Correlating Soil Microbial Properties with Crop Yields in the Canadian Prairies: Two Case Studies.

Soil microorganisms mediate many important biological processes for sustainable agriculture.  However, correlations between soil microbial properties and crop productivity cannot always be demonstrated.  Two case studies, one with wheat (Triticum aestivum L.) and another with canola (Brassica napus L.), were used to relate soil microbial properties to crop yields in the Canadian prairies.  The wheat data were collected from an irrigated cropping systems study, conducted to compare the effects of rotation length (3-, 4-yr) under conventional (CONV) or conservation (CONS) soil management, and 5- and 6-yr CONS soil management rotations on the productivity of spring wheat, dry bean (Phaseolus vulgaris L.), potato (Solanum tuberosum L.), and sugar beet (Beta vulgaris L.).  Wheat monoculture (1-yr rotation) under CONV management was also included.  CONS management included reduced tillage, application of compost and cover-cropping.  Soil microbial biomass C (MBC) and bacterial diversity (H’) were determined from 2002 to 2011 in the wheat phase of each rotation, and these properties were related to wheat yields.  Most of the microbiological properties were greater under CONS management than CONV management, and greater in longer rotations than in shorter rotations.  Over the 10-yr period, MBC in wheat rhizosphere and bulk soil were positively correlated with wheat yield (r = 0.41** and 0.39**, respectively).  Bacterial diversity did not correlate with wheat yields.  The canola study was conducted at seven sites in two years to determine if maximum canola yields would be attained by applying higher-than-normal input levels.  The treatments consisted of two canola seeding rates (75 or 150 seeds m^-2), two nitrogen rates (1x and 1.5x soil test recommendation) and three nitrogen form-fungicide (prothioconazole) combinations (uncoated urea-no fungicide, uncoated urea + fungicide, and 50% polymer-coated urea + fungicide) in a 2 x 2 x 3 factorial arrangement.  Soil MBC, β-glucosidase enzyme activity and H’ were determined in canola rhizosphere and in bulk soil and related to canola yields.  The effects of seeding rate, N rate and N form on soil microbial biomass, enzyme activity or bacterial diversity were usually not significant.  When effects were significant, doubling the seeding rate from 75 to 150 seeds m^-2 usually increased these microbial characteristics in canola rhizosphere or bulk soil.  Increasing the rate of N application to 1.5x the recommended rate had mostly positive effects in canola rhizosphere and negative effects in bulk soil.  The effects of N form (including addition of fungicide) were inconsistent.  Soil MBC and β-glucosidase enzyme activity correlated positively with canola grain yield at the five sites where yields were less than 4000 kg ha^-1 (r = 0.51** to 0.76**), but no or negative correlations were observed at the two sites with more than 4000 kg ha^-1 yields.  The diversity of soil bacteria was mostly not correlated with grain yields.  Some of these relationships could be explained by canola root damage by root maggots because root damage was usually negatively correlated with the soil microbial characteristics.  These results show that crop/soil management practices that enhance soil microbial communities and their activities will also enhance crop growth, health and yields.