What are the best statistical methods for mapping soil contamination in an urban area?

2 Geostatistical simulation

Another statistical method for mapping soil contamination is geostatistical simulation, which is the process of generating multiple realizations of soil parameters that are consistent with the observed data and the spatial correlation structure. Geostatistical simulation can be done using various techniques, such as sequential Gaussian simulation, sequential indicator simulation, and conditional simulation. The advantage of geostatistical simulation is that it can quantify the uncertainty and variability of soil data, and provide a range of possible scenarios of soil contamination. Geostatistical simulation can produce probability maps of soil contamination that can indicate the areas with high or low risk of exceeding a certain threshold.

3 Spatial regression

A third statistical method for mapping soil contamination is spatial regression, which is the process of modeling the relationship between soil parameters and other explanatory variables, such as land use, soil type, elevation, and distance to sources. Spatial regression can be done using various techniques, such as ordinary least squares, spatial lag, spatial error, and geographically weighted regression. The benefit of spatial regression is that it can identify the factors that influence soil contamination, and estimate the magnitude and direction of their effects. Spatial regression can produce coefficient maps of soil contamination that can show the spatial variation and significance of the explanatory variables.

4 Spatial analysis software

To apply these statistical methods for mapping soil contamination, you will need to use some spatial analysis software that can handle large and complex soil data sets. Some of the most popular spatial analysis software are ArcGIS, QGIS, R, and Python. These software have various functions and packages that can perform spatial interpolation, geostatistical simulation, spatial regression, and other spatial analysis tasks. They also have graphical user interfaces and visualization tools that can help you create and display high-quality maps of soil contamination.

5 Data quality and sampling design

Prior to applying any statistical method for mapping soil contamination, it is essential to guarantee that your soil data are of high quality and accurately represent the spatial variability and heterogeneity of the soil system. To achieve this, you should employ suitable methods and equipment for collecting, storing, and analyzing soil samples. Additionally, it is important to remove or correct any errors, outliers, or missing values in the soil data. Furthermore, you must select a suitable spatial scale and resolution for the soil data and the maps. Moreover, you should use a stratified or adaptive sampling design that can capture the spatial patterns and hotspots of soil contamination. Finally, you must use a sufficient number of soil samples to provide precise and dependable estimates of soil parameters.

6 Remediation planning and evaluation

The ultimate goal of mapping soil contamination is to support the remediation planning and evaluation of the contaminated sites. Remediation is the process of restoring the soil quality and reducing the risk of exposure to pollutants, which can be done with various methods such as excavation, capping, bioremediation, phytoremediation, and chemical treatment. The choice of method depends on the type, level, and extent of soil contamination, as well as the environmental, social, and economic factors. The maps of soil contamination can help identify areas that need remediation and prioritize them based on risk level and objectives. Additionally, they can be used to select the most suitable remediation method and design a scheme based on the spatial distribution and characteristics of soil contamination. Finally, maps can be used to monitor and evaluate the remediation performance and effectiveness based on changes in soil contamination over time and space.

7 Here’s what else to consider

This is a space to share examples, stories, or insights that don’t fit into any of the previous sections. What else would you like to add?