Abstract
The National Forestry Commission of Mexico continuously monitors forest
structure within the country’s continental territory by the
implementation of the National Forest and Soils Inventory (INFyS). Due
to the challenges involved in collecting data exclusively from field
surveys, there are spatial information gaps for important forest
attributes. This can produce bias or increase uncertainty when
generating estimates required to support forest management decisions.
Our objective is to predict the spatial distribution of tree height and
tree density in all Mexican forests. We performed wall-to-wall spatial
predictions of both attributes in 1-km grids, using ensemble machine
learning across each forest type in Mexico. Predictor variables include
remote sensing imagery and other geospatial data (e.g., vegetation
indexes, surface temperature). Training data is from the 2009-2014 cycle
(n>26,000 sampling plots). Spatial cross validation
suggested that the model had a better performance when predicting tree
height r2=0.4 [0.15,0.55] (mean [min, max])
than for tree density r2=0.2[0.10,0.31]. Maximum
values of tree height were for coniferous forests, coniferous-broadleaf
forests and cloud mountain forest (~36 m, 30 m and 21 m,
respectively). Tropical forests had maximum values of tree density
(~1370 trees/ha), followed by tropical dry forest (1006
trees/ha) and coniferous forest (988 trees/ha). Although most forests
had relatively low values of uncertainty, e.g., values <40%,
arid and semiarid ecosystems had high uncertainty in both tree height
and tree density predictions, e.g., values >60%. The
applied open science approach we present is easily replicable and
scalable, thus it is helpful to assist in the decision-making and future
of the National Forest and Soils Inventory. This work highlights the
need for technical capabilities aimed to use and resignify all the
effort done by the Mexican Forestry Commission in implementing the
INFyS.