OR/14/025 Introduction

LAWLEY, R. 2014. User Guide: Soil Parent Material 1 km dataset. British Geological Survey Internal Report, OR/14/025.

This document provides information for users of the Soil Parent Material 1 km dataset. The Soil Parent Material 1 km (SPM1k) database is part of a series of GIS maps designed to help environmental scientists and consultants assess the characteristics of the ‘near-surface’ weathered zone. In particular, the data focuses upon the material from which top soils and subsoils (A and B horizons) develop (i.e. from the base of pedological soil down to c. 2–3m).

This Soil Parent Material1km resolution dataset is provided as a free-to-use dataset under the Open-Government License. The 1km dataset is derived from the standard 1:50 000 scale Soil Parent MaterialV6 dataset by spatial generalising (onto a regular vector grid of 1 km cell size) the dominant Parent Material type onto a 1 km resolution vector grid. Each grid cell being attributed seven fields of simple information derived from the dominant parent type.

The attribution of the dataset mirrors that available within the UK Soil Observatory.www.UKSO.org. The 1 km resolution makes the data an ideal vector layer to use with other freely available layers from the UKSO.

Background

In 2005 the British Geological Survey initiated a development programme to produce thematic datasets that identified and portrayed a variety of parent material and soil properties in Great Britain.

A ‘parent material’ is a geological deposit over, and within which, a soil develops. Typically, the parent material is the first recognisably geological deposit encountered when excavating beneath the soil layer. It represents the very-near-surface geology. In general, the geological deposits closest to the ground surface are the most weathered, whilst the deeper deposits are less so. The interface between soil and parent (geology) can vary from a sharp, clearly defined boundary, to a diffuse continuum with no distinctive point of transition.

“The major soil groups [in the UK]… are distinguished by broad differences in the composition or origin of the soil material” (Avery, 1980). A soil’s parent materials play a vital role in determining soil-type. Typically, parent material characteristics control three primary characteristics of their overlying soils:

• Texture
• Gross soil Chemistry (geo-genic component)
• Permeability-Porosity (drainage)

A range of material characteristics (physical and chemical) are implicitly defined in the typical geological rock/deposit descriptions provided in many BGS products and most geologists are able to interpret the information from their own experience and knowledge. However, these characteristics are easier to use (by non-geologists) when explicitly defined, as they are in the Soil Parent Materialdatabase.

The Soil Parent Material 1 km data has explicitly defined attributes for the following characteristics:

• European Soil Bureau PM type
• Soil Texture
• Soil Group (generalised soil texture)
• PM texture
• PM Carbonate Content
• Soil Layer thickness

Who would benefit from the dataset?

It is envisaged that ‘Parent Materials 1 km’ is of interest to a wide range of users working in the field of environmental science (from ground engineering and landscape modelling to climate change impact assessment). The dataset has been designed for a diverse user-base and simply presents BGS’ geological and pedological spatial data in a manner that is more flexible for GIS-aware environmental scientists. A basic understanding of geology and soils is recommended so that users can fully utilise the dataset but any user with a grasp of physical geography should be able to work with the dataset. The database sets out simplified, qualitative descriptions of 6 basic PM and soil characteristics available from the UKSO and mySoil smartphone app. The 1 km dataset is designed to work with similar 1km resolution data available from BGS and other organisations working in soil research. More expert users may want to use the high resolution 1:50 000 scale dataset to integrate with climate, land use and terrain data to create high-resolution digital soil models.

What the dataset shows

The Soil Parent Material 1km dataset is a vector grid (sometimes called a fishnet-grid) with 1 km x 1 km cell dimensions. It is derived from the standard 1:50 000 scale Soil Parent Material v6 dataset by spatial generalised of the dominant Parent Material type (less spatially prevalent parent materials being absorbed into the dominant class). Dataset attribution is inherited from the dominant parent and comprises the following information.

European Soil Bureau definition of Parent Material type (ESB_DESC)

The European Soil Bureau (ESB) definition of Parent Material type provides the ‘most applicable’, European definition of parent material type for each map unit. The ESB ‘code’ and description for a parent material is defined in the Georeferenced Soil database for Europe (Finke et al, 2001) and is summarised in Appendix 1. Multi-lithic parent materials (e.g. layered units of two or more contrasting lithologies) are represented by two or more codes separated by ‘_’. The ESB description is useful for users considering integration of this database into European soil/geology databases. Alternative codes and definitions are available within the high resolution, licensed dataset, that provide opportunities for reclassifying the parent type via a number of hierarchical classification systems.

Carbonate Content (CARB_CNTNT)

This field classifies all forms of carbonate content in each parent material (calcite, dolomite, siderite) as a simple ranking of: none, low, moderate or high (with unknown or variable for heterolithic and multilithic parent units). As a very general rule soils that have formed over carbonate-rich parent material typically have large carbonate contents which means they strongly buffer acidity from agricultural or atmospheric sources. The Parent Material Carbonate Content classification is based on expert judgment of the likely content of carbonate (Mg/CaCO₃) in the total rock-mass of each parent material. For example, High carbonate content would be expected for a chalk or limestone parent material type; Low carbonate content would be representative of weakly, calcite-cemented sandstone. The term variable is used where the parent material comprises two or more distinct lithologies e.g. interbedded limestone and mudstone where it is not possible to clearly identify the precise parent type at any given location. Users should note that land management practices and natural weathering processes of parent materials play a significant role in controlling carbonate content (it has not been possible to factor in these influences within the scope of this dataset) and so the classes shown in this dataset should be regarded as wholly ‘indicative’.

Parent Material Grain Size (PMM_GRAIN)

This is a generalised textural description of the Parent Material. The grain-size values are derived from analyses of samples taken from the parent material and also expert judgement (based on field survey and description of the deposits). Where a parent material comprises more than one type of deposit, a range of classes is recorded. Deposits that are known to be heterolithic are identified by the term ‘mixed’. Note that geologists use different grain-size descriptions for rocks originating from sedimentary and igneous lithologies and that peat is not attributed with a grain- size term (peat is used to denote areas of peatland where we hold no alternative grain-size information).

Soil Group (SOIL_GROUP)

This is a generalised description of soil texture found over different parent material types, in terms of Heavy, Medium or Light soils as broadly indicated in the Defra Cross Compliance Guidance for Soil Management (Defra, 2006).

These generalised soil texture groupings are derived by comparing samples of measured soil textures (%clay, %silt and %sand) against a simplified grain-size ternary plot. Note that this map uses terms that refer to light soils (i.e. sand/silt rich), medium soils (i.e. loams) and heavy soils (i.e. clay-rich). The classifications are based on analyses of archived BGS G-BASE and geotechnical samples as well as expert judgment (where sample data is not available). Most parent materials generate a broad range of soil textures so these groupings should be regarded as ‘indicative’. The ‘Mixed or Organic’ classification denotes highly variably textured soils or the presence of peat (no texture data is available for peaty soils).

Soil Texture (SOIL_TEX)

This is a general classification of soil texture derived from measured samples of soils overlying different parent materials. Soil texture classes are based on a UK classification of soil texture. These soil texture classifications are derived by comparing samples of measured soil textures (%clay, %silt and %sand) against the soil grain-size ternary plot designed by The National Soil Resources Institute (Hodgson, 1997). The classifications are based on analyses of archive G-Base and geotechnical samples as well as expert judgment (where sample data is not available). Note that the map uses terms that refer to: Sandy soils, Silty soils, Clayey soils and Loamy soils with additional indicators for the presence of Chalk fragments (Chalky) and Peat (Peaty). Most parent materials generate a broad range of soil textures so these terms should be regarded as ‘indicative’. The ‘Varied, Locally Peaty’ classification denotes variably textured soils that are typically found in alluvial systems.

Soil layer thickness (SOIL_DEPTH)

This is a general classification of soil thickness (soil depth) typically found on different Parent Material types. These soil thickness classifications are derived by using expert judgement to compare records of measured soil depth observed from borehole records with engineering strength characteristics of the underlying Parent Materials. Five categories of soil thickness are provided ranging from deep to shallow. As a rule of thumb, these categories are based on the ability to manually dig (with a spade) without being hindered by a substrate that is too strong to excavate (i.e. layers of solid rock, dense gravel, or very stiff clays). An example of a deep soil would be one developed over unconsolidated, clayey Quaternary deposits, and an example of a shallow soil would be one found in areas underlain by chalk or limestone. Users should note that land management practices and natural weathering and accumulation (or erosion) processes of soil materials play a significant role in controlling field-scale soil depth (it has not been possible to factor in these influences within the scope of this dataset).

How the dataset was created

Data was collated and interpreted from a number of different sources currently held by BGS. The primary datasets used for the ‘Soil Parent Material 1 km’ dataset are:

• Soil Parent Material Map V6 dataset (The main spatial framework)
• DiGMapGB PLUS ‘Excavatability’ dataset
• GBASE (Geochemical and related observational data)
• Single Onshore Borehole Index (SOBI)
• BGS National Geotechnical Properties Database (NGPD)
• Industrial Minerals Assessment Unit (IMAU) Sand and Gravel dataset

These datasets have contributed underlying information to the SPM1km attributes as follows:

	ESB type	Carbonate Content	Parent Material Grain size	Soil Group	Soil texture	Soil Thickness
PMMv6 50k	Y	Y	Y	Y	Y
DiGMapGBPLUS Excavatability						Y
GBASE		Y		Y	Y	Y
SOBI		Y	Y	Y	Y	Y
NGPD			Y	Y	Y	Y
IMAU		Y	Y	Y	Y	Y

The Soil Parent Material 1 km dataset has a simple sequential workflow for each attribute component and an automated spatial generalisation procedure from the 1:50 000 scale SPMM v6 dataset:

1. Identify measurements and observations from archive sample records and rock descriptions that relate to the selected attributes (e.g. carbonate chemistry).
2. Categorise the observations into classes/groups of similar behaviours (e.g. High, Low), where possible using published standards of classification such as BS 5930, Eurocode 7 or similar.
3. Assess whether the range of available evidence statistically supports the range of possible classes (using typical tests such as ANOVA), or whether expert judgement is required to subdivide/combine categories (e.g. it is not viable to predict an accurate CaCO₃ content (as %) across Great Britain, but generic classes of High, Medium, Low can be delineated/estimated).
4. Assess the range of lithostratigraphic variation of geological materials across Great Britain, in terms of the categories and classes identified in steps 2 & 3. Using combined expert-judgement and spatial analyses create a ‘lookup’ dataset of expected attributes (material characteristics and behaviours) per lithostratigraphic unit.
5. Combine the spatial (map) components of the Soil Parent Material v6 datasets with the ‘lookup’ dataset created in step 4 (using standard GIS ‘spatial and/or attribution join’ techniques to create new ‘map’ of fill characteristics.
6. Apply a fish-net generalisation to the dataset, by identifying the dominant parent material type (by area) on a 1 km by 1 km vector grid. Each grid cell is assigned the classification of the dominant feature type.
7. Submit data for standard BGS digital-data checking procedures (to assess completeness and cleanliness of content).

Categorisation of the data and assignment of classes have been mostly made by expert judgement in order to create ‘look up tables’ of lithostratigraphic rock type and associated engineering properties. The methods used are similar to those described in the methodology report for the underpinning Soil Parent Material v6 dataset (Lawley, 2008). Assignments of characteristics have been made at the ‘formation’ level of lithostratigraphic classification of rock/deposit type.

Spatial data-creation and fill-attribution has been completed using standard ESRI ARCGIS (EDITOR Version) software.