Landuse Change, Productivity & Development - Case studies
Final Report of Theme 5.1 to the National Land & Water Resources Audit
August 2001
5. Case studies
Summary
Four regions - Loddon and Corangomite in Victoria, Blackwood Catchment and Chapman River Catchment in Western Australia - were used to investigate the utility of different information in determining the scale of land use impacts on land degradation. Two of the regions - Loddon and Blackwood - suffer from dryland salinity problems; in addition the Loddon area also experiences irrigated salinity and significant peri-urban growth. Soil acidification was an issue in the Chapman River Catchment.
At the Statistical Division (SD) level the data shows the major economic drivers on land use patterns so that changes in the value of wool, dairy products and cereals for grain were followed by changes in the activity of the industries. However, interpretation of land use change from changes in value of agriculture production at the SD level is difficult. The Statistical Local Area (SLA) level information not only showed, at a smaller scale, the long term changes in wool, wheat, and dairy industries but also showed influences on productivity, intensification and diversification.
In Victoria information was also available at a parish level, which provided more information on the location of dairying and the switching between livestock and cropping land use. The groundtruthing of data suggests there there are numerous land use change drivers at work, which appear to push different sub-regions in different directions. Parish data may also indicate the emergence of new and supporting industries eg feed grain production for the dairy industry.
In Western Australia information was available from the ABS census at small areas (10km x 10km pixels). There, producer perceptions of salinity, which presumably reflect the situation on the ground, were that an increasing proportion of their holdings was being affected by salinity. However this has not yet shown up, in a reduced area used for higher value operations such as cropping although there was a reduced number of pixels with relatively high yields in the Blackwood Catchment and Chapman Catchment. Although wheat yields were increasing steadily in both Catchments some areas still had low water use efficiencies for wheat production, which have worrying implications for salinisation trends and raises concerns about the sustainability of cropping in those regions.
The case studies indicate that there can be many local factors affecting, and being affected by, land use. Land use changes due to more complex environmental and social factors require a range of datasets that are appropriate to the scale of the relevant factors to assist interpretation and to derive consistent narratives. Such consistent narratives will be the basis of responsive actions.
The broad overview that was presented at the national level necessarily involved simplifications in presentation. At the same time case studies were undertaken to assess the ability of the aggregated data to detect changes at local and regional scale. Case studies, besides helping to flesh out aggregate statistics, also allow an analysis of missing information. The case studies were chosen as areas where different land conditions may be affecting land use and productivity. They were, in Victoria, the irrigated Loddon-Campaspe (which has problems with salinity) and Goulburn Valley regions and the dryland Western District; and in Western Australia, the Blackwood Catchment (concerns about salinity) and the Greenough Catchment because of recent increases in productivity. The map in Figure 5.1 indicates the location of the case study regions and of the indicator SLAs for describing temporal changes.
Figure 5-1 Location of case study regions.
Victoria
Particular attention was given to trends shown in the concorded data that may have indicated a land use change. For example, two of the case study areas showed significant declines in the value of production from cropping. The case studies investigated whether this trend indicated a land use change. Examination of the case study areas showed that looking at a range of statistics created a consistent picture.
Data at Statistical Division level on land uses, farm finances and populations were provided in the Revaluing Rural Australia (RRA) tool. It should be noted that in 1994, Victoria had radical changes in boundaries to its local government areas subsequently reflected in its SLA boundaries. The ABS concordance used in the RRA was based upon redistributing data according to changes in population. This is appropriate for some factors, but is limited for agriculture, which has location specific effects, eg irrigation, and is not spatially uniform. Consequently, there are likely to be inaccuracies arising in this concordance. The utility of this data was checked by groundtruthing in a series of local focus groups.
Parish data were obtained for 1986 and 1996 to create a time series of volume (area and production) in the subject regions. Stock numbers and areas of land use were mapped to illustrate spatial changes in areas under cultivation and numbers of stock. To calculate change between 1986 and 1996, the areas of Crop, Pasture and Horticulture were divided by the total area of holdings for that year. The change indicated on the maps is therefore the change in the proportion of total holdings that comprises Crops, Pasture or Horticulture.
5.1 Loddon
Background
The Loddon-Campaspe Statistical Division (SD) is located in the central north of Victoria, covering 14,503 square kilometres and stretching from the Macedon Ranges outside Melbourne to just south of the River Murray. The Campaspe River comprises much of the eastern border. The Division encompasses the local government regions of Loddon, Maryborough, Mount Alexander, Greater Bendigo and Macedon Ranges whose populations total 157,320.
Within the Tragowel Plains, the land is almost flat, and bounded to the east by granite hills. This area includes the Pyramid Hill and Boort Irrigation Areas. In the south, the land rises gradually towards the Great Dividing Range. The soils are classified as Prior Stream Woodland (loam) and Treeless Plain soils (clay). Highly saline groundwater and poor surface drainage affect much of the area. The annual average rainfall is 300 to 600 mm per annum.
There are 2 315 establishments in the region with agricultural activity, averaging and area of 417 hectares. The major agricultural industries are Cereals, Milk, Pigs, Wool, Wheat and Lamb.
Land Use and Industry Change in the Loddon 1986 - 1996
The Loddon-Campaspe area, more than the Western Districts, suffers from severe land degradation problems in the form of both irrigated and dryland salinity as well as significant peri-urban growth.
The Loddon-Campaspe appears as an area in decline during this timeframe. Total value of agriculture was roughly constant from 1980 to 1990 at approximately $550M, followed by a fall of 20% to $443M in 1992, then a dramatic fall of 33% in 1996 to $286M. Feedback from interviews with local experts in the Loddon-Campaspe SD presented with the RRA source graphs was that the long term decline in wool is clear with its last big year in 1990 then, as wool prices halved in 1992, increases in sheep and lamb slaughtering from 1992 to 1995. The long term decline of wheat is notable, while the rise in cereals in 1996 could be attributed to good climatic conditions as well as an increased demand for feed grain from the major dairy industry to the north of the SD. However, short-term changes such as the shifts from livestock to cropping in 1993, back to livestock in 1995 and then back to cropping again in 1996 were not reflected in the data.
The parish data indicate sheep numbers (Figure 5-2) as a proportion of all stock have remained reasonably static across much of Loddon-Campaspe, with declines only occurring in the dairy area in the north and in the south east where a meat cattle industry exists. This indicates that sheep numbers have declined most in areas where an alternate agricultural industry already exists.
Figure 5-2 Changes in the value of the dairy and wool industries and numbers of stock in the Loddon-Campaspe SD between 1985 and 1996. (source ABS)
The area under cereal cropping (Figure 5-3) declined, with a corresponding increase in the area under pasture in the northern irrigated area as well as in patches in the southwest and east. The pasture area figures likewise show major declines in many parishes not matched by increases in areas of other agricultural activity.
Overall, long term trends in most commodities were not disputed although the fall in milk production in 1996 shown in Figure 5-2 was assumed to be a case of incorrect data.
A major initiative to increase the competitiveness of existing farm businesses in the region and to attract new businesses was the Loddon Murray 2000 Plus project funded under the Australian government’s Regional Partnership Program. A review (Barr 2000) found that the Loddon Murray Region is characterised by high rates of entry and exit from agriculture. Exit rates from agriculture are twice as high amongst small farms, yet even the rate of exit of larger farms is still above the national average. Exit rates seem uniform across the region, suggesting that exit is driven by factors other than differential farm economic performance. On the other hand, entries to agriculture on small farms are twice that of larger farms and clearly highest in horticultural block regions.
There were a large number of farms of sub economic viability in parts of the region. However, policies aimed at increasing exit rates from agriculture will have little impact upon the structure of the regions without matching programs which either increase the competitiveness of existing businesses or attract new businesses with significant capital backing. Transferable water entitlements (TWE) have had a significant impact on the structure of irrigated agriculture in the region. Permanent TWE is gradually shifting water from low value enterprises to high value enterprises. Large movements of water are occurring during years of low allocation, with water moving from mixed farms to dairy farms.
Figure 5-3 Changes in the proportion of area under crop and pasture in the Loddon-Campaspe SD between 1985 and 1996. (source ABS)
Loddon - North Shire
Land use within this shire shows (Figure 5-4) changes in area sown to broadacre crop to vary widely around the average of about 23 per cent of total agricultural area. (The data for 1994-95 year was too incomplete to include here.) The area of sown pastures appears to be in decline over the time frame of this study. Nevertheless in an area with a high proportion of land irrigated (about 25 percent) most - nearly 82 per cent - is under sown pastures. There is very little semi-intensive or horticultural crops. The intensity index appears moderate to low and moderately responsive to changes in sown pastures and crops. Productivity of the major commodities (as measured by total DSE/ha) shows fluctuations from year to year, with the droughts of 1982 and 1994 showing up in the data for 1982-83 and 1994-95.
Figure 5-4 Changes in land use classes, productivity and derived intensity index for the SLA of Loddon-North. (source ABS)
Summary
Loddon displays some trends, such as the declines in the value of wool and cereals for grain, and an inconsistent picture for milk. The groundtruthing and Parish data suggest there are numerous land use change drivers at work in this area, which appear to be pushing different regions within the SD in different directions. The diverse nature of land use in the Loddon SD may mean that signals from the value of production data are more difficult to interpret and therefore more difficult to link to land use change. Given that about a quarter of the area is irrigated, the overall intensification and productivity of this area is low, certainly not exceeding the levels in the Western District where there is negligible irrigation. Although it is possible that further intensification and productivity gains are restricted in this area by factors such as salinity problems or underlying structural problems, local opinion does not agree.
Clearly this area needs revitalising. One initiative is a combined DNRE/Commonwealth National Partnership Program, Loddon Murray 2000 Plus project, to address some of the needs for development.
5.2 Corangomite
Background
The Western District SD is located in the south western corner of Victoria, covering 23 367 square kilometres, with a population of 100 200. It encompasses the local government regions of Glenelg, Southern Grampians, Moyne, Corangamite and Warrnambool. Portland, in the south-west of the area, is the only deep water port between Melbourne and Adelaide.
The main geographical features of the region include a volcanic plain through the centre of the region containing many volcanic lakes and an impeded drainage system. The south of the region contains an upthrust of lower cretaceous origin, and hilly to steep topography. Highlands to the north are composed of weathered Ordovician sediments which have a hilly topography, and contain outcrops of Devonian granite which are significant in terms of salinity. The annual average rainfall is between 600 and 1 800 mm per annum, which is higher than Victoria’s average.
There are 5,150 establishments with agricultural activity in the region, averaging 339 hectares. Milk is the major agricultural industry, followed by wool, beef, lambs and pasture.
Land Use and Industry Change in the Western District SD 1986 - 1996
Total Value of Agriculture in the Western District SD decreased 12% from $709M in 1980 to $620M in 1985, then increased 23% to $767M in 1990 before dropping sharply by 19% in 1992 to $617M, with a gradual recovery of 24% over the next four years to 1996 to a peak of $768M.
Feedback from interviews with local experts in the Western District SD, when presented with RRA source graphs, indicated that the long term trend of increasing milk production was consistent as was the area of crops increasing since 1994 as farmers moved out of wool to grow fodder for the growing dairy sector. Sheep and lamb slaughterings react to wool prices, as farmers change strategies to maximise returns. So that as wool prices fell after its peak in 1990 to 1994, lamb prices fell to 1992 and rose to 1995 when there was a minor peak in 1995 wool prices which is reflected in the data.
At a parish level there are clear patterns in the changes in livestock industries (Figure 5-5). There is an increase in milk production, including in dairy cattle numbers, in the south-eastern corner at the expense of sheep numbers. Sheep numbers also declined in the western section of the SD where there was a move to meat cattle.
The area under cereal crops (Figure 5-6) also declined in the north-east of Western District SD, while pasture changes provided no discernible patterns.
Again, the data suggests that the impact of major industry changes show up in the SD data from ABS, but that changes due to more complex environmental and social factors are much harder to discern. This requires more detailed and localised information.
As an example, The South West Victorian Monitor Farm Project by DNRE identifies that grazing profitability in 1994-95 and 1995-96 was similar. This stability conceals a degree of variability depending upon farm enterprise mix. Prime lamb improved (this is indicated in the data) while wool (not indicated in the data) and beef farming fell. Factors underlying these changes include higher sheep and lamb prices, weakening of the wool market and reduced total costs. Earlier, the 1987/88 report noted the main reason for improved livestock performance was a 74% increase in wool price per kilogram when the Gross Value of Production (GVP) increased by 36% from $215M in 1985 to $293M in 1990.
In Western District as the GVP of wool declined, the GVP of other grazing industries appear to have increased, indicating a shift in industries (though probably not a major change in land use). Associated with the wool GVP decline is the increase in GVP of sheep and lambs slaughtered.
Figure 5-5 Changes in value of the dairy and wool industries and numbers of stock in the Western District SD between 1985 and 1996.
Figure 5-6 Changes in proportion of crop and pasture in the Western Districts SD during 1985 and 1996.
Corangomite - North Shire
This shire shows (Figure 5-7) some changes in the proportion of total area of agricultural holdings within the shire. (The data for 1994-95 year was too incomplete to include here.) Broadacre crops peaked in 1985 at nearly 15 per cent and then declined to a steady proportion of around 7 per cent. There are only small areas of irrigation, mostly on to pastures. The intensity index shows moderate levels and some changes as the crop and pasture areas changed. By contrast productivity of the major commodities (as shown by DSE/ha) appears to be generally increasing slightly with time. The effects of the droughts in 1982 and 1994 are very evident.
Figure 5-7 Changes in the land uses, productivity and derived intensity index for the SLA of Corangomite North. (source ABS)
Summary
The Western Districts showed some similarity with the pattern of changes to Loddon, with the major declines in the wool industry seem to be a dominant factor in this region. However, the decline in wool and cereal production is offset here by increases in the dairy industry and the value of livestock slaughtered. Parish data, supported by the groundtruthing, suggests a consolidation of the dairy industry within the region. Further detail indicates that the increases in cropping after 1993 is due to the emergence of feed grain production for the dairy industry.
Identifying the Drivers of Changing Land Use
During the groundtruthing exercise, participants were asked the question: “What compels you to change your enterprise mix?” The overwhelming response “Economics - the need to improve returns on investment.”
Derived from the basic premise that declining profits have forced a decision to change, other factors then operate. In this context, drivers observed by the project team and other work, include:
Financial: Availability of off farm income; Increasing/decreasing productivity; Water pricing policy (in irrigation regions).
Social: Population movements; New entrants into agriculture.
Environmental: New Technology; Climatic variability/reliability/risk; Resource condition.
Conclusions from Victorian case studies
The total value of agricultural commodities produced and the trend in 1996 indicate that the Western Districts is strongly performing, while Loddon-Campaspe has experienced decline in overall value of agricultural commodities. The growth in the Western District is being driven by increases in the value of both Livestock Products and Livestock Slaughtering. The Loddon-Campaspe SD appears to be declining in all commodity groupings.
The SLA and SD level information appeared to show the long term changes such as decline in wool, the long term decline of wheat, and increase in the major dairy industry. However, parish level data provided more information on the location of dairying and the switching between livestock and cropping land use. The case studies indicate that changes due to more complex environmental and social factors require a range of datasets to assist in interpreting changes in the value of agricultural production to derive a consistent narrative for land use changes.
Data at Statistical Division level on land uses, farm finances and populations were provided in the Revaluing Rural Australia (RRA) tool provided enough data to stimulate inquiry, but not enough information to be able to investigate the questions it provoked.
Interpretation of land use change from changes in value of agriculture production at the SD level is difficult. There are a variety of reasons why changes in value of production may not be due to actual changes in land use. However, the case studies appear to indicate that using a range of datasets can assist in interpreting changes in the value of agricultural production. Where a consistent narrative can be constructed around these several sets of data, changes in the value of production can indicate land use changes.
Western Australia
The case studies in Western Australia are for the Blackwood catchment in the south-west and Greenough catchment near Geraldton to the north of the areas of intensive agriculture in Western Australia.
As a result of a geocoding exercise undertaken by Agriculture Western Australia and the Australian Bureau of Statistics it was possible to break down AgStats for some years to a grid with cell size of approximately 10 km x 10 km.
A major feature of the timeframe has been development of broadacre cropping. In the Blackwood Catchment total area sown to crops has increased by 23 per cent from 382,527 ha in the 1984-85 to 470,196 ha in 1996-97. Likewise in Chapman Valley the area has increased by 35 per cent from 83,556 ha in the 1984-85 to 113,068 ha in 1996-97.
Two SLAs of comparable size were chosen to compare and contrast the catchments (Table 5-1). Although of comparable area and value of agriculture produced, there are large differences in the land use and number of farms operating in each.
Table 5-1 Comparison of two shires used to represent case studies in Western Australia (data from AgStats for 1997)
| Attribute | Boyup Brook | Mingenew |
|---|---|---|
| Total agricultural area (ha) | 189,645 | 193,852 |
| Number of agricultural holdings | 253 | 53 |
| Value of livestock slaughtered ($000s) | 12,275 | 2,250 |
| Value of livestock products ($000s) | 19,100 | 3007 |
| Value of crops ($000s) | 10,553 | 39,098 |
| Total value of agriculture ($000s) | 41,929 | 44,354 |
5.3 Blackwood Catchment
The Blackwood Catchment has been well studied because of extensive development of dryland salinity. The catchment is extensive and covers parts of many Statistical Local Areas (SLAs) or shires. They include from the west: Augusta-Margaret River; Nannup; Donnybrook-Balingup; Bridgetown; Boyup Brook; West Arthur; Williams; Kojonup; Woodanilling; Wagin; Narrogin; Broomehill; Katanning; Dumbleyung; Wickepin; Gnowangerup; Kent; and Kulin. A representative shire that is largely within the boundaries of the catchment is Boyup Brook.
Boyup Brook
One method of showing changes within a shire over time is to show the different land uses as a proportion of the total area of holdings recorded for each year within that shire. Figure 5-8 shows considerable variation in the proportions of broadacre crop and sown pastures, although most of the shire is likely to be timbered or naturalised pasture. (The data for 1994-95 year was too incomplete to include here.) There is little semi-intensive or horticultural crop within the shire and minimal irrigation. The intensity index is moderately high but appears to be declining in recent years. However, the productivity of the major commodities (as measured by DSE/ha) appears to be increasing slightly although the impact of the major drought in 1994 shows out.
Figure 5-8 Changes in broad land uses, productivity and intensity index for one Statistical Local Area (Boyup Brook) in the Blackwood Catchment during 1982-83 to 1996-97.
Small area detail across the Catchment
Comparing the two maps in Figure 5-9 indicates where cropping is most intense and where this has increased in the timeframe of 1984-85 to 1996-97. Cropping intensity as shown by the darkness of colour in the pixels increases from east to west in this catchment. The most intense areas of broadacre cropping within the catchment boundaries fall in the Statistical Local Areas (SLAs) of Dumbleyung, Wickepin, Katanning and the western part of Kent.
By 1996-97 the number of pixels with more than 2 500 hectares each in crop had increased from about 40 to about 56, often surrounding those pixels already with high intensity. In addition the intensity had increased considerably in the middle of the catchment (the shires of Wagin, Wodanilling, Kojonup and West Arthur).
Figure 5-9 Small area (10 km pixel) of broadacre cropping in the Blackwood Catchment in 1984-85 and 1996-97. (source AgWA and ABS)
Salinity as estimated by producers in their Agricultural Census returns indicate a considerable and widespread increase in salinity in the Blackwood Catchment when shown at small scale 10 km x 10 km cells (Figure 5-10). In the 1984 census 192 cells representing 28,311 hectares were reported with salinity as a problem and by 1993 this had increased to 216 cells representing 60,450 hectares.
Wheat productivity is shown as the percentage difference between the average yield (t/ha) for each cell and that for the whole SLA in which they are located. In 1984 season there were 26 cells across the Blackwood Catchment with yields better than 20 per cent greater than the SLA average and 61 cells achieving 5-20 per cent above the average. In the 1996 season there were 33 cells achieving yields 20 per cent above shire average and only 42 getting between 5 and 20 per cent of the average. There were some discernible patterns arising from a comparison of the two years. First, there are fewer cells in the western edge of the cropping part that had sown wheat in 1997 at a time when the total area of cropping was increasing. This could be due to the increasing salinity, although increased diversity in crop rotations is desirable. Secondly, the productivity of three sub-regions appear to have changed between the two years:
the northern middle region (the western part of Kulin SLA) has gone from a strong positive yields to average or below average yields;
•immediately to the west (Wickepin SLA) appears to have improved its relative yield over the same time and;
• in the centre of the Catchment (western Dumbleyung and eastern Wagin) there has also been an improvement in relative yields over the time.
These last changes are more difficult to ascribe to salinity since they may reflect avoiding saline areas and putting better management into the remaining areas.
Lime is the best ameliorant for acidity in soils, which may increase through widespread legume production or acidifying fertilisers. From 1995 to 1997 there was lime was applied in many more cells and often in higher amounts (84 cells representing 13 892 tonnes in 1995 up to 116 cells representing 22 251 tonnes in 1997). The heaviest amounts applied appear in the western portion (SLAs of Margaret River, Nannup, Donnybrook-Balingup, and Bridgetown) where grazing pastures and horticulture (particularly viticulture) are major land uses. However, the increase in area using lime was most noticeable in the eastern half of the Catchment where cropping is the dominant land use.
The water use efficiency by wheat crops was highest in the south (70 per cent) and east (50-70 per cent) of the Catchment. The relatively low efficiencies in the west (less than 40 per cent) are of concern because they imply that more salt is being moved to groundwater. In part it may be consequence of more out-of-growing season rainfall because the area has been achieving high (greater than 50 kg/ha/year) increases in wheat yield.
Figure 5-10 Comparison of relative wheat productivity with salinity and lime application in the Blackwood catchments at two times.
5.4 Greenough/Chapman catchment
At the northern end of the Western Australian sheep-wheat belt, the Chapman Valley is much smaller than the Blackwood Catchment. It is enclosed by the Statistical Local Areas (SLAs) or shires of: Northampton; Chapman Valley; and Greenough.
An adjacent shire of comparable size to Boyup Brook and similar distance in from the coast is Mingenew which is used here for indicative comparison.
Mingenew
Mingenew shire by contrast to Boyup Brook has a much higher proportion (50 per cent vs 15 per cent) of total area of holdings being used for broadacre crops (Figure 5-11). The area given to sown pastures is similar. (The data for 1994-95 year was too incomplete to include here.) Again there is considerable variation in the proportions of broadacre crops and sown pastures over the time frame of the study, while the intensity index, although moderately high, appears to be declining slightly. There are no semi-intensive or horticultural crops, or irrigation in this shire. Productivity, as calculated from the major commodities in terms of dry sheep equivalents per hectare (DSE/ha), appears to be have increased steadily in recent years. This is most likely through more efficient cropping. Land degradation is not noted to be a problem here, nor are there alternative uses being promoted.
Figure 5-11 Changes in the proportion of land uses, productivity and derived intensity index -value for Mingenew SLA during 1982-83 to 1996-97. (source ABS)
Small area detail across the Valley
The Chapman Valley (Figure 5-12) is a much smaller catchment than the Blackwood and covers part of only 3 shires. Cropping is distributed over the whole of this catchment. The number of pixels with more than 2 500 hectares of crop grown within the catchment increased from 6 in 1984-85 to 11 in 1996-97. At the same time there were several pixels that appeared to reduce the intensity of their cropping. This was part of a region wide change as outside the boundary of the catchment cropping appeared to increase its intensity to the north, west and south.
Figure 5-12 Small area (10 km pixel) of broadacre cropping in the Chapman Valley in 1984-85 and 1996-97. (source AgWA and ABS)
Salinity as estimated by producers in their Agricultural Census returns indicate some increase in salinity in the Blackwood Catchment when shown at small scale 10 km x 10 km cells (Figure 5-13). In the 1984 census 13 cells representing 777 hectares were reported with salinity as a problem and by 1993 this had increased to 18 cells representing 1 598 hectares. Wheat productivity is shown as the percentage difference between the average yield (t/ha) for each cell and that for the whole SLA in which they are located. In 1984 season there were 8 cells across the Chapman Valley with yields better than 20 per cent greater than the SLA average and 7 cells achieving 5-20 per cent above the average. In the 1996 season there were only 3 cells achieving yields 20 per cent above shire average but still 7 cells getting between 5 and 20 per cent of the average. There did not appear to be a clear pattern in the changes beyond a general reduction in the number of cells that deviated both above and below the average.
This area has a low variability (coefficient of variability less than 0.1) in wheat yields, but high (greater than 60 kg/ha/year) gains in wheat (see Figures 4-4 and 4-5). Considerable lupin production in the area of about one third of total broadacre crop in any one year, could be contributing to acidification. There was an increase in lime application from 1995 (8 cells representing 2,263 tonnes) to 1997 (10 cells representing 5,352 tonnes) and all appeared to be applied to cropping areas.
Figure 5-13 Comparison of relative wheat productivity with salinity and lime application in the Greenough/Chapman Valley catchment at two times.
Water use efficiency was low (30-39 per cent of potential) in the SLA of Chapman Valley but was higher in Northampton (40-49) and Greenough (50-59 per cent).
Conclusions from Western Australian case studies
Although the sample times for the small area data (shown at 10km x 10 km cells) for salinity, lime application and wheat yield productivity do not precisely coincide, there are some conclusions that can be drawn. At the small scale, producer perceptions of salinity, which presumably reflect the situation on the ground, were that an increasing proportion of their holdings was being affected by salinity. One would expect this to show up in a reduced area used for higher value operations such as cropping. This did not appear to occur, although there was a reduced number of pixels with relatively high yields in the Blackwood Catchment and Chapman Catchment.
In the Chapman Catchment, where a high proportion of wheat is rotated with lupins each year, there was evidence that lime applications had increased to counter acidification and in response to extension programs highlighting the problem. In the Blackwood Catchment most lime appeared to be applied either to pastures or to higher value horticultural crops. Responses to increasing acidity and salinity appear to be showing up crop areas and yields at small areas, smaller than SLA, but shown up at 10km x 10 km cells.
At the SLA level wheat yields were increasing steadily in both Catchments. However, some areas had low water use efficiencies for wheat production, which have worrying implications for salinisation trends. Although some of this may be an artefact of the method of calculation by ignoring use by other means eg pasture and weeds, it does raise concerns about the sustainability of cropping in those regions.
As salinity and acidity show up at small scales, the changes in a minor or early development (ie affecting less than 25 per cent of the holdings in an area) will be effectively diluted when aggregated with hundreds of unaffected responses at the SLA level.