**5. Conclusions**

288 Remote Sensing of Biomass – Principles and Applications

Fig. 7. Cokriging results for Pantanillos area. Above: Original estimate (left) and processed (right) biomass (ton/ha). Bottom: Variance of estimation error for AGB (ton2/ha2), original

In order to apply this approach in the estimation of AGB in any area the following steps

a. Selection of the geographic area: It is first necessary to define the limits of the study area and the stands of interest. One may not be able to divide into stands if one does not have enough previous information and the results can be aggregated in the proper

b. Gathering and processing satellite images: The images used should be subject to the

(left) and processed (right).

spatial division later (see step f).

following treatments:

should be done:

The use of covariates extracted from SRTM elevation model and LANDSAT images allows for estimates with a greater level of detail than those obtained by using only data field. This can be corroborated by comparing the estimates by cokriging with those that would be obtained with univariate kriging. We hypothesize that the set of covariates accounts, through their spatial dependence, for two fundamental aspects to explain existing biomass:


Geostatistical Estimation of Biomass Stock in Chilean Native Forests and Plantations 291

Bitterlich, W. (1984). The Relascope Idea. Relative Measurements in Forestry.

Caldentey, J. (1989). Beziehungen zwischen Klimaelementen und der Produktivitat von

Caldentey, J. (1995). Acumulación de biomasa en rodales naturales de *Nothofagus pumilio* en

Chavez, P. (1996). Image-based atmospheric corrections - revisited and improved.

Chilès, J.P. & Delfiner, P. (1999). *Geostatistics: Modeling Spatial Uncertainty*, Wiley, New York. CNE/GTZ/INFOR. (2008). Disponibilidad de residuos madereros. Residuos de la industria

Cohen, W. & Spies, T. (1992). Estimating structural attributes of Douglas-fir/ western

CONAF-CONAMA-BIRF. (1999). Catastro y evaluación de recursos vegetacionales nativos de Chile. Informe nacional con variables ambientales. Santiago. Chile. 90 pp

Coulston, J. (2008). Forest Inventory and Stratified Estimation: A Cautionary Note. Research

Curran, P.J.; Dungan, J. & Gholz, H. (1992). Seasonal LAI in slash pine estimated with

David, M. (1976). The Practice of Kriging, In: *Advanced Geostatistics in the Mining Industry*, M. Guarascio, M. David & C.J. Huijbregts, (Eds.), 31-48, Reidel, Dordrecht Donoso, C. (1981). Tipos Forestales de los Bosques Nativos de Chile. Documento de Trabajo

Donoso, C., (1993). Bosques templados de Chile y Argentina. Variación estructura y dinámica. Ecología forestal. Ed. Universitaria. Santiago, Chile 484 pp Donoso, S.; Peña-Rojas, K.; Delgado-Flores, C.; Riquelme, A. & Paratori, A. (2010). Above-

Eklundh, L.; Harrie, L. & Kuusk, A. (2001). Investigating relationships between LANDSAT

Fazakas, Z.; Nilsson, M. & Olsson, H. (1999). Regional forest biomass and wood volume

FIA. (2001). Bosque nativo en Chile: situación actual y perspectivas. Fundación para la

CORMA, (February 2011). Available from http://www.corma.cl/corma\_info.asp

LANDSAT TM. *Remote Sensing of Environment*, 39, 3–13

in central Chile. Interciencia, 35(11):65-69

*Pinus radiata* Plantagen in Chile. Inaugural- Dissertation zur Erlangung der Doktorwurde der Forestwissenschaftlichen Fakultat der Ludwig-Maximilians-

Tierra del Fuego, Chile. Investigación Agraria: Sistemas y Recursos Forestales 4(2):

*Photogrammetric Engineering & Remote Sensing*, Vol. 62, No. 9 (September 1996), pp.

primaria de la madera – Disponibilidad para uso energético. Santiago, octubre

hemlock forest stands from LANDSAT and Spot imagery. *Remote Sensing of* 

Note SRS–16. Southern Research Station Forest Service. Department of Agriculture.

N°. 38. Investigación y Desarrollo Forestal (CONAF, PNUD-FAO) (Publicación

ground biomass accumulation and growth in marginal *Nothofagus macrocarpa* forest

ETM+ sensor data and leaf area index in a boreal conifer forest. *Remote Sensing of* 

estimation using satellite data and ancillary data. *Agricultural and Forest* 

Innovación Agraria. Ministerio de Agricultura. ISBN 956-7874-14-X, Santiago,

Commonwealth Agricultural Bureaux. ISBN 0-85198-539-4. 241 pp

Universitat Munchen. 131 pp

pp. 165-175

1025-1036

2007.54 pp

*Environment*, 41 (1), 1–17

United States. 6 pp

FAO Chile) 70 pp

*Environment*, 78, 239–251

Chile. 111pp

*Meteorology*, 98–99, pp. 417–425

The approach to this scale of work, using medium spatial resolution sensors, is to estimate AGB independently of species or plant community types and then overlap stands boundaries, or any other administrative division, to obtain the associated total AGB stocks. Coregionalization models (direct and cross variograms) are simplifications of reality; in particular, they may not detect anisotropies when having a too small amount of field data, allowing only for an omnidirectional inference. The number of field data should be such that the inference of AGB variogram is feasible, which is usually achieved with more than 100 data in each stratum. In order for this number to be reduced, we recommend the sampling of the area (site) to be as regular as possible, i.e. avoiding samples clustered in space.

In the future, we suggest identifying other covariates correlated with AGB, for example, an indicator of the amount of sunlight available or a local indicator of the ground geometry (concave, convex or plane) that is related to the availability of water. These indicators could be calculated from a digital elevation model, already available at several spatial resolutions.
