Resumo. The gallery forests have great genetic diversity and important ecological functions, such as protecting the headwaters, controlling erosion, and functioning as buffer zones and filtering chemicals. Litterfall, which can be used as an indicator of ecological productivity, is widely collected using litter traps that are randomly distributed in a forest plot. However, vegetation distribution may present spatial dependence, thus the yield of the litterfall can be mapped using geostatistical techniques allowing the delineation of management zones. Therefore, the objective of this study was to evaluate the efficiency of geostatistical methods using the zoning management productivity of litterfall on a gallery forest in central Brazil. The study was conducted on the gallery forest along the ‘Lava-pés’ stream in Goiás State, Brazil where the experimental site (3 ha) was structured in a grid of 60 litterfall traps, each of 0.33 m 2 held 0.65 m above the ground, georeferenced, spaced at 32 x 32 m intervals. Litterfall was monthly collected from December 2011 to November 2012. All litterfall samples were manually separated into three fractions: leaves (LE), branch bark (BB), and reproductive parts (RP) and they were expressed in kg ha -1 . Statistical analyses consisted of data description and geostatistics. The litterfall of for LE and total showed strong spatial dependence. The BB and RP showed pure nugget effect. The total litterfall maps obtained by the Kriging interpolation method indicated zones in the map ranging from 900 to 10,900 kg ha -1 per yr. The kriging interpolation technique delineate management zones of productivity in the gallery forest litterfall studied, which allowed the specific forest management of litterfall.