1Resource Management Officer (Water)
PO Box 865 Dubbo NSW 2830
Ph. 02 68811274, Fax. 02 68811295
2Spatial Information Officer
LMB 21 Orange NSW 2800
3Resource Information Officer (GIS)
Department of Land & Water Conservation
PO Box 512 Albury
4Resource Analysis Manager
Department of Land & Water Conservation
PO Box 1840 Dubbo
Remote sensing and GIS technologies have been used to facilitate the development of a Cap management strategy for the Barwon-Darling River in western NSW. The Murray-Darling Basin Cap on Diversions (Cap) limits annual diversions to those that would have occurred with 1993/94 levels of irrigation development. In the Barwon-Darling River, diversions are yet to be managed to achieve the Cap and the Barwon-Darling River Management Committee has until September 2001 to recommend a Cap management strategy to the Minister for Land and Water Conservation.
One of the major impediments to the development of a Cap management strategy has been a lack of accurate data on the level of irrigation development that existed in 1993/94 – the year upon which the Cap is based. Remote sensing and GIS technologies, in addition to a comprehensive landholder survey, were used to determine on-farm water storage capacities, irrigation infrastructure and cropped areas. These data are being used to upgrade the Integrated Quantity and Quality Model (IQQM), used by the Dept of Land and Water Conservation (DLWC) to model water use.
Remote sensing and GIS technologies served not only as a tool for collecting the necessary data, but through the transparency of the process, engaged the community and improved community confidence in the data that the DLWC is now using to determine water usage in the valley.
Water management in NSW
The NSW Government has embarked on an extensive water management planning process involving community-based water management committees. Committee membership includes all major stakeholder groups: water users, conservationists, Aboriginal community representatives, Local Government, catchment management and various state government agencies. The Committees are responsible for advising the Minister for Land and Water Conservation on water management, through the development of Water Sharing Plans.
The Water Sharing Plans set the Bulk Access Regime (BAR) – the water sharing principles and rules that determine the amount of water available for extraction. A major component of the BAR for those rivers in the Murray-Darling Basin (MDB), is a Cap management strategy. The Cap management strategy ensures no net growth in water use over the long term.
The Murray-Darling Basin Cap
The Murray-Darling Basin Cap on Diversions (Cap) was introduced in 1995 to limit growth in water use across the Basin. In NSW, the Cap is based on the long-term average annual diversion that would occur with the level of irrigation development that existed in 1993/94. A Cap management strategy is yet to be developed or implemented on the Barwon-Darling. As a result, growth in irrigation development has continued since 1993/94. It is therefore highly probable that growth in water use has also occurred and that diversions on the Barwon-Darling exceed the Cap.
Developing a Cap management strategy for the Barwon-Darling
To get diversions back to Cap, water use on the Barwon-Darling will need to be reduced. The magnitude of this reduction will depend on the level of growth in water use that has occurred since 1993/94. If a significant level of growth has occurred and a substantial reduction in water use is required to achieve Cap, the socio-economic implications could be severe.
The Barwon-Darling River Management Committee (BDRMC) is responsible for developing a strategy to manage water use to Cap on the Barwon-Darling. To achieve this, the Committee must have access to, and confidence in, the data concerning irrigation development and water use. It is therefore vital that accurate data and an objective and transparent process are used.
This paper reports on the project that was undertaken by the NSW Departments of Agriculture (NSW Agriculture) and Land and Water Conservation (DLWC), to achieve the above.
The Barwon-Darling River history of irrigation development project
The primary objective of the project was to establish the level of irrigation development that existed in 1993/94 and each subsequent year until the present date. The data derived from the project would then be used to upgrade the DLWC’s Integrated Quantity and Quality model (IQQM), which would be run to determine the growth in water use and the level of cutback required to achieve Cap. The data would also be used to provide the basis for socio-economic assessment.
To ensure that the community would have confidence in the data, the project had to demonstrate objectivity and transparency, and be comprehensive. To achieve these three requirements, satellite imagery and a landholder-audited survey, were chosen as the key data collection tools. The process of data collection and verification by the irrigation community is illustrated in Figure 1.
Figure 1: Major steps in the project.
The irrigation industry on the Barwon-Darling River is predominantly summer-crop based, principally cotton. Summer images were therefore purchased. The preferred period was late January to late February, as this is when the cotton crop shows the highest reflectance, and therefore the greatest contrast with the surrounding landscape.
Landsat imagery was chosen for the majority of imagery. Landsat provided a balance between cost and resolution. SPOT (Le Systeme Pour l’Observation de la Terre) imagery was also used to enhance the accuracy of the digitising process (discussed later).
With the objective of determining the level of irrigation development that existed in 1993/94 and current, imagery for these years was required. It was also desirable to determine any trends in cropping practices leading up to these significant dates. Imagery was therefore also purchased for the 6 years prior (a balance between sufficient data and cost). In total, data for each summer from 1987/88 to 2000/01 were collected.
Data determined directly from the satellite imagery and GIS were; area developed, area cropped, and on farm storage area. In addition the satellite imagery was used to verify the presence or absence of development and cropping. It provided us with an independent assessment of the status of the development on each farm at the time each image was taken. This allowed us to verify construction times of on-farm storages and paddocks developed for irrigation.
The irrigator surveys were conducted using print outs of the satellite imagery for each year as a visual aid. Irrigators were asked to provide information in relation to summer crop types and subtypes, crop yields, winter crop details, irrigation methods, field slope, lasering dates, on farm storage (OFS) capacities, OFS management, pump types, pump capacities, pump installation dates, floodplain water harvesting, property ownership and service centre.
In general irrigators with large irrigation enterprises (greater than 50 ha of summer crops) were visited and interviewed in person. The smaller irrigators and sleeper licence holders were interviewed by telephone.
An important part of the landholder survey was the auditing and sign-off process. Each landholder was sent a document detailing the complete history of development and growth on their property, based on the survey results and satellite imagery. Each landholder was required to sign the document as a complete and accurate record of the development on their property and return it for collation and input into the model. This process served to instill confidence in the project team, as well as the irrigation community, as to the accuracy of the data to be used in the IQQM.
In addition to obtaining data about the areas developed for cropping and the on farm storage capacity it was essential to be able to determine the changes in management practice that are occurring as the industry develops. For example it has become evident that a large proportion of the increase in storage capacity has been installed in order to ensure a higher level of security for a similar crop area, not in order to increase the area of crop grown.
Without the comprehensive assessment of OFS capacity and crop areas across all seasons from the late 80’s to now that was facilitated by GIS and remote sensing it is likely these relationships would have been missed. This would result in inaccurate modelling of the use patterns and therefore incorrect treatment in relation to Cap.
Digitising the shapes and patches.
To determine areas accurately each land management unit was digitised on screen as a patch. This normally related to a single paddock on the property, however, when part of a paddock was managed differently, that area was separated from the remainder of the patch by dividing it. Subdividing and adding patches as required allowed the accuracy of the data to be maintained when only a portion of a patch was cropped or the development for that patch changed. For example the construction of an on farm water storage (OFS) on a portion of a patch originally developed for irrigating. Each patch was given a unique identifying number, which were used to link the patches to the season tables.
Enhancing the Accuracy of Digitising:
Digitising of the majority of patches was done using the 1998/99 satellite imagery. In order to enhance the accuracy of this digitising, we purchased a black and white SPOT image with 10m x 10m pixels and merged this with the coloured Landsat image of 30m pixels from a similar date. This gave us a colour image with 10m x 10m pixels improving the resolution of the image and thereby the accuracy of the digitising.
The satellite imagery was used to verify the presence or absence of development and cropping. It provided us with an independent assessment of the status of the development on each farm at the time each image was taken. This allowed us to verify construction times of on-farm storages, and paddocks developed for irrigation.
Data was collated using the Arcview GIS platform. The data was arranged in a series of dbf tables; one for each year that we collected the data for. We called these the Season Tables. The Season Tables included fields containing data about crop type/subtype, yield, irrigation method, slope, drainage, on farm storage capacity, property ownership, property name, and service centre. Figure 2 shows the simple data structure used to store the data in Arcview.
Entering data and summarising data over a number of different season tables can be very laborious, time consuming and prone to errors when done manually. Scripts, written in both Arcview and Excel, were used to facilitate the data entry, data checking and reporting process.
Figure 2: Schematic representation of the simple data structure used to store the data in Arcview:
The project undertaken for the Barwon-Darling River Management Committee, and the irrigation community on the Barwon-Darling River, relied heavily on satellite imagery and GIS technology. The use of these technologies provided not only high quality data, but gained the confidence of the community through the objectivity and transparency allowed by these technologies when coupled with other techniques such as survey and auditing.
Water management, particularly that which may involve reduced water access or allocation, is a controversial issue. The use of satellite imagery and GIS technology can help to alleviate controversy over accuracy of data. Involving the community in the validation of data collected by satellite imagery and GIS technologies is an important part of the process and should be considered when designing similar projects.