**4.2 Methods**

278 Remote Sensing of Biomass – Principles and Applications

(Caldentey, 1995; Donoso et al., 2010; Garfias, 1994; Gayoso et al., 2002; Herrera and Waisberg, 2002; Sáez, 1991; Schlegel, 2001, Schmidt et al., 2009) (see Table 2). For native species, the biomass estimation methods most used in Chile are regression or allometry for the average tree (see section 3.0). In recent years, further attention has been paid in less expensive methods to assess AGB in native forests, based on remote sensing data and

In Chile, forest plantations are the main supply source of industrial raw material, with a total of near 40 million cubic meters. *Pinus radiata* D. Don is the main commercial species behind these massive commercial activities, with 80% of participation, followed by eucalyptus with 20% participation (CORMA, 2011). The plantations are concentrated in the central coastal area of Chile, between latitude 32 º S and 42 º S (Figure 2), covering a wide range of soils types (Schlatter, 1977; Schlatter and Gerding, 1984), environmental conditions (Fuenzalida, 1965; Madgwick, 1994) and silvicultural management regimes (Lewis and Ferguson, 1993; Lavery, 1986; Gerding, 1991). The land ownership structure is highly concentrated in two large forestry companies, which together owned 53% of the total planted area (CNE, GTZ / INFOR, 2008; Leyton, 2009), and the problem of quantifying wood supply has been primarily addressed by the private sector. Therefore, little public information exists on the area, location, age, species and management regime of plantations (CORMA, 2011). In contrast, the private sector has a large network of forest inventory information and has built growth simulators for the main species for different areas (i.e. *Radiata* and *Eucasim* models), types of site and management schemes and bucking (Fundación Chile, 2005; Morales et al., 1979). Some studies on the availability of wood for the forest industry (CORMA, 2011), carbon sequestration by plantations (Gilabert et al., 2010), AGB stocks for energy projects (INFOR, 2010, CNE / GTZ / INFOR, 2008) have been made by combining regional forest inventory data and these simulation

(Mg/ha) Author

2002

*N. dombeyi* <sup>285</sup>Herrera and Waisberg,

spatially explicit methods (Peña, 2007; Valdez et al., 2006).

Region Forest Type AGB

Table 2. Above ground biomass of different types of forest and locality.

<sup>X</sup>Mixed *N. obliqua*, *N. alpina* and

**4. Remote sensing based biomass estimation** 

**4.1 Description of the study areas and field data collection** 

VIII *N. alpina* second growth forest 104 Garfias, 1994 X Evergreen 194 to 663 Schlegel, 2001

XII *Nothofagus pumilio* 380 to 447 Caldentey, 1995

We consider four study areas, two predominantly plantations (pine and eucalyptus) and two covered with mostly second growth Chilean oaks: *Nothofagus obliqua*, *Nothofagus alpina*  and *Nothofagus dombeyi*. In both cases, one of the areas has a farm spatial scale and the other a municipality level (Table 3). Figure 2 shows the geographic location of the four areas of

**3.2 Plantation description** 

models.

Chile

study.
