Australian Natural Resources Atlas

The table below shows the number of soil profiles from each State and Territory, including those collected and held by CSIRO Land and Water.

A detailed description of the database and how it was compiled is presented below.

The soil profiles (in SITES format) can be downloaded directly from the Australian Natural Resources Data Library with the permission of the data custodians.

 Table of contents

Compiling the database

Soil profile data were collated in digital form from data held by State and Territory agencies and CSIRO in a wide range of formats. To be accepted into ASRIS, the minimum requirement for each profile description was:

• map coordinates defining the site location and
• at least one observation about the site.

All data were transformed to the standard SITES format, which comprises several levels:

site (must have a location)
observation
horizon
sample (may contain lab results)

Only information for which a field exists in the SITES structure was transferred from the data held by the State agencies. The ASRIS database is not a mirror of the State's datasets.

There is considerable variation in the degree of description, and the attributes described for each profile:

• most sites had morphological description of the profile
• many had data on soil texture and pH
• other chemical data was limited
• soil physical data was very uncommon.

Table soil_lab_results.xls shows the number of horizons (not profiles) in the database containing data for selected attributes.

There is rarely an explicit statement of reliability associated with the soil profiles - click here for more information on data quality.

Further data preparation involved

• conversion to common units/scales - for example, ensuring that all depths were in metres (not centimetres or millimetres),
• investigating method differences for laboratory data and calibrating comparable methods where possible - see section on soil chemical data
• interrogating unusual observations and deleting obvious outliers (for example, pH values outside the range 0-14).

Download a detailed technical document that describes the data preparation for specific soil properties used in modelling.

Multiple observations at the same site are possible. These may be either time series at the same point, or observations taken at slightly different locations but recorded as a single site. It is not always possible to distinguish these, particularly as the date of sample acquisition is often not recorded.

From the maps of sample density, it is apparent that sampling density is very uneven. There are areas of high intensity where intensive local surveys have been undertaken and sparse areas where only exploratory surveys have been commissioned. This is to be expected because the ASRIS database represents a collection of diverse data sets from the State and Territory agencies and CSIRO Land and Water.

This heterogeneity of sources and sampling techniques has some major implications for the ways in which the compiled data can be used. Data quality and constraints in using the compiled database are discussed below.

 Table of contents

Data quality

The soil profile data in the ASRIS database has been collected over fifty years or more, by different agencies and for different purposes. As a result, the type and quality of the data varies enormously. There is rarely an explicit statement of reliability, or even survey purpose, associated with the individual data points. At best, information is available about the survey. Within the State agencies, there is a good general knowledge of specific surveys, their quality and utility, but it is very difficult to quantify and record.

In addition to the question of the reliability of individual data points, there is the issue of comparability of data from different surveys. This is a significant issue when using the database for regional and national scale studies.

Some of the main issues affecting the degree to which data from different profiles and/or different surveys can be compared are:

1. differences in the purpose for which the sample was collected. Some are derived from model-based or design-based surveys, some for specific studies (for example, studies of acid sulphate soils), other are dictated by ease of access (for example, along roads).

2. variation in the degree of description, and the attributes described for each profile. Most sites had morphological description of the profile. Many had data on soil texture and pH. Other chemical data was limited. Soil physical data was very uncommon. The table shows the number of profiles in the database containing data for specific attributes.

3. inconsistencies in the way soil horizons are described and named, not always following the agreed standard for description (as set out in the Australian Soil and Land Survey Field Handbook ) including:

• no horizon designator
• non-standard horizon designator (eg E)

4. differences in the methods used for laboratory determinations of specific properties - see section on soil chemical data.

5. inconsistent or non-existent soil taxonomic descriptions. Click here for more information on soil classification. The table contained in this information shows the extent to which common soil classifications are represented in the database.

6. failure to distinguish between a null field and zero result

Preparation of the data for modelling soil properties involved considerable cleaning and checking, and the resulting CSIRO point model dataset is a more consistent and generally higher quality subset of the ASRIS soil profile database.

Number of horizons per laboratory soil test available to ASRIS database.

 Table of contents

Soil chemical data

Soil chemical data are available for a wide range of attributes. However, not all profiles contain data on soil chemistry; and there is no consistent suite of properties for which all profiles have data.

Methods used for laboratory determinations of specific soil properties vary between laboratories and over time. A code is included in the database to describe the method used.

Download a table that lists the properties for which data exists in the ASRIS soil profile database, the analytical methods used and the relevant codes as described in the Australian Laboratory Handbook of Soil and Water Chemical Methods (Rayment and Higginson 1992).

Because different methods have been used, or no method has been recorded, in many cases results for the same property cannot be compared directly. As part of ASRIS, the laboratory methods used were reviewed to determine where different methods could be calibrated and compared. Download the results of this study.

An example - pH measurement

The table below shows the methods used for pH determination in samples in the ASRIS database. Note that 20% of the data did not have a method recorded. In most cases, for example where the soil / liquid ratio differs, results are not directly comparable

Conversions have been derived between methods 4A1 (pH of 1:5 soil/water suspension) and 4B1 (pH of 1:5 soil/0.01 M CaCl2 extract) (Peverill et al 1999). These differ slightly depending on the dataset on which they were based. A new conversion from pHwater to pHCaCl2 was developed using the data from ASRIS (download a detail technical document on model development).

These problems with comparison of data from different laboratories and methods mean that the amount of data available for modelling is sometimes much less than expected. For example, of the 151 810 horizons with pH measurements, around a third can be expected to be from A horizons, giving about 50,000 possible points. Only determinations using methods 4A1 and 4B1 can be compared - this is just over 70% of the data, reducing the number to 35 000. After removal of points that lie outside the modelling area, have no or inconsistent horizon designation or some other inconsistency, the total number of points available for modelling pH of the topsoil was 24 319.

Comparison of laboratory methods used for determination of pH for samples in the ASRIS database

 Table of contents

CSIRO point model dataset

One of the main reasons for compiling the ASRIS soil profile database was to provide data for modelling soil properties on regional to national. Models were constructed for:

• topsoil (layer 1): the first A horizon
• subsoil (layer 2) the first B horizon.

A subset of 135 490 observations with horizons which fit these descriptions were extracted from the database. Further data preparation involved removal of points outside the ASRIS extent, conversion to common units/scales, investigating method differences and calibrating comparable methods where possible, interrogating unusual observations and deleting obvious outliers and null observations. This was done for each soil property separately and is described in a detailed technical document.

The final dataset used in modelling contained 130992 observations, distributed across the 18 regions which comprise the ASRIS region, as shown in Table BB.

The CSIRO point model dataset is a more consistent subset of the ASRIS soil profile database. It contains corrections to the data which have not been incorporated into the ASRIS database, or into the original data held by the States. Note that because it has been extracted for a specific purpose, only the first A and first B horizons have been included. This dataset is available for direct download from the Audit Data Library under the same access conditions as the ASRIS soil profile database.

The table and figure below show the breakdown across the 18 regions covering the ASRIS extent.

 Table of contents

Soil Information Transfer & Evaluation System (SITES) Entity Relationship Diagram

SITES (Soil Information Transfer and Evaluation System) was developed by ACLEP, and includes:

• a National Data Model for soil profile and associated analytical data;
• a Data Exchange Protocol to enable routine data transfer; and
• a generic Windows-based product (pcSITES) which provides an interface to existing databases (PC or networked).

The SITES data model is shown in the entity relationship diagrams (below). The model is fully consistent with the Australian Soil and Land Survey Field Handbook, and the Australian Laboratory Handbook of Soil and Water Chemical Methods. It is described in detail in: "ACLEP (1997) Soil Information Transfer and Evaluation System: Version 1.2. ACLEP Technical Report No.5. CSIRO Land and Water, Canberra."

For more information see http://www.clw.csiro.au/aclep/

SITES entity relationship diagram

Soils Points & Area Collection Elements

 Table of contents

Further Information

Download a document describe the scale of soil maps used in Australian Soil Resources Information System (PDF ??MB) ----------can't find allscales.pdf-----------

• technical document detailing the development of soil property prediction using soil maps (PDF 9.1 MB)
• Download a technical document on point modelling for soil properties (PDF ?? MB) download ASRIS_storyboard\brent.pdf: can't find this file
• technical document detailing estimation of particle size distribution (PDF 1.3 MB)
• technical document detailing laboratory methods and correlation procedure (MS Word 260 KB)
• spreadsheet detailing laboratory methods used and number of samples (MS Excel 86 KB)
• table detailing land use classes used in Australian Soil Resources Information System modelling (MS Word 32 KB)
• statement on data quality for South Australian data sets (MS Word 24 KB)
• statement on data quality for Western Australia data sets (MS Word 24 KB)

Proposed Polygon Standards

• technical document detailing the polygon data standard for land resource data sets (MS Word 77 KB)
• technical document detailing the polygon attributes proposed as part of the polygon standard for land resource data sets (MS Word 24 KB)

 Table of contents