Abstract Knowledge of biomass properties and/or fuel characteristics is the basis for bioenergy or chemical conversion processes. Biomass properties are variable and associated with plant species and assortments, therefore, identifying the associations will be important for understanding mechanisms behind the variations and for improvements in predicting biofuel characteristics and management of sustainable feedstock production. A data synthesis was carried out based on data of energy properties including contents of carbon (C), hydrogen (H), oxygen (O), nitrogen (N), sulfur (S) and ash (A) (CHONSA) in variable biomass species and assortments. Data were collected from 260 peer-reviewed literatures and composed of 742 data objects of more than 144 species. Principal component analysis (PCA) was used to analyze the data. The analyses were carried out at five levels of plant classifications, respectively: levels 1–3 for life-forms, level 4 for assortments and level 5 for species. All the loading plots produced for the energy properties showed the following groupings: C, H and O; ash; N and S. The score plots showed a large natural variation between life-forms for herbaceous materials but more homogeneous for woody ones. Thus, an ordering of energy properties for variable biomass was achieved and correlations between biomass properties and species groups were shown. The general trend was: wood→herbaceous dicots+C4 graminoids→C3 graminoids along the PCA axis from CH rich to ash rich, which is consistent with the empirical knowledge that wood is a high quality but grass a low quality fuel during a combustion. A robust analysis confirmed the findings by PCA. The analyses in this study indicated that the C, H, and O contents are much less variable than those of N, S and total ash. The ash content had the largest role in differentiating biomass energy properties between species and between assortments.