Abstract

The Department of Land and Water Conservation is the custodian of the NSW Soil and Land Information System (SALIS) and is charged with mapping soils across NSW through the Soil Landscape Mapping Program. This includes Soil Landscape information and soil profile information as well as significant data holdings from other agencies, universities and consultants, some held confidentially.

Approximately 50% complete, Soil Landscape Maps are the primary land resource assessment data set in NSW. A feature is the wide range of attributes that can be filtered and recombined to produce derivative maps of specific soil and land characteristics. The community benefits of this program have been documented as being over 40 times the cost.

SALIS currently contains over 48,000 soil profile descriptions and 24,000 laboratory test results, mostly from eastern and central NSW, plus 13,000 observations from the NSW Land Degradation Survey, in an Oracle 8i Spatial database. Soil Landscapes maps and other spatial data types are being progressively added. Soil profile information is available over the Internet via the Soil Profile Attribute Data Environment (SPADE) spatial viewer at http://spade.dlwc.nsw.gov.au. Current development aims to make all public SALIS data, including maps, accessible over the Internet.

Introduction

Human quality of life depends on the capacity and quality of our soils to produce most of our food and fibre. It can be argued that as a community we have been ignorant in the management of soil resources. Soils are essential to ecosystem health and are important for storing, filtering and sequestering nutrients and pollutants. Massive soil erosion has occurred historically in the 1890s through to the 1960s (Russell and Isbell, 1986). Although the current situation is certainly not as critical as in previous times of rabbit plague and drought, much land remains at risk from soil erosion. For example, up to one million tonnes of topsoil is estimated to have been lost from the Mooki-Namoi catchments during recent floods (Lang, 2001). Other, more recent and often more subtle and complex soil and land degradation issues such as dryland salinity, mobilisation of salt stores to water bodies, soil acidification, soil compaction and structural breakdown, sodicity, pollution and alienation of quality soil through urbanisation of highly productive land are issues that can threaten what is essentially a non-renewable resource base.

There are many types of soils in New South Wales, ranging, for example, from friable well-structured loams formed on basalt that support rainforests, organic peats of coastal swamps and the brown organic soils of alpine regions, through to the highly productive loams and clays of the cereal belt, the saline and gypseous scalded clays and red dunes of the Western Division. This geographical diversity is reflected by the variation in soil capability, in particular the reaction to change or degradation. Due to their inherent properties reacting and adapting differently to disturbance and environmental changes, soils in NSW vary in their resilience and have a remarkably diverse range of recovery and improvement properties. Many soils in New South Wales have properties that make management more difficult compared with soils in other parts of the world. For example, NSW soils are often old, shallow, fragile and infertile compared to soils in other countries.

There is a need to manage what is arguably our most fundamental and non-renewable natural resource (Chapman et al., 1999) in a sustainable manner, as the majority of terrestrial biodiversity is found in soils (Blum, 2000). This means matching soil capability to land use and land management. A comprehensive, useful, accessible resource of soil and land information is key to this process, consequently the NSW government has a number of soils information programs to encourage evidence-based policy, knowledge-based land use planning decisions and increasingly wise land management.

The Department of Land and Water Conservation (DLWC) has responsibility for Land and Resource Assessment across NSW. This paper outlines some of the main activities of the NSW Government in relation to provision of spatial soils information.

Market failure in Land Resource Assessment and the response of Government

PDP Australia Pty. Ltd. (1993) undertook a review of Land Resource Assessment in Australia on behalf of the Federal Government and found that:

Market failure in Land Resource Assessment was a major problem.

The benefits of resource information are shared between many users and no one user would be prepared to pay the "efficiency price"

The Australian Collaborative Land Evaluation Program (ACLEP) commissioned ACIL (1996) to undertake a review of the costs and benefits of Land Resource Assessment in Australia. One of DLWC’s land resource assessment products, the Gosford-Lake Macquarie Soil Landscapes Map and Report (Murphy, 1993) had benefits that conservatively exceeded costs by over 44:1, whereas the benefit:cost ratio for the same area in terms of independent survey was considered to be 17:1. Whilst this is reassuring for long-term land resource assessment, the question remains about market failure of soils information and, in particular, why this level of potential benefit has not necessarily been translated into improvements in the sustainability of land management.

Reviews of users and potential users of Soil Landscape information by Mulligan and Mesiti (1994) and Rawson (1998) found that the information was valued, but that substantial barriers remain concerning the use of soil information within the general community.

Soil information can potentially answer two important kinds of question:

I have a parcel of land. What characteristics does it have? How can I make best use of it? Will others (such as my bank manager) consider this to be a sustainable option both in social, financial and environmental terms?

I have a particular venture in mind. Where is the best place to locate this enterprise for maximum economic, ecological and social advantage? What will I need to do in order to manage, maintain and/or improve the land on a long-term basis?

The barriers against the effective use of soil information by the community can be illustrated using the second of these examples: purchasing a property for a particular purpose.

Time and expense is required to gather information about land conditions: soils, landform configuration, vegetation, climate, land use and management. The effort and expense of finding and using this information competes with other requirements, such as conveyancing and effluent disposal, but is often small in comparison with investment decisions. However, soil and land information may also mean the difference between success and failure of the whole enterprise. But before making effective use of this information the prospective user must overcome several fundamental barriers:

Making the intellectual connection between biophysical conditions and the possible success or failure of a particular enterprise. This relates to general awareness of the relationships between natural resource activities and environmental health, and is a major challenge for an increasingly urbanised population.

Knowing that the information exists in the area of interest in the first place.

Knowing where to search for such information. For example, where can I find information about soils?

Having found the information, recognising what role might it play in reducing uncertainty in decision-making (Pannell, 2000). For example, what difference is this information likely to make with regard to my decision? What are the risks I must endure on this decision for it to pay off? What if the advice is wrong? Is the information reliable enough?

DLWC, the NSW Government’s custodian of soil and land information, is in the prime position to address all four of these problems. DLWC is pursuing a program of Soil Landscape mapping that is progressively covering the Central and Eastern Divisions of NSW. This work provides both point-based soil descriptions and descriptions of the landscape in which these soils are found. The Soil Landscape Mapping Program provides a detailed spatially-based information resource suitable for land use planning from a regional scale right down to the property level. This information is being brought together in combination with other kinds of natural resource information into a centralised repository of natural resource data. This substantial data warehouse is being made progressively accessible to the NSW community via a single information gateway, the Soils NSW Access Portal (SNAP), along with other information products. Each of these advances is addressed in detail below.

Soil Landscape Mapping

In 1987 in agreement with other Government Departments, the then NSW Soil Conservation Service (SCS) commenced a program to formally describe and map the soils of Eastern and Central NSW. The western area of the State had previously been mapped using the closely related methodology of Land Systems (Walker 1991).

Currently, DLWC has 16 Soil Surveyors based in regional locations across eastern and central NSW, carrying out a program of integrated Land Resource Assessment called Soil Landscape Mapping. Soil Landscapes are “areas of land that have recognisable and specifiable topographies and soils, that are capable of being presented on maps and can be described by concise statements” (Northcote 1979). This holistic mapping methodology relies on the fundamental fact that the processes that form soils and the landscapes in which they occur are broadly the same, allowing the spatial extent of soils to be mapped using the characteristics of the landscape. In turn, this allows land management to be tailored to the properties of both the landform and its soils, providing a powerful tool for promoting environmental sustainability. The Soil Landscape Mapping process involves both the detailed description of the landscape, including its topography, geology, climate, vegetation, land use and limitations, and the collection of hundreds of detailed soil descriptions.

Both Soil and Landscape constraints and properties can be portrayed on a single mapping unit. Consequentially interactions between soil and land characteristics can be geographically enunciated. (Murphy et al. 2001). A particular feature of the Soil Landscape mapping program is that as many biophysical soil and landscape attributes as practical are measured or assessed. The data set collected includes over 100 landscape and soil variables, ranging from geology and regolith, native vegetation structure and conspicuous species, land-use, land degradation, patterns of soil distribution, details of soil materials (horizons) and morphological descriptions of type profiles. In addition for each soil material encountered in each major soil profile type samples are taken for laboratory testing for 23 properties including chemical, engineering and physical tests. Samples from soil materials in type profiles are indexed and archived for future research and testing purposes. So far the NSW State Soil Survey Sample Repository includes over 10,000 samples.

Soil Landscape mapping covers 13.5 million ha which is around 30% percent of the target area. In addition, a further 20% of NSW has been mapped using Soil Landscapes at a reconnaissance for regional assessment purposes (Figure 1). Using current resources the program will be completed in 2010, although it could be completed sooner than this date using enhanced resource assessment methods. For example, as part of the Southern Brigalow Interim Bioregion Assessment, extrapolation of training areas and rule sets from previous soil landscape mapping in accordance with environmental variables such as Digital Elevation Models (DEMs), lithology maps, Airborne Gamma Radiometrics, and climate surfaces to map reconnaissance-level Soil Landscapes at 1:100,000 scale over 54,400 square kilometers (Goldrick et al. 2001).

Figure 1: Extent of Soil Landscape Mapping in New South Wales.

Clearly, information certainty and reliability are important issues. These issues are linked to scale and presentation of variability within map units, realising that soil features often vary at scales more intense than the original scale of mapping. However, this variation can be described through the analogous use of landscape factors to describe soil distribution, for example by using specific landscape factors that are known to co-vary with soil parameters. To use a hypothetical example, the relationship between landform and soils within a single Soil Landscape could be summarised as shallow rocky soils on crests and deep sandy soils on flats.

However, whilst consideration of map purity in prediction of important factors is vital, it is often difficult to assign reliability over multiple map units for multiple variables. For example, Soil Landscape mapping includes consideration of hundreds of soil and land variables ranging from predicted exchangeable calcium, soil depth and percentage clay in all horizons, to rockfall hazard, erodibility and plant available water capacity. Furthermore, quantification of environmental variables and map purity, and the ability to predict likely conditions at any particular site, are often dependent on sparse data and the innate skills of the soil surveyor in correctly perceiving and interpreting landscape and soil-forming processes. Soil Landscapes aim to predict correctly for 85% of each soil landscape, a figure that has been verified on testing (Dewar et al. 1996). Of course, for intensive developments further on-site examination of soil properties is often required, and in such situations environmental consultants often find existing Soil Landscape maps provide them with timesaving indications of what to expect.

Capture and Storage of Soil and Land Information

Although Land Resource Assessment information is inherently useful when produced in its traditional forms (hard-copy reports and maps) a whole gamut of new uses are made available if the information is stored and analysed electronically. This is particularly the case if the information is stored in one place, allowing a single consistent dataset to be built up across large areas.

The NSW Government’s first venture into the field of centralised soil information management was the NSW Soil Data System (SDS). SDS was essentially a centralised electronic filing system for soil profile information for NSW, with geolocated soil descriptions entered via optically scanned Soil Data Cards. More recently, advances in data management technology have seen SDS replaced by the NSW Soil And Land Information System (SALIS). SALIS is fully spatially enabled through the use of the Oracle 8i Spatial database, allowing both point and polygon (map) information to be stored and analysed. Both SALIS and its predecessor accept information both from internal DLWC sources and from over 200 external Members in the NSW community, including scientists, geotechnical consultants, engineers and educators.

Data is contributed to SALIS for the following reasons:

• It allows information to be made available for use by others. It can be used, reused and reinterpreted time and time again for purposes far beyond the original intention for collection. This is particularly important from an altruistic viewpoint, as it adds to the body of knowledge about soil and land conditions in NSW and is of assistance for future natural resource management. Other users prefer to hold their data confidentially and use SALIS as a broker.
• It contributes to data security. SALIS information is held safely on a permanent basis and can be valuable for benchmarking. Too often soils information is used for its primary collection purpose and then filed away, minimising any possibility for further usage. It has been estimated by Naylor (pers. comm., 1998) that over 100,000 soil profile records are held in relatively organised institutional filing systems across NSW. A similar amount of data has probably been discarded over recent years, lost as those responsible for it resign or retire. In addition to the above data there is probably 10 to 100 times more information of this type in field notebooks, soil test results and related legacy records that cannot be georeferenced.
• It avoids duplication of effort. Existing data can be referred to in many instances without new data having to be collected over the same area.

Around half of the soil profile point data in SALIS is held confidentially (accessible only by its owner or system administrators) with the remainder being publicly available. Around one third of the data in SALIS is contributed via DLWC’s Soil Surveyors, with the rest coming from other DLWC staff or from other agencies and external clients. At present there are just over 48,000 soil profile descriptions in SALIS, and the database is growing rapidly with over 17,000 profiles being entered during the last twelve months. Figure 2 shows the location of Soil Profile Points in SALIS.

Figure 2: Soil Profile Points in SALIS as at 10^th May 2001.

The versatility of SALIS allows it to store other data sets, not just soil profile descriptions like its predecessor. Besides soil profile descriptions there are also approximately 24,900 soil laboratory test results and 13,000 data points from the NSW Land Degradation Survey (Graham, 1989). SALIS is also intended to hold soil and land map and attribute data, the first of which are DLWC’s Soil Landscape mapping units.

Points to Polygons: Critical Mass and Data Types

Issues of critical mass are important for geographic databases. Profile points are useful when people wish to know about the soil and land at a particular location. However, using only point data there will always be gaps between points in the database, and users would need to guess soil composition at their area of interest from the closest points available. In many situations the variation in soil properties is greater than the difference in soil variables shown between points (Figure 3 shows the distance between points at varying scales), thus the user is multiplying the risks by using point data alone.

Figure 3: The distance between points near Taree at a small scale (NSW) and then at a larger scale. This large scale map shows a realistic perspective of distance between points.

Despite the vortex of scale that applies to soil point maps, soil profile descriptions and observations form the backbone against which soils maps must ultimately rely. Soil maps remove the data point gap uncertainty problem by interpolating between points. At present the greatest priority for SALIS is to populate the database with Soil Landscape maps information. Plans for population of SALIS with Soil Landscape report information have been prepared and are awaiting approval.

Retrieving and Interpreting Soil and Land Information

All Public soil profile data from SALIS is available at cost of transfer, or can be obtained free of charge using the SPADE (Soil Profile Attribute Data Environment) Internet spatial viewer at http://spade.dlwc.nsw.gov.au (Milford et al. 2001). Soil profile information can be delivered either as self-contained reports or in spreadsheet form.

DLWC’s Soil Landscape mapping information is currently produced as 1:100,000 map sheets and comprehensive hard-copy reports. In addition to the base maps and reports, a wide range of Derivative Maps can be produced, each mapping the extent and severity of a particular soil or land quality, limitation, hazard, suitability or capability (Examples can be seen in Table 1 and Figure 4). These can range from maps of topsoil pH through suitability for underground pipes and cables to predicted pre-1788 native vegetation. Derivative Maps are currently produced using DLWC’s Geographic Information System (GIS) resources, however the capability to store and analyse this information using SALIS means that in the near future these products can be generated and accessed directly from the database and made available by using new technologies, such as the Internet.

• On-site sewage disposal & wastewater site location strategies for local government (Chapman et al., 2001).
• Location of grapes, tea tree, olives & other exotic plantations.
• Location and planning of private & public forestry plantations (Bureau Resource Sciences, 2000)
• Planning forestry operations, including EPA soil regolith stability classification for forest harvesting (Murphy et al, 1998).
• General planning & urban & agricultural land capability assessment by local government.
• Used by diverse range of state government agencies & consultants (eg NSW Agriculture, State Forests, DUAP).
• Specialist acid sulfate soil risk maps were developed from soil landscape maps by DLWC soil surveyors. (Naylor et al.,1998) These form the basis for local government planning.
• Aquaculture (including oyster farming) suitability planning and assessment.
• Pipeline & optical fibre cable routing.
• Groundwater vulnerability mapping & mapping of sand bodies.
• University & TAFE rural studies.
• Property planners for physical farm planning
• Used by consultants in EIS as a framework for detailed soil survey.
• Property plans & property agreements.
• “Farming for the future” & other community extension programs & field days.
• Compilation of erosion & mass movement hazard maps for use in vegetation clearing assessment.
• Catchment Planners use maps to identify catchment soil types & relationships with geology, hydrology, & vegetation issues, to assess catchment land capability & to delineate problems such as salinity risk areas & dispersible soils.
• Preparing Rivercare plans.
• Planning soil conservation works.
• Agro-forestry assessments.
• Used by Landcare Groups to assist with landscape management issues.
• Assist to delineate salinisation zones.
• Identification of groundwater dependent ecosystems.
• Mapping of non-riverine sand & gravel sources.
• Assessment of wetland types.

Table 1: Some Examples of Natural Resource Management Applications of Soil Landscape Mapping

Figure 4: Capability for septic tank absorption trenches.

Future Developments in Data Access

The Soils NSW Access Portal (SNAP) is a key focus for making DLWC’s soil and land information more accessible and usable by the NSW community. Building on the foundation provided by the SPADE soil profile spatial viewer, further developments are now underway to provide spatial information (both points and map polygons) on the Internet in the same way through the SNAP gateway.

In order to address the widely varying requirements of the various kinds of prospective users (ranging from school children through land holders, consultants, scientists and planners) SNAP is planned to provide two ways of accessing this information. The needs of general non-technical users will be addressed through a ‘Soils Map Server’ Website that will provide access to static, non-interactive Derivative Maps and other information for each 1:100,000 map sheet area.

Technical users will be catered for by an enhanced spatial viewer providing online active maps with multiple layers of spatial information, and the capability to query and analyse underlying data. It is also proposed to supply advisory information, including brochure-based soil and land management advice, via another Website. Taken together, these three initiatives should address a broad spectrum of requirements for both current and prospective users of NSW’s soil and land information.

Conclusion

The information base provided by soil profile information and Soil Landscape mapping make available a powerful resource for improving the sustainability and profitability of land management across NSW. A comprehensive Statewide coverage will provide additional benefits in both regional, State and national resource inventory, planning and policy, but will also address the needs of many disparate users in the NSW community. In the foreseeable future NSW citizens with Internet connections wanting soil and/or land information will have a wealth of information available to answer their questions. The information will be accessible, formatted or formattable to address their specific purpose, and the information can be assessed for its reliability. In the meantime there is much work to be done.

References

ACIL (1996). An economic framework for assessing the benefits and costs of land resource assessment in Australia. In Australian Collaborative Land Evaluation Program Newsletter 5:4, ACLEP.

Blum, W.E.H. 2000, Challenges for Soil Science at the dawn of the 21^st century, Soil 2000: New Horizons for a New Century, Australian and New Zealand 2nd Joint Soil Conference Volume 1: Plenary Papers, J.A. Adams and A.K. Metherell (eds), 3 – 8 December 2000, Lincoln University, New Zealand Society of Soil Science.

Bureau of Resource Sciences (2000). Identification of plantation opportunities in New South Wales - Southern NSW CRA Region, Bureau of Resource Sciences.

Chapman, G. A., Edye J.A., Kenway, S., Milford, H.B., Murphy, C.L., McGaw, A.J.E. and Simons, N.A. (2001). Soil information for on-site effluent management. In Proceedings of the Geospatial Information and Agriculture Symposium, Sydney 2001.

Chapman, G.A., Zierholz, C. and Davies, B. (1999). The Importance of Soil and Landscape Characteristics in Determining Fire Regimes and Appropriate Management Strategies. In Bush Fire Management Conference Proceedings, Nature Conservation Council of NSW.

Dewar, R.B., Anwar, S., Parker, J., Chapman, G.A. and Houghton, P.D. (1996). Landscape modelling for natural resource assessment, NSW Department of Land and Water Conservation.

Graham, O.P. 1989, Land Degradation Survey of NSW 1987-1988: Methodology, Soil Conservation Service of NSW (SCS) Technical Report No. 7, SCS, Sydney.

Goldrick, G., Chapman, G.A., Simons, N.A., Milford, H.B., Murphy, C.L., McGaw, A.J.E., Edye, J.A., and Macleod, A.P. (2001). New technology and soil landscape mapping in N.S.W. In Proceedings of the Geospatial Information and Agriculture Symposium, Sydney 2001.

Lang, Des (4^th January 2001). A million tonnes of soil washed away in flood. In Namoi Valley Independent.

Milford, H.B., Simons, N.A., McGaw, A.J.E., Nixon, K.J., Chapman, G.A., Murphy, C.L. and Edye, J.A. (2001). The Soil Profile Attribute Data Environment (SPADE) : NSW Soil Information Online. In Proceedings of the Geospatial Information and Agriculture Symposium, Sydney 2001.

Mulligan S. and Mesiti L. (1994). Soil Survey Program Questionnaire, School of Agriculture and Rural Development, University of Western Sydney, Hawkesbury.

Murphy, C.L., Macleod, A.P., Chapman, G.A., Milford, H.B., McGaw, A.J.E., Edye, J.A. and Simons, N.A. (2001). NSW State Soil Landscape Mapping Program and Derivative Products. In Proceedings of the Geospatial Information and Agriculture Symposium, Sydney 2001.

Murphy, C., Fogarty, P. and Ryan P. (1998). Soil Regolith Stability Classification for State Forests in Eastern New South Wales, Technical Report No. 41, Department of Land and Water Conservation, Sydney.

Murphy, C.L. (1993). Soil Landscapes of the Gosford-Lake Macquarie 1:100 000 Sheet Report, Department of Conservation and Land Management.

Naylor, S.D., Chapman, G.A., Atkinson, G., Murphy, C.L., Tulau, M.J., Flewin, T.C., Milford, H.B. and Morand, D.T. (1998). Guidelines for the Use of Acid Sulfate Risk Maps, 2^nd edition, Department of Land and Water Conservation, Sydney.

Northcote, K.H. (1979). A Factual Key for the Recognition of Australian Soils, 4th edn, Rellim Technical Publications, Adelaide, SA.

Pannell, D. (2000). What is the Value of a Sustainability Indicator? On the NSW Agriculture web site at http://www.agric.nsw.gov.au/reader/3631

PDP Australia Pty Ltd (1993). Land Resource Assessment in Australia, A Review of Commonwealth Support. PDP Australia Pty Ltd.

Rawson, J.W. (1998). Department of Land and Water Conservation Soil Landscape Program Market Survey, Polymex, Australia.

Russell, J.S. and Isbell, R.F. (1986). Australian soils: the human impact, University of Queensland Press.

Walker, P.J. (1991). Land Systems of Western New South Wales, Technical Report No. 25, Soil Conservation of NSW, Sydney.