Non-destructive methods for estimation of vegetative biomass have been developed using several remote sensing strategies as well as physical measurement techniques. An effective method for estimating biomass must be at least as accurate as the accepted standard for destructive removal measurement techniques such as a forage harvester or quad harvest strategies. In large part vegetative biomass is considered a function of canopy or plant height. Subsequently, a method or piece of equipment which can estimate a height component is typically implemented for collecting measurements and from those measurements a relationship is created between height and mass. A number of sensing technologies have been examined for such applications. This study examined several types of ground-based sensing strategies for use in estimation of in-field forage biomass. A forage production trial consisting of multiple fertilizer treatments and mixed as well as mono culture species treatments was employed as an evaluation platform for the performance of the sensor estimation as compared to physical removal harvests. Predictive models were constructed and comparisons of sensor based estimates made to physically measured biomass harvested by hand from quad harvests as well as machine harvests using a forage harvester. Statistical analysis for both methods of harvest and sensor estimated harvests were performed as would normally be done according to treatment structure. Mean estimates were examined for evaluation of differences between biomass evaluation methods for each treatment. This strategy was employed in order to evaluate the difference if any on the overall research implications for data which was generated from physical collection techniques as well as sensor estimated data. Ultimately differences were minimal and did not contribute to disparity in implications for research aspects of the trial. Additionally, statistical analysis was performed on a subset of the data for repeatability. Paired identical plots were compared using Limits of agreement analysis to evaluate the repeatability of each technique. This analysis produced more narrow error bands for the sensing data as compared to the harvested data which suggests the sensing data is at least as stable as the physically harvested data. Subsequently, using ground-based mobile sensing for data collection which are incorporated into a biomass estimation model could prove to be effective in rapid accurate in field biomass estimation.