Sources of error that have been identified on the gamma-ray soil density gauge are: (1) sensitivity to local density variations, especially at the soil surface, (2) sensitivity to soil composition, and (3) inaccurate calibration techniques. An attempt was made to solve these problems by developing for each gauge calibration models that enable stable and homogenous non-soil standards to be used for calibration. Calibration standards and calibration models for each gauge were found and tested. In the process of developing and testing the new nuclear gauge calibration model, the dual-gauge principal of compensating these gauges for variations in soil composition was discovered. The air-gap method of using this principal is explained and was optimized by using the calibration model developed in the current study.