The Statewide Landcover and Trees Study (SLATS) Monitoring Land Cover Change and Greenhouse Gas Emissions in Queensland. Tim J Danaher, Gerard R Wedderburn-Bisshop, Lazaros E Kastanis, John O Carter Climate Impacts and Grazing Systems Resource Science and Knowledge Natural Sciences Precinct Queensland Department of Natural Resources 80 Meiers Road Indooroopilly, 4068 email -
[email protected] URL: http://www.dnr.qld.gov.au/slats/
Abstract A major vegetation monitoring initiative, the Statewide Landcover And Trees Study (SLATS) is currently in progress in Queensland, Australia. SLATS aims to develop accurate vegetation cover mapping so that tree clearing policies and guidelines are based on factual information. Landsat Thematic Mapper imagery is being used to compare the vegetation cover between 1988, 1991, 1995, and 1997 over the entire State of Queensland. A more detailed 1991 baseline survey is also in progress. The primary vegetation and land cover data sets produced will be used to regularly report the extent, cover and trend of Queensland's vegetation cover and land use and provide estimates of greenhouse gas emissions for the Land Use and Forestry sector. The average annual tree clearing rate for Queensland (1991-95) has been calculated as 262,000 ha per year. The clearing figures are available as tables and maps, categorised by tenure, biogeographic region, catchment, local government area and native pasture community type. This information and a range of satellite imagery map products have been developed in consultation with the project advisory committee and are being supplied to government, landholders and the public. A preliminary analysis of biomass cleared using Landsat TM derived change analysis, 1.1 km resolution National Oceanic and Atmospheric Administration (NOAA) satellite imagery and generalised equations for biomass and tree basal area, indicates that approximately 20 million tonnes of live trees are cleared each year. However with the inclusion of woodland thickening, the Land Use Change and Forestry sector would be a net sink not a net source of emissions. As part of a greenhouse new initiative the satellite and woodland monitoring program is set to continue for a further four years and become part of a comprehensive vegetation management system.
Background The National Greenhouse Gas Inventory (NGGIC, 1994) estimates that land clearing for agriculture accounted for 24% of the total national greenhouse gas emissions on a carbon dioxide equivalent basis. The size of this source in comparison to others has focussed attention on land clearing as a major contributor to Australia’s total greenhouse gas emissions. In Queensland, this attention is particularly strong because this State was identified as having the highest source of emissions as a result of tree clearing. Queensland is one of the last States to clear its land for agriculture and pastoralism, the southern Australian States having completed the bulk of their land development decades ago. Most of this clearing for pastoralism occurs in woodland areas, although continued population growth is increasing the demand for urban land which is resulting in forest clearing in coastal and southeast Queensland. With declining terms of trade for agricultural commodities, producers are under pressure to become more efficient. Land clearing in some situations can triple beef
productivity, so there can be considerable economic benefit from clearing. In fact a large proportion of the Queensland beef industry, worth $1.2-1.4 billion per annum, was built on cleared land. Media attention became increasingly focussed on tree clearing, biodiversity and greenhouse issues. This, together with heightened public awareness of conservation issues, provided conservation groups with a platform to apply political pressure to implement strong counter measures such as total bans on tree clearing. Estimates of tree clearing in Queensland, based on surveys of clearing contractors, herbicide sales and other sources, fluctuated wildly from 200,000 ha to 1,000,000 ha per year, depending on the position being argued. This large variation in quoted figures was a source of contention and hampered informed public debate. The Queensland Department of Primary Industries (DPI) estimated that the rate of carbon storage due to regrowth of trees, forest thickening and growth of woody weeds would more than balance the release of greenhouse gases through clearing, and the land use and forestry sector may be a net sink for greenhouse gases in Queensland (Burrows, 1995). In response, the Queensland Government established the Statewide Landcover and Trees Study (SLATS). It aims to provide factual information on land cover and trends in land clearing, tree growth and regrowth for use in policy and community planning decisions regarding sustainable land management and the National Greenhouse Response Strategy. It was originally known as the Queensland Vegetation and Land Use Monitoring Project (Danaher et al., 1996).
Description SLATS set out to improve the greenhouse gas (GHG) inventory for the land use and forestry sector in Queensland and to provide information for review of tree clearing policies in Queensland. In order to determine project priorities, the uncertainties involved in greenhouse gas emission calculations were evaluated (Danaher et al.,1996). The resulting program set out to address the largest of these uncertainties. The main components of the SLATS project are: • The development of a satellite based monitoring system using Landsat Thematic Mapper (TM) satellite imagery for 1988, 1991, 1995 and 1997 to detect change in woody vegetation cover across the entire State; • Mapping the extent of woodlands in Queensland by conducting a detailed baseline landcover survey using 1991 satellite imagery; • Establishment of permanent vegetation monitoring sites to assess woodland thickening trends across the State; • Assessment of the impacts of tree clearing and woodland regrowth on Queensland's greenhouse emissions; • Extension of the broadscale and local tree clearing guidelines covering leasehold land in Queensland; • Production and distribution of products for vegetation management planning and policy including reports, satellite image maps and digital products to government, land holders, other industry and the general public. The SLATS project employs fourteen Queensland Department of Natural Resources (DNR) scientific staff for image processing, field work, greenhouse gas and climate change activities. Two DPI staff are involved in the woodland ecology aspects of the project. Thirteen Silicon Graphics Unix workstations are used for the image and geographic data processing. The project uses ERDAS Imagine along with in house developed image processing software and ArcInfo for vector geographic information systems (GIS) analysis. There are a high quality, consistent set of image processing procedures which has been evolved by the SLATS team for analysing the satellite imagery. Most of these procedures have now been automated and standardised into documented macros and scripts to minimise errors, although considerable visual editing is still performed by trained professionals for an accurate result. These
procedures are all documented on the SLATS local intranet Web pages and are under continuous upgrade and development. The project utilises a huge amount of data to monitor change over the entire State for 1988, 1991, 1995 and 1997. A total of 88 Landsat TM scenes have been acquired for each of these years to cover the entire State. A Cray J90 supercomputer coupled with a Storage Technology 4400 robotic tape library with two Timberline tape drives (one gigabyte (Gb) capacity uncompressed) and two Redwood tape drives (50 Gb capacity) is used to store and retrieve the satellite imagery, derived products and other modelling data. Data is automatically moved either on or off line as required using Cray's Data Migration Facility (DMF) software. The potential storage capacity of the tape library is 770 terabytes.
SLATS Advisory Committee A project advisory committee representing state and local government, producer groups, the conservation movement, research organisations and academia was established. Figure 1 lists the organisations involved. The role of this group is to advise the SLATS project on the needs of the interest groups represented and to suggest future directions of investigation. In addition, this group assists in the dissemination of information generated by the SLATS project through their networks. Over the life of this project the advisory committee has helped develop SLATS products, disseminate information and assist in maintaining ongoing funding for the project. Land and Water Resources Research and Development Corporation
• • • •
Producer Groups United Graziers Association Cattlemen’s Union Canegrowers Graingrowers Association
Local Government Associations
Queensland Conservation Council
Project Advisory Committee
Queensland Government • Department of Environment • Department of Primary Industries • Department of Natural Resources
Emeritus Professor advising on botanical issues
Figure 1 : SLATS advisory committee
Methods Imagery must undergo several pre-processing steps prior to use in change detection and mapping procedures. All imagery was purchased from the Australian Centre for Remote Sensing (ACRES) pre-processed to level 5. Due to the multi-temporal nature of the project, radiometric correction of the imagery is essential. The imagery is first corrected for variation in solar zenith angle and then standardised for atmospheric variations between dates by radiometrically registering all dates to a reference year. Corrections are applied to improve scene to scene matching along each path and a bi-directional reflectance distribution function
correction is also applied. Collett et al. (1998) discusses in detail the radiometric correction procedures. Geometric correction firstly involves registration of all scenes to the 1991 reference year. Semi- automated image correlation methods are being used to register imagery to better than 0.4 pixels root mean square error. The majority of subsequent processing uses these registered images. To enable rectification of change detection classifications, vegetation mapping and other products, scene to map transformations are calculated. Preliminary transformations are generated using ground control points identified on 1:100,000 and 1:250,000 scale topographic maps. The accuracy of these points are later upgraded while in the field using differential global positioning systems (Fugro and Garmin type, with sub 10m accuracy) to maintain a constant spatial accuracy of at least 1:100,000 scale. (Kuhnell et al., 1998). A semi automated technique for change detection has been developed using band reflectance differences and Normalised Difference Vegetation Index (NDVI) differencing methodology. The resulting changes are classified to indicate the direction and magnitude of change. Refer to Paudyal and Kuhnell (1997) for details of this procedure. A woody vegetation mask is used to isolate changes in woody vegetation and help improve the accuracy of the final change classification. Woody regrowth is also classified although this is more difficult to measure due to slow rates of regrowth and the low density of some regrowth stands. A detailed baseline landcover survey is done using 1991 imagery. This involves the development of wooded / non wooded masks, foliage projective cover (overstorey and shrub) and tree basal area layers. A full description is given by Kuhnell et al. (1998). Water, bare ground and shadow masks are created for each scene and crop masks have also been produced using an existing crop coverage (Kuhnell and Danaher, 1996) which is updated using visual interpretation. Settlement areas are delineated using an analysis of digital cadastral boundaries based on lot size. Field validation of remote sensing analysis is an essential part of the SLATS methodology (Kuhnell et al., 1998) and involves approximately 70,000 km of travel each year. Each scene is ground truthed; • • • •
to verify vegetation change classification accuracy; to measure differential GPS control points; to collect vegetation site data for calibration and validation purposes; by liaising with local DNR offices and landholders.
A field based woodland monitoring program, run by DPI Tropical Beef Centre staff in Rockhampton, measure woodland thickening trends across Queensland. As part of the SLATS project, DPI has extended its permanent on-ground woodland monitoring system, known as TRAPS (Back et al. 1997) to include a total of 118 measurement sites. These sites are measured in fine detail by tree bands and callipers. Woodland thickening should not be confused with regrowth after clearing; woodland thickening refers to increases in the woody biomass due to changes in vegetation density from causes such as a reduction in the frequency of fires. Thickening is understood to be occurring across vast areas of Queensland’s woodlands. In addition, another 20 sites across the State have been sampled to confirm woodland thickening by analysis of carbon isotope ratios using the δ13C technique (Burrows et al., 1998). This technique discriminates carbon produced by trees with C3 photosynthetic pathway from carbon produced by C4 grasses which dominate tropical and sub-tropical areas. This technique uncovers areas where grass dominated plant communities have been replaced by thickening trees and shrubs. In order to calculate changes in greenhouse gas emissions resulting from clearing, we need to know the biomass of vegetation cleared. There are known relationships between biomass and tree basal area for some species (Burrows, 1976; Madgwick et al.,1991), so a combination of
the vegetation change mapping and land cover mapping are used to estimate the biomass of cleared vegetation (Lucas et al., 1998).
Vegetation Change Rates Until recently, clearing rates in Queensland were estimated to range from 200,000 to 1,000,000 ha per year. SLATS reports have now settled this, by providing (in 1996) preliminary 1991 to 1995 clearing rates of 308,000ha per year ± 25% and revised figures of 262,000 ha per year ± 10% ; see the SLATS Interim Report (DNR, 1997). Results for three central Queensland scenes indicate that regrowth may be occurring at a rate approximately equal to 43% of the clearing rate, (although this estimate has a high degree of uncertainty). Preliminary indications suggest that the tree clearing rate between 1991 to 1995 has dropped by over 20% when compared to the 1988 to 1991 clearing rates. Farmers, conservationists, governments and scientists now have factual information when formulating policy and making land management decisions. For the SLATS Interim report (QDNR, 1997), a raster of cleared areas was intersected with geographic information system (GIS) overlays of tenure type, biogeographic regions and provinces, catchments, local government areas and native pasture communities. For each polygon the clearing rate (km2 year-1) and also the clearing rate as a proportion of polygon area per year was calculated. This proportional analysis highlighted significant clearing areas in smaller polygons. Once the 1991 baseline landcover mapping is complete it will be possible to provide clearing as a proportion of the area of 1991 woody vegetation in each polygon. Clearing by vegetation communities will also be calculated once current and pre-European vegetation communities are mapped. Clearing totals were accumulated for each of the GIS overlays and statistics for each are provided in table and map form. An example of these maps is given in Figure 2. Other interesting facts detailed in the report were that: • more than half of all clearing was occurring in the Brigalow Belt biogeographic region; • the highest percentage of a biogeographic region cleared was in the Desert Uplands; • over 40% of all clearing was occurring on freehold land vis a vis leasehold land; • the catchment with the highest rate of clearing is the Suttor River and the clearing in the Sunshine Coast catchments is also significant on a proportional basis of the areal percentage of clearing in a shire; • Jericho Shire is the local government area with the highest clearing rate.
Greenhouse Gases Emissions A preliminary analysis of biomass cleared combining the latest SLATS Landsat TM based change analysis, 1.1 km resolution NOAA satellite imagery, generalised equations for tree basal area (Lucas et al., 1998) and biomass per unit basal area, showed that approximately 20 million tonnes of live trees are cleared each year. This does not include standing dead trees, fallen timber, litter and tree roots. It also over-estimates biomass in areas of regrowth clearing. This 20 million tonnes of biomass is equivalent to 33 million tonnes of carbon dioxide (CO2) emissions when timber is burned or eventually decays.In contrast, much of Queensland’s remaining forests and woodlands have been shown to be a sink of the major greenhouse gas carbon dioxide, storing an estimated 100 million tonnes of CO2 per year (Burrows et al., 1998). This additional storage of carbon occurs in both live and dead trees, and possibly the soil. The major cause, woodland thickening, which is the increase in size and density of forests mainly as a result of the decreasing frequency of fire caused by grazing management practices (Burrows et al., 1997). Grazing management tends to reduce the available grass fuel load and graziers tend to regulate the occurrence of fires. The net result of fewer fires is an increase in the number and size of small tree, shrubs and woody weeds over much of Queensland’s savanna woodlands. Some would argue that it is actually a form of land degradation and that increased use of fire should be used to slowly return these thickened areas to open grasslands. The sink provided by woodland thickening is finite, as carbon stored by woodlands will eventually reach a new equilibrium consistent with current fire regimes. Climate variability and CO2 fertilisation may also be contributing factors to changing tree density.
Despite the considerable scientific evidence documented by Burrows et al.(1998), which shows woodland thickening is real, it still has not been included in the National Greenhouse Gas Inventory. Although there are uncertainties associated with these estimates the sink is almost certainly larger than the sources from land clearing. If thickening is included, the Land Use and Forestry Sector will be a net sink, the relative contribution of other sectors will therefore be increased, but Australia’s baseline per capita emissions would be reduced in total. However, any future emissions from thinning and removal of biomass (e.g. increased fire and climate change) would have to be monitored and added to the sources of emissions. The sectoral and national implications of the Kyoto agreement in relationship to the above remains a complex issue.
SLATS Products Promotion of the SLATS project, notably through “Infomercials” run on regional television and availability of many of the products via on-line internet catalogues and direct intranet downloads, has resulted in a large increase in demand for SLATS data and products. Demand for data (mostly digital) from land management professionals within the State Government has been very strong, and demand for hardcopy maps from landholders and farm managers is increasing rapidly. A number of SLATS products suitable for public use were identified in conjunction with the SLATS advisory board and were developed during the project. They consist of hardcopy Landsat images with property boundary and road overlays at 1:250,000 and 1:100,000 , Landsat images with property boundaries at 1:50,000, vegetation change maps at 1:250,000 and land cover and vegetation density maps at 1:100,000 and 1:250,000 scale. Some data sets are also available in digital form. The SLATS web site at http:/www.dnr.qld.gov.au/slats/ has full details of product availability. These SLATS products form a complete statewide series. Already over 250 hard copy products and 300 compact disks of digital data have been distributed to clients. Additional hard copy products integrating the Department of Environment regional ecosystems boundaries and Landsat TM imagery are being considered. The full range of products is covered in more detail by Walls et al. (1998).
The Future In the two years that SLATS has been in progress there has been a huge demand for the information and products produced. The availability of factual information has moved the debate on land clearing forward with all parties adopting the SLATS figures. The project has also provided crucial information for the establishment of Australia’s national greenhouse position. SLATS is sometimes criticised for the fact that it has aimed at monitoring after clearing has occurred and that there is a lack of close integration between satellite monitoring, vegetation community information and the tree clearing permit system. These points will be addressed as a comprehensive vegetation management planning system is developed. A vision for an ideal (and achievable) framework for vegetation management decisions based on the best science, geographical data, local knowledge and information technology would be as shown in Figure 3. If the whole State’s biodiversity resources were mapped, landholders, conservationists, government and academia could possibly decide what levels of development were acceptable, and some type of agreement reached. If any landholder wanted to clear beyond the regional conservation limits, he/she might possibly be able to buy or trade another development right somewhere else in the region (that had not been cleared at that stage).
State and Regional Vegetation Management Plans developed for the whole State. Conservation limits defined
Development request from land holder
• • • • •
Primary risk assessment using geographical data: Available potential within vegetation communities and bioregion Wildlife and endangered vegetation assessment Salinity hazard assessment Water course buffers Pasture soil/rainfall cost/benefit modelling e.g. GRASSMAN
• • • • • • •
Farm utilisation options examined, e.g. Farm production/grazing, shade Greenhouse sink Conservation area Forestry Water catchment protection Wildlife corridor potential Method of clearing
• •
Property Vegetation Management Plan developed Development rights (with defined limits) described These rights potentially tradeable, giving increased investment certainty and conservation guarantees
Figure 3 : Vegetation Management Planning Decision Framework This framework would allow certainty for farmers’ development and investment decisions, and have positive, verifiable environmental outcomes for conservationists, the community and government. Farmers’ export potential would be guaranteed through the demonstrated ability to address international standards for biodiversity management and greenhouse issues. In some areas of the State where leasehold land is the predominant tenure, parts of this framework are already in place. The whole system would rely on GIS technology for spatial information storage and query, remote sensing would continue to be used to monitor vegetation changes and ground checks would still be required to ensure accuracy. How this could be achieved across all land tenure types, not only on leasehold State land (which comprises 75% of the State). is still being debated. Voluntary codes of practice or a legislative framework are two possible options being investigated by working groups. Other issues that need to be investigated include incentive schemes and development of a carbon credits scheme which is adaptable to the uncertainties in biological carbon stores. While this is being resolved many of the components required for such a comprehensive planning system are being put together.
A monitoring system is essential to ensure that Queensland’s vegetation retention policies and practices are ecologically sustainable and the biodiversity status of regional ecosystems is maintained. Through a greenhouse new initiative of the Queensland Government, SLATS will continue to be funded for a further four years. Satellite imagery will be used to routinely monitor regions where clearing has previously been detected every two years. The whole State will be analysed every four years. Some historical Landsat Multi-Spectral Scanner (MSS) imagery (early 1970’s, 1979/80 and 1984) will be acquired and analysed to provide a better historical context of clearing rates, regrowth age, soil carbon run-down and greenhouse calculations. Woodland thickening trends will continue to be monitored using permanent field monitoring sites and laser profiling techniques (Tickle et al., 1998) will be investigated as a means of additional monitoring. The new initiative will have a targeted extension program to promote responsible vegetation management practices, the Queensland tree clearing guidelines, tree management planning, the SLATS project and on-farm use of SLATS satellite products across the State. The mapping of pre-European remnant vegetation and regional ecosystems is also an essential component of a new planning framework. In another major new initiative, facilitated by SLATS, the Queensland Government has funded the Department of Environment (DoE) for survey and mapping of vegetation and regional ecosystems for a further two years. This information will facilitate the refinement of local tree clearing guidelines and the identification of regional tree retention requirements across all tenures. Integration and use of this information is being trialed in the development of a regional vegetation planning GIS system. It is a one year project funded by the Natural Heritage Trust (NHT, 1997). The project will cover the north-west section of Jericho Shire, in the Desert Uplands Biogeographic Region of Central Queensland. It will integrate DoE vegetation community boundaries, be updated using the latest SLATS satellite monitoring data and include tree management plans for all properties in the pilot area. It will be used for assessment of vegetation changes proposed within the Regional Tree Retention Plan against policies and local guidelines and existing vegetation status as recorded within the GIS. The project will identify the framework needed and cost to expand the process into a state-wide program. In conclusion, the SLATS project is providing a solid foundation to implement a fully integrated approach to vegetation management based on appropriate technology, best science and embracing ecologically sustainable development principles.
Acknowledgments The authors would like to thank the Queensland Government for their vision and support of this work, especially a $7.2 million Special Treasury Initiative grant without which all this would not be possible, the DNR Resource Management and Resource Sciences and Knowledge directorate, all the DNR Climate Impacts and Grazing Systems staff and DPI Tropical Beef Centre staff who have contributed to this work, the project advisory committee for their advice and enduring support of the project, Mr Gus McGown for his constant advice and support, and the Commonwealth Bureau of Resource Sciences which contributed 10% of the project’s firstterm funding.
References Back, P.V., Anderson, E.R., Burrows, W.H., Kennedy, M.K.K. and Carter, J.O. (1997). TRAPS transect recording and processing system: Field guide and software manual. (QDPI: Rockhampton). pp. 45. Burrows, W.H. (1976). Aspects of nutrient cycling in mallee and mulga communities. PhD thesis, ANU. Burrows, W.H. (1995). Greenhouse revisited - an alternative viewpoint on land use change and forestry from a Queensland perspective, Climate Change Newsletter, 7:6-7. Burrows, W.H., Anderson, E.R., Back, P.V. & Hoffmann, M. (1997). Regrowth and Woody Plant Thickening/Invasion Impacts on the Land Use Change and Forestry Inventory, In: Proceedings IPCC Workshop on Biomass Burning, Land-Use Change and Forestry, Rockhampton, Australia (IAE/OEDC: Paris) (In Publication). Burrows, W.H., Compton, J.F. and Hoffmann, M.B. (1998). Vegetation thickening and carbon sinks in the grazed woodlands of north-east Australia, In: Proceedings Australian Forest Growers Conference, Lismore, 1998. (In publication) Collett, L. and Goulevitch, B. (1998). SLATS Radiometric Correction : A Semi Automated Multi Stage Process for the Standardisation of Temporal and Spatial Radiometric Differences Proceedings, In: 9th Australasian Remote Sensing and Photogrammetry Conference, Sydney, Australia, July 1998. (These proceedings). Danaher, T., Carter, J., Brook, K., Kelly, G. & Goulevitch, B. (1996). Monitoring Vegetation Cover Change And Its Impact On Queensland’s Greenhouse Gas Emissions, In: 8th Australasian Remote Sensing Conference, Canberra 25-29 March, 1996, 2:128-134. Lucas, R., Tickle, P. and Carter, J. (1998). Australia’s National Biomass Inventory : Advancing the Use of Remotely Sensed Data for Estimating Biomass and Biomass Increment, In: Proceedings, 9th Australasian Remote Sensing and Photogrammetry Conference, Sydney, Australia, July 1998. (These proceedings). Kastanis, L., Wedderburn-Bisshop, G. and Danaher, T. (1997) From Global Greenhouse To Sustainable Tree Management, In: Proceedings, Athens International Conference Urban Regional Environmental Planning And Informatics to Planning in an Era of Transition, Athens, Greece, 22-24 October 1997, pp. 707-725. Kuhnell, C., and Danaher T. (1996). Mapping broadacre cropping areas in Queensland using Landsat TM and NOAA imagery, In: 8th Australasian Remote Sensing Conference Proceedings, (2)16-23. Kuhnell, C., Goulevitch, B., Danaher, T. and Harris, D. (1998). Mapping Woody Vegetation Cover over the State of Queensland using Landsat TM Imagery. In: Proceedings, 9th Australasian Remote Sensing and Photogrammetry Conference, Sydney, Australia, July 1998. (These proceedings). Madgwick, H.A.I., Fredrick, D.J. and Thompson Tew, D. (1991). Biomass relationships in stands of eucalyptus species. Bioresource Technology, 37:85-91. Natural Heritage Trust (1997). Facilitating Sustainable Pastoral Development and Native Tree Retention in Queensland, Proposal submitted by the Queensland Department of Natural Resources.
National Greenhouse Gas Inventory Committee (1994). Australian Methodology for the Estimation of Greenhouse Gas Emissions and Sinks, Workbook for Carbon Dioxide from the Biosphere, Workbook 4.0. National Greenhouse Gas Inventory Committee (1996). State and Territory Greenhouse Gas Inventory 1988 and 1990 Queensland, Department of Environment Sport and Territories. pp. 249. National Greenhouse Gas Inventory Committee (1997). Land Use Change and Forestry, Workbook for Carbon Dioxide from the Biosphere, Workbook 4.2 Revision 2, Environment Australia. pp. 76. Paudyal, D., Kuhnell, C. & Danaher, T. (1997). Detecting Change in Woody Vegetation in Queensland using Landsat TM Imagery, In: Proceedings, North Australasian Remote Sensing and GIS Conference (NARGIS), Cairns, April 28-30,1997. (In publication) Queensland Department of Natural Resources (1997). Interim Report - The Statewide Landcover and Trees Study (SLATS), October 1997. pp. 37. Scanlan, J.C. and McKeon, G.M. (1990). GRASSMAN - a computer program for managing native pastures in eucalypt woodland. Queensland Department of Primary Industries. Tickle, P., Witte, C., Danaher, T. and Jones, K. (1998). The Application of Large-Scale Video and Laser Altimetry to Forest Inventory, In: Proceedings, 9th Australasian Remote Sensing and Photogrammetry Conference, Sydney, Australia, July 1998. (These proceedings). Walls,J., Kastanis, L. and Plunkett, J. (1998). Development of Integrated Products for Tree Management Planning. In: Proceedings, 9th Australasian Remote Sensing and Photogrammetry Conference, Sydney, Australia, July 1998. (These proceedings).
