The data used to estimate the carbon stock changes in this study were obtained from measurements in two subsequent years in a chronosequence. As a result, we could not analyze the possible year to year variation in growth rates, recruitment and mortality within each age class. This question will be resolved once we have more data available in subsequent years.
Another limitation is we could not obtain the secondary forests plots of more than 35 years. The oldest secondary forests considered in the study were of 35 years after abandonment of slash and burn agriculture. It was difficult obtain the sampling plots older than 35 years as the current human settlements in the study area has a history of about 30–40 years. Therefore, we could not have more field information about the changes that can occur in older phases secondary forest succession before they get to mature forests. The trajectories of carbon accumulations were derived from regression models based on the information obtained from secondary forests of up to 35 years and mature forests. The establishment sampling plots for 35 years age only in the second measurement year precluded to calculate the growth, recruitment and for this age class. The changes that occur in this age class can be obtained by the subsequent measurement.