This research addresses the reoccurrence of forest fires and their size with regional-spatial information. This study is attained to the Sustainable Development Goal in the year 2030 (climate action and life on land) and is consistent with the mission of JISDeP. Probit and tobit regression analyses were applied to the regional-spatial panel data from 2015 to 2018 in Indonesia with the observations of forest-fire events, peatland, forest area, and timber concession on an annual basis. Such analyses would characterize the possible determinants for the forest fire reoccurrence together with their sizes. Comparatively, this study tries to fill the gap by examining the reoccurrence of forest fires. This research tries to fill in the gap on studies about land and forest fires by combining quantitative analysis using probit and tobit regression and using spatial approach of peatland, forest and timber concession area. The regression results reveal the following outcomes. The first outcome is whether forest fire reoccurrence positively (negatively) associated with peatland and forest areas (timber concession). Second, forest fires tend to decrease with the repetition of past forest fires but increase with timber concession, peatland, and forest areas. Overall, these results imply that the reoccurrence of forest fires and their sizes is highly concerned with timber concession and types of areas, suggesting that Indonesia should organize the policies regarding forest timber concession and areas to reduce forest fires and the associated damage.