This paper provides an overview of an Australian national database established as a repository for statistically valid nitrogen, phosphorus, potassium, and sulfur crop response experiments with accompanying soil test data conducted for cereal, oilseed, and pulse crops since the late 1950s. Soil test calibration relationships can be derived using an online interrogation tool which utilises a series of queries (e.g. soil texture or pH) to define variability in crop responsiveness. The calibration data derived from the tool will be used to underpin soil test interpretations as part of fertiliser recommendations made to grain producers.

Historic field experiment data for Australian broad-acre crops have been collated within an online database that can be interrogated through a web browser interface. Interrogations produce calibrated critical soil test values that represent soil N, P, K, and S levels expected to generate grain yields at 80, 90, and 95% of full yield potential. Interrogation criteria include crop type, soil type, geographic region, and a range of other experiment metadata that may influence the level of available soil nutrients required to reach those grain yields.

A national database of Australian field experiments was used to examine the relationships between relative grain yields of wheat or barley and available soil nitrogen (N) determined before or at sowing (plant-available N measured as nitrate-N and mineral-N as kg N/ha or mg N/kg and other general indicators of soil fertility status or of mineralisable N). Crop yield potential (Y_max) was clearly the dominant factor affecting the critical soil value for 90% of maximum yield (CV90), with no detectable changes to CV90 in response to differences in grain protein. Management factors affected plant-available N but not CV90. The over-riding influence of Y_max on crop N demand, CV90, and N fertiliser requirement suggests improved guidelines for in-season N management are required in regions where recovery of topdressed N by the crop is reliable.

A comparison of soil phosphorus (P) tests (viz. Colwell-P, BSES-P, Olsen-P, DGT-P, CaCl₂-P, and Mehlich 3-P) was undertaken using 164 historical P fertiliser response experiments with wheat (Triticum aestivum) grown on a wide range of soil types in New South Wales, Victoria, and South Australia. Soil P tests, except Mehlich 3-P, showed moderate calibration relationships (R-values ≥0.43, P < 0.001) with relative grain yield. Critical soil test values for all soil P tests were higher for Calcarosols than for other soil types. Colwell-P critical values and confidence intervals (CI70 and CI95) for calibration relationships derived for different P buffer capacity classes were consistent with published data.

A national database of Australian field experiments was used to examine the relationships between soil P concentrations (Colwell NaHCO₃ extraction) and relative grain yields of grain legume, oilseed, and summer grain crops. Data were limited for several species, so similar relationships derived for the more extensively research winter cereals in the same regional/local production systems were used to benchmark relative P responses. The analyses highlighted the importance of P in subsurface layers as a modifier of crop yield potential and soil P concentrations below which fertiliser responses are indicated, while the benchmarking provided an insight into relative P responsiveness of key species grown in regional cropping systems.

Potassium (K) is an essential plant nutrient required in large quantities, second to nitrogen, by crop and pasture plants. The bicarbonate-extractable K (Colwell soil-test K), the most extensively used soil test, are frequently low for Australian soils. Data are available for several crop species grown on a range of soil types. Critical Colwell STK concentrations for wheat varied from 40 to 64 mg K/kg soil, for canola ranged from 43 to 47 mg K/kg, and in lupin was ~25 mg K/kg.

Soil testing is used to measure plant-available soil sulfur (S) status. Sampling to a depth of 30 cm improves the ability of the KCl-40 soil S test to measure soil S availability for crops compared with sampling to a depth of 10 cm. KCl-40 S critical soil test values and ranges (0–30 cm) were 2.8 (2.4–3.2) mg kg^–1 for wheat and 6.8 (6.4–7.2) mg kg^–1 for canola grown in Western Australia.

This paper reports on the extension process undertaken as part of the Making Better Fertiliser Decisions for Cropping (BFDC) project and provides a perspective on how industry-wide change could be approached within a specific R&D area. The Australian grains and fertiliser industries are committed to the uptake of soil test calibrations derived from BFDC, yet a difficulty is always the facilitation of extension activities and the uptake of the database by individuals. To address this challenge we discuss the steps taken to widen our audience by facilitating alternative methods of user access to the BFDC National database.

The new Better Fertiliser Decisions for Crops (BFDC) National Database and BFDC Interrogator provide the user with the ability to make specific searches for soil test–relative yield responses for N, P, K, and S on defined soil types and geographical regions, and to apply filters for factors such as absolute yield, subsoil nutrients, and pH where this latter information is available. The influence of specific management and environmental factors on nutrient response is best assessed from specific studies due to the paucity of metadata for most historic experiments. The soil test corresponding with a 90% relative yield for a crop is best described by a critical range rather than a specific value.