A cost and time-effective method to evaluate soil microbial respiration for soil quality assessment

Soil microbial respiration (SMR) is considered a robust estimate of potential soil biological activity. SMR measurement by incubation studies in laboratory conditions is a well-established technique, however is costly and time consuming depending on the selected incubation period. Based on 210 data available from Iran and Italy, we developed and validated (randomly selecting 70 and 30% of total data, respectively) a multiple linear regression analysis procedure for evaluating SMR. CumR (cumulative respiration during a 28 days incubation period) was evaluated from laboratory incubation measured at day 1 (R1d), day 2 (R2d) and day 4 (R4d), and CbasR (basal respiration at the last day of incubation, 28 days) from laboratory incubation at R4d. The best model prediction for CumR was found for the three parameters model (R1d, R2d and R4d; Adjusted R-2 = 0.90, RMSE = 46.5), and was still accurate in the two parameters model (R1d and R4d; Adjusted R-2 = 0.89, RMSE= 48.1) and in the one parameter model (R4d; Adjusted R-2 = 0.86, RMSE= 57.2). Model prediction for CbasR from R4d was acceptable (Adjusted R-2= 0.71, RMSE= 3.64). Since the main aim of the study was to provide a time and cost-effective method for deriving CumR and CbasR, we can speculate that model prediction based on one parameter model (R4d) is still sufficiently accurate. The models were validated with a large dataset of laboratory measures from Italy (1050 samples) and very close relationships were found between actual and predicted values for both CumR and CbasR. Therefore, the proposed models can be used for the application of composite indexes for soil quality assessment and recommended for further research with the same methodology using other datasets.