A key influence on farming practice in the region is rainfall and its distribution. For the purposes of interpreting results from the project, spatial patterns are also important.

Method 

We downloaded Bureau of Meteorology monthly rainfall data for the available Wimmera/Mallee stations (Figure 1, from http://www.bom.gov.au/climate/data/weather-data.shtml) and kriged monthly data to a 0.1 x 0.1 degree grid (global variogram using VESPER 1.6, Australian Centre for Precision Agriculture 2004). The monthly data was averaged across years and used to produce the following graphs on Google Earth. Focus paddock locations are given in Figure 2.

Figure 1. Bureau of Meteorology rainfall stations in the Mallee and Wimmera with data at least until 2004 (most go to 2009). The average start year is 1916. M, B and H indicate Mildura, Birchip and Horsham.

Figure 2. Location of the focus paddock pairs (Google Earth).

Growing season and summer rainfall

Rainfall, particularly in the growing season, has reduced dramatically. Historically (1990-1995, Figure 3) most paddocks (for names, Figure 2) have had April-October rainfall between 220 and 290 mm. In the ten years between 1996 and 2005 (Figure 4), the average April-October rainfall for the paddocks dropped to 200mm for the northern 8 paddocks, and 250mm for the southern 4 paddocks. The 2006-2008 April-October rainfall (Figure 5) has been 130-170mm.

The pattern of April-October rainfall has remainedessentially the same. A dry patch over the Tyrrell basin (Manangatang and Lake Tyrrell) has remained. There is some evidence for the areas around Yaapeet and Birchip becoming drier than surrounding areas to the east and west.

The range in summer rainfall (November-March) across the area is less, for the focus paddocks between 120 and 140mm. The main change in recent years is for rainfall to be higher in the east, rather than in the south. For the focus paddocks this has meant the Yaapeet and Patchewollock paddocks (to the west) have received less rain over summer than the eastern four pairs of paddocks.

Figure 3. Average total rainfall in the growing season, 1900-1995. Flexible farming paddocks are outlined in red.

Figure 4. Average total rainfall in the growing season, 1996-2005. Flexible farming paddocks are outlined in red.

Figure 5. Average total rainfall in the growing season, 2006-2008. Flexible farming paddocks are outlined in red.

Two-monthly rainfall

Climate change modelling for the Wimmera and Mallee (Department of Sustainability and Environment 2008) forecasts drying in autumn, winter and particularly spring. The two-monthly rainfall data shows drying north and west of Mildura, and east of Birchip in January-February for 2006-2008, and some drying in the northwest in the preceding ten years (Figure 6). January-February rainfall has actually been wetter since 1996 than the period beforehand.

March-April rainfall was dramatically drier between 1996 and 2005, compared to both historic rainfall data and the last three years (Figure 7). The recent reliability of March-April rainfall, particularly along the Mallee highway west of Ouyen, has underpinned dry sowing strategies in the past three years.

Figure 6. Average January-February rainfall (sum of both months), across the years 1900-1995.

Figure 7. Average March-April rainfall (sum of both months), across the years 1900-1995.

May-June rainfall has historically been quite even across the Mallee and Wimmera (Figure8). Since 1996, and especially in the last three years, there has been a drying trend from the north. July-August rainfall has changed little in the northern Wimmera and Mallee until 2006, since when there has also been drying from the north (Figure9). Rainfall in the central Wimmera has fallen considerably since 1996.  Dryness around Birchip and Yaapeet stands out in these four months. Despite the fall in rainfall, there have still been reasonable amounts between May and August (at least 0.5mm/day), enough to sustain a crop.

Figure 8. Average May-June rainfall (sum of both months), across the years 1900-1995.

Figure 9. Average July-August rainfall (sum of both months), across the years 1900-1995.

The dryness of September-October 2006-2008 contrasts dramatically with 1996-2005, which is little different to 1990-1995 (Figure 10). The paddocks have received 25-35mm less than average during these months. That this is a relatively recent change gives some hope that it may not persist into the future!

November-December rainfall has increased, both since 1996 and 2006 (Figure 11). The pattern of increase particularly to the east is quite evident.

Figure 10. Average September-October rainfall (sum of both months), across the years 1900-1995.

Figure 11. Average November-December rainfall (sum of both months), across the years 1900-1995.

Conclusion

According to rainfall records, the period of the project (2006-2008) has been relatively unusual for the farmers involved. Farmers have had more reliable March-April rainfall than they were used to in the preceding decade, but suffered much less reliable September-October.  May-August has been on a drying trend since 1996 and farmers may have been expecting this.