A new method for the estimation of biomass yield of giant miscanthus (Miscanthus giganteus) in the course of vegetation

Abstract The paper presents an innovative approach to the estimation of biomass of giant miscanthus ( Miscanthus giganteus ) at any point in time during the period of its vegetation. The innovative aspect consists in a simplification of required input data, as the model of biomass increase was developed on the basis of simple biometric measurements of such parameters as shoot length, diameter and mass. The study was based on field measurements conducted on plants in the period of 2012–2014 ( 3 rd - 5 th year of cultivation), at varied weather conditions. The field experiment was performed on a cultivation conducted with the extensive method. Based on the developed descriptive statistics of biometric traits of miscanthus, the correlation of two features of miscanthus shoot volume index and shoot mass was determined statistically correctly. After transformation, it was defined as a mixture of two normal distributions independent from the age of the plantation and from the time of measurement during the vegetation period. The number of shoots per plant was estimated by shifted Pascal distribution. In the next stage of the study the material acquired from the field measurements and performed calculations allowed to develop the final formula of a simple model for the estimation of current biomass of giant miscanthus. Knowledge of the structure of plant density on a plantation permits the estimation of the current amount of miscanthus biomass on its area. The accuracy of estimation of current biomass of miscanthus, analysed in the study, depends on the number of shoots on which biometric measurements are performed. The most rational approach is to make the measurements on 10 shoots of miscanthus. Any further increase of the number of shoots subjected to measurement does not provide any significant increase of accuracy of estimation of biomass on the plantation. The results obtained allow very good estimation, with as much as 80% probability, of the biomass of the plant, with relative error of estimation not exceeding 17%. An important aspect of the constructed model is the fact that it is based on measured simple biometric features of the plant and does not require any estimates of the weather conditions during the vegetation.