal sensing instruments such as an EM38 can provide data on the spatial distribution of soil salinity, which is associated with CASS, and can be complemented by digital elevation models (DEM). We used EM38 measurements of the apparent soil electrical conductivity (ECa) in the horizontal and vertical modes in combination with a high resolution DEM to delineate the spatial distribution of CASS. We used a fuzzy k-means algorithm to cluster the data. The fuzziness exponent, number of classes (k) and distance metric (i.e. Euclidean, Mahalanobis and diagonal) were varied to determine a set of parameters to identify CASS. The mean-squared prediction error variance of the class mean of various soil properties (e.g. EC1:5 and pH) was used to identify which of these metrics was suitable for further analysis (i.e. Mahalanobis) and also determine the optimal number of classes (i.e. k=4). The final map is consistent with previously defined soil-landscape units generated using traditional soil profile description, classification and mapping. The DSM approach is amenable for evaluation on a larger scale and in order to refine CASS boundaries previously mapped using the traditional approach or to identify CASS areas that remain unmapped.

Mapping Spatial Variability of Soil Salinity in a Coastal Paddy Field Based on Electromagnetic Sensors

In coastal China, there is an urgent need to increase land area for agricultural production and urban development, where there is a rapid growing population. One solution is land reclamation from coastal tidelands, but soil salinization is problematic. As such, it is very important to characterize and map the within-field variability of soil salinity in space and time. Conventional methods are often time-consuming, expensive, labor-intensive, and unpractical. Fortunately, proximal sensing has become an important technology in characterizing within-field spatial variability. In this study, we employed the EM38 to study spatial variability of soil salinity in a coastal paddy field. Significant correlation relationship between ECa and EC1:5 (i.e. r >0.9) allowed us to use EM38 data to characterize the spatial variability of soil salinity. Geostatistical methods were used to determine the horizontal spatio-temporal variability of soil salinity over three consecutive years. The study found that the distribution of salinity was heterogeneous and the leaching of salts was more significant in the edges of the study field. By inverting the EM38 data using a Quasi-3D inversion algorithm, the vertical spatio-temporal variability of soil salinity was determined and the leaching of salts over time was easily identified. The methodology of this study can be used as guidance for researchers interested in understanding soil salinity development as well as land managers aiming for effective soil salinity monitoring and management practices. In order to better characterize the variations in soil salinity to a deeper soil profile, the deeper mode of EM38 (i.e., EM38v) as well as other EMI instruments (e.g. DUALEM-421) can be incorporated to conduct Quasi-3D inversions for deeper soil profiles.

Irrigation salinity hazard assessment and risk mapping in the lower Macintyre Valley, Australia

In the Murray-Darling Basin of Australia, secondary soil salinization occurs due to excessive deep drainage and the presence of shallow saline water tables. In order to understand the cause and best management, soil and vadose zone information is necessary. This type of information has been generated in the Toobeah district but owing to the state border an inconsistent methodology was used. This has led to much confusion from stakeholders who are unable to understand the ambiguity of the results in terms of final overall risk of salinization. In this research, a digital soil mapping method that employs various ancillary data is presented. Firstly, an electromagnetic induction survey using a Geonics EM34 and EM38 was used to characterise soil and vadose zone stratigraphy. From the apparent electrical conductivity (ECa) collected, soil sampling locations were selected and with laboratory analysis carried out to determine average (2-12m) clay and EC of a saturated soil-paste extract (ECe). EM34 ECa, land surface parameters derived from a digital elevation model and measured soil data were used to establish multiple linear regression models, which allowed for mapping of various hazard factors, including clay and ECe. EM38 ECa data were calibrated to deep drainage obtained from Salt and Leaching Fraction (SaLF) modelling of soil data. Expert knowledge and indicator kriging were used to determine critical values where the salinity hazard factors were likely to contribute to a shallow saline water table (i.e., clay ≤35%; ECe>2.5dS/m, and deep drainage >100mm/year). This information was combined to produce an overall salinity risk map for the Toobeah district using indicator kriging. The risk map shows potential salinization areas and where detailed information is required and where targeted research can be conducted to monitor soil conditions and water table heights and determine best management strategies.

The location- and scale- specific correlation between temperature and soil carbon sequestration across the globe

Much research has been conducted to understand the spatial distribution of soil carbon stock and its temporal dynamics. However, an agreement has not been reached on whether increasing global temperature has a positive or negative feedback on soil carbon stocks. By analysing global maps of soil organic carbon (SOC) using a spherical wavelet analysis, it was found that the correlation between SOC and soil temperature at the regional scale was negative between 52° N and 40° S parallels and positive beyond this region. This was consistent with a few previous studies and it was assumed that the effect was most likely due to the temperature-dependent SOC formation (photosynthesis) and decomposition (microbial activities and substrate decomposability) processes. The results also suggested that the large SOC stocks distributed in the low-temperature areas might increase under global warming while the small SOC stocks found in the high-temperature areas might decrease accordingly. Although it remains unknown whether the potential increasing soil carbon stocks in the low-temperature areas can offset the loss of carbon stocks in the high-temperature areas, the location- and scale- specific correlations between SOC and temperature should be taken into account for modeling SOC dynamics and SOC sequestration management.

Evaluating a Bayesian modelling approach (INLA-SPDE) for environmental mapping

Understanding the uncertainty in spatial modelling of environmental variables is important because it provides the end-users with the reliability of the maps. Over the past decades, Bayesian statistics has been successfully used. However, the conventional simulation-based Markov Chain Monte Carlo (MCMC) approaches are often computationally intensive. In this study, the performance of a novel Bayesian inference approach called Integrated Nested Laplace Approximation with Stochastic Partial Differential Equation (INLA-SPDE) was evaluated using independent calibration and validation datasets of various skewed and non-skewed soil properties and was compared with a linear mixed model estimated by residual maximum likelihood (REML-LMM). It was found that INLA-SPDE was equivalent to REML-LMM in terms of the model performance and was similarly robust with sparse datasets (i.e. 40-60 samples). In comparison, INLA-SPDE was able to estimate the posterior marginal distributions of the model parameters without extensive simulations. It was concluded that INLA-SPDE had the potential to map the spatial distribution of environmental variables along with their posterior marginal distributions for environmental management. Some drawbacks were identified with INLA-SPDE, including artefacts of model response due to the use of triangle meshes and a longer computational time when dealing with non-Gaussian likelihood families.

Mapping soil water dynamics and a moving wetting front by spatiotemporal inversion of electromagnetic induction data

Characterization of the spatiotemporal distribution of soil volumetric water content (θ) is fundamental to agriculture, ecology, and earth science. Given the labor intensive and inefficient nature of determining θ, apparent electrical conductivity (ECa) measured by electromagnetic induction has been used as a proxy. A number of previous studies have employed inversion algorithms to convert ECa data to depth-specific electrical conductivity (σ) which could then be correlated to soil θ and other soil properties. The purpose of this study was to develop a spatiotemporal inversion algorithm which accounts for the temporal continuity of ECa. The algorithm was applied to a case study where time-lapse ECa was collected on a 350 m transect on seven different days on an alfalfa farm in the USA. Results showed that the approach was able to map the location of moving wetting front along the transect. Results also showed that the spatiotemporal inversion algorithm was more precise (RMSE = 0.0457 cm3/cm3) and less biased (ME = −0.0023 cm3/cm3) as compared with the nonspatiotemporal inversion approach (0.0483 cm3/cm3 and ME = −0.0030 cm3/cm3, respectively). In addition, the spatiotemporal inversion algorithm allows for a reduced set of ECa surveys to be used when non abrupt changes of soil water content occur with time. To apply this spatiotemporal inversion algorithm beyond low induction number condition, full solution of the EM induction phenomena can be studied in the future.

Temperature-dependent hysteresis effects on EM induction instruments

Diurnal drifts of DUALEM were studied at single locations and along a transect.ECa of the perpendicular arrays were more stable than the horizontal coplanar arrays.In-phase of the horizontal coplanar arrays were more stable than the perpendicular.DUALEM measurements show instrument- and temperature-dependent hysteresis effects.Shading the instruments and applying drift correction procedures should be adopted. Non-invasive electromagnetic (EM) induction has been used in agriculture, earth science and archaeology. This is because the apparent electrical conductivity (ECa or quadrature response mSm1) and the apparent magnetic susceptibility (in-phase response ppt) which they measure are related to soil properties such as clay, soil mineralogy, salinity and soil moisture as well as buried metal objects. Although the accuracy issues of the single-coil array EM38 meter have been widely discussed, the accuracy issues of the next generation, multi-coil (perpendicular-PRP and horizontal-HCP) array DUALEM meter, particularly the instrument drift, have little been reported. In this study, the diurnal drifts of a DUALEM-421S and a DUALEM-21S were studied at single locations (for a 24h period) and along a 480-m transect (at five typical operation times). Based on the experiment results of the two DUALEM instruments, it was found that the ECa readings of the PRP arrays were more stable than those of the HCP arrays. The reverse was true for in-phase measurements. Specifically, during the diurnal cooling and heating phases, ECa measurements of the HCP arrays and in-phase PRP arrays showed different correlations with ambient temperature, which can be defined as instrument-specific and temperature-dependent hysteresis effects. In addition, the stability of ECa and in-phase measurements increased with array length and was much less compared to the theoretical values. It was suggested a similar experiment should be conducted for the DUALEM instruments before the DUALEM surveys and repeated DUALEM surveys for mapping the spatio-temporal variations in soil properties should be carried out at the similar temperature (i.e., similar ambient temperature and within the same warming or cooling phase). In addition, shading the instruments with non-conductive thermal insulation should be adopted and drift correction procedures should be applied to improve the quality of the measurements of the EM instruments.

Identifying soil landscape units at the district scale by numerically clustering remote and proximal sensed data

Proximal and remote ancillary data were combined to map soil landscape units.Fuzzy k-means analysis was used to cluster the EM induction and gamma-ray data.Linear mixed models and REML analysis were used to find the optimal class.Topsoil and subsoil properties were predicted with smallest errors when k=4.Distribution of four classes is analogous to the major soil landscape units. Identifying soil landscape units at a district scale is important as it allows for sustainable land-use management. However, given the large number of soil properties that need to be understood and mapped, cost-effective methods are required. In this study, we used a digital soil mapping (DSM) approach where remote and proximal sensed ancillary data collected across a farming district near Bourke, were numerical clustered (fuzzy k-means: FKM) to identify soil landscape units. The remote data was obtained from an air-borne gamma-ray (γ-ray) spectrometer survey (i.e. potassium-K, uranium-U, thorium-Th and total counts-TC). Proximal sensed data was collected using an EM38 in the horizontal (EM38h) and vertical (EM38v) mode of operation. The FKM analysis (using Mahalanobis metric) of the kriged ancillary (i.e. common 100m grid) data revealed a fuzziness exponent (ź) of 1.4 was suitable for further analysis and that k=4 classes was smallest for the fuzziness performance index (FPI) and normalised classification entropy (NCE). Using laboratory measured physical (i.e. clay) and chemical (i.e. CEC, ECe and pH) properties we found k=4 was minimised in terms of mean squared prediction error (i.e. ź2p,C) when considering topsoil (0-0.3m) clay (159.76), CEC (21.943), ECe (13.56) and pH (0.2296) and subsoil (0.9-1.2m) clay (80.81), CEC (31.251) and ECe (16.66). These ź2p,C were smaller than those calculated using the mapped soil landscape units identified using a traditional approach. Nevertheless, class 4A represents the Aeolian soil landscape (i.e. Nb4), while 4D, represents deep grey (CC19) self-mulching clays, and 4B and 4C yellow-grey (II1) self-mulching clays adjacent to the river and clay alluvial plain, respectively. The differences in clay and CEC reveal why 4B, 4C and 4D have been extensively developed for irrigated cotton production and also why the slightly less reactive 4B might be a source of deep drainage; evidenced by smaller topsoil (2.13dS/m) and subsoil (3.76dS/m) ECe. The research has implications for providing meaningful DSM of soil landscape units for farmers at districts scales where traditional methods were restrictive in terms of time and cost.