Stump diameter (SD) has been rarely considered as an important tree parameter in forestry. It is until recently that SD has been found to be important predictor of tree diameter at breast height (D) and forest stand parameter such as volume and biomass. This study, developed D-SD relationships for nine different forest cover type in Tanzania mainland. A total of 32265 sample trees covering miombo woodlands, humid montane, lowland forests, bushlands, grasslands, mangroves, cultivated land, wetlands forests and plantations (Pines and Eucalyptus species) were used for fitting D-SD models. The findings revealed a linear relationship between D and SD for all forest covers. In addition, we found forest covers having similar D-SD allometry while others had unique D-SD allometry. This prompted fitting Generalised Linear Model where three forest cover groups were generated, i.e. group 1 (bushlands, woodlands, lowlands and grasslands); group 2 (mangroves, cultivated land, plantation and wetlands); and group 3 (humid montane). We fitted linear model to each forest cover group. Large variations in D were adequately explained by SD for each forest cover group. We further compared AGB values estimated from the measured D and estimated D from the D-SD equation. The estimated AGB from both approach did not differ significantly. We therefore, recommend the developed D-SD relationships models be applied to predict D of the missing trees for which their stumps still exist.