Dahlen Long-term Nitrogen and Phosphorus Recovery Experiment

Since establishment, crops planted included barley (1996, 2000, 2004, 2008), chickpeas (1997), canola (1998, 2002, 2006), wheat (1999, 2003, 2007), and lentils (2001, 2005). Canola was chosen as the 2010 crop, following lentils in 2009, and barley in 2008.

With good seasonal conditions, canola showed a strong response to P with the first 9 kg P/ha doubling yields and the next 18 kg P/ha plots raising yields to over 3 t/ha. At the highest P rate (36 kg P/ha), establishment was poor, probably due to fertilizer damage, and yields were somewhat lower. The response to N showed a 17% yield increase to 40 to 80 kg N/ha, and split N applications showed a small benefit over applying all the N at sowing. The small N response is probably due to N₂-fixation from the lentil crop in 2009 and possibly carryover from a failed barley crop in 2008 due to dry conditions. When first established in 1996, the site had a soil test P (Colwell) of around 22 mg/kg and a moderate to low P buffering index (i.e., a moderate to low capacity to “bind” plant available soil P into unavailable forms). Since 1996, there have been 15 crops sown and 12 harvested, with crop failures in 2002 (wheat), 2006 (canola), and 2008 (barley). At the end of the 2006 canola crop, a nutrient audit was undertaken at the site, with the inputs of P from fertilizer balanced against the P taken away in the grain. Use of 9 kg P/ha/year was in approximate P balance, with a slight surplus of 4 kg P/ha/year. This surplus appears as a rise in soil P test value from 22 to 33 mg/kg over the 15 years.

The results indicate that over the relatively low yield decade of the 2000s, the use of 9 kg P/ha approximately balanced the grain P removal and raised soil test levels to a desirable value from what was a low value. That fertilizer strategy matched input with output and a separate economic analysis of this site reported some time ago showed that 9 kg P/ha also provided the best returns. ANZ-08