monitoring of forest biomass, which will ultimately result in
the production of accurate and comprehensive information
for use in forest management.
Conclusions
This study proposed the use of a
MVRVR
model for
AGB
estima-
tion. In addition, to validate the use of multi-temporal
ALOS
PALSAR
FBD
data, a comparison of
MLR
,
MLPNN
,
SVR
, and
MVRVR
models was performed.
AGB
values were obtained from forest
inventories which were in the range of 11 to 392 Mg/ha. The
results from the various models showed that the
MVRVR
model
featured the highest R2 value and the lowest errors; the
MLR
model had the lowest R2 value and the highest errors with a
high overestimation in the biomass range between 0 and 200
Mg/ha. It was also indicated that the
MLPNN
model had a rela-
tively good saturation point, but the results was not able to
follow the trend of validation data. The superior performance
of the
MVRVR
model is related to the fact that it is able to fol-
low the trend of validation data using limited ground data.
In addition, the output values of the validation data were
synthesized using only a limited number of relevance vectors
(12 percent of training data). Furthermore, the
MVRVR
model
has the highest saturation point (297.81 Mg/ha) in compari-
son with those of the
MLPNN
(264.69 Mg/ha),
MLR
(255.66
Mg/ha), and
SVR
(224.75 Mg/ha) methods. It is evident that
although each approach has positive properties, none of them
is able to thoroughly solve the problem of underestimation. In
future work, we will focus on machine learning techniques,
such as deep learning algorithms, in an attempt to capture all
the details relating to
AGB
data, with the aim of avoiding the
problem of underestimation at higher levels.
Acknowledgments
The authors would like to thank Dr. Sumantyo and Dr. Pour-
shakouri for supplying the
ALOS
PALSAR
data and the forest
biomass data, respectively.
References
Amini, J., and J.T.S. Sumantyo, 2009. Employing a method on
SAR and optical images for forest biomass estimation, IEEE
Transactions on Geoscience and Remote Sensing, 47(12):4020–
4026: doi: 10.1109/TGRS.2009.2034464.
Camps-Valls, G., L. Bruzzone, J.L. Rojo-Alvarez, and F. Melgani,
2006. Robust support vector regression for biophysical variable
estimation from remotely sensed images, IEEE Geoscience
and Remote Sensing Letters, 3(3):339–343: doi: 10.1109/
LGRS.2006.871748.
Carreiras, J.M.B., M.J. Vasconcelos, and R.M. Lucas, 2012.
Understanding the relationship between aboveground biomass
and ALOS PALSAR data in the forests of Guinea-Bissau (West
Africa), Remote Sensing of Environment, 121:426–442: doi:
10.1016/j.rse.2012.02.012.
Chave, J., C. Andalo, S. Brown, M.A. Cairns, J.Q. Chambers, D.
Eamus, H. Folster, F. Fromard, N. Higuchi, T. Kira, J.P. Lescure,
B.W. Nelson, H. Ogawa, H. Puig, B. Riera, and T.
Yamakura, 2005. Tree allometry and improved estimation of carbon
stocks and balance in tropical forests, Oecologia, 145(1):87–99:
doi: 10.1007/s00442-005-0100-x.
Del Frate, F., and D. Solimini, 2004. On neural network algorithms
for retrieving forest biomass from SAR data, IEEE Transactions
on Geoscience and Remote Sensing, 42(1):24–34: doi: 10.1109/
TGRS.2003.817220.
Dobson, M.C., F.T. Ulaby, T. Le Toan, A. Beaudoin, E.S. Kasischke,
and N. Christensen, 1992.
Dependence of radar backscatter on coniferous forest biomass, IEEE
Transactions on Geoscience and Remote Sensing, 30(2):412–415:
doi: 10.1109/36.134090.
Englhart, S., V. Keuck, and F. Siegert, 2011. Aboveground biomass
retrieval in tropical forests - The potential of combined X-
and L-band SAR data use, Remote Sensing of Environment,
115(5):1260–1271: doi: 10.1016/j.rse.2011.01.008.
Englhart, S., V. Keuck, and F. Siegert, 2012. Modeling aboveground
biomass in tropical forests using multi-frequency SAR data - A
comparison of methods, IEEE Journal of Selected Topics in
Applied Earth Observations and Remote Sensing, 5(1):298–306:
doi: 10.1109/JSTARS.2011.2176720.
Fransson, J.E.S., and H. Israelsson, 1999. Estimation of stem volume
in boreal forests using ERS-1 C- and JERS-1 L-band SAR data,
International Journal of Remote Sensing, 20(1):123–137: doi:
10.1080/014311699213640.
Geisser, S., 1993. Predictive Inference, Chapman and Hall, New York,
260 p.
Guo, Q., W. Li, H. Yu, and O. Alvarez, 2010. Effects of topographic
variability and lidar sampling density on several DEM
interpolation methods, Photogrammetric Engineering & Remote
Sensing, 76(6):701–712.
Hame, T., Y. Rauste, O. Antropov, H.A. Ahola, and J. Kilpi, 2013.
Improved mapping of tropical forests with optical and SAR
imagery, Part II: Above ground biomass estimation, IEEE Journal
of Selected Topics in Applied Earth Observations and Remote
Sensing, 6(1):1939–1404: doi: 10.1109/JSTARS.2013.2241020.
Heimann, M., and M. Reichstein, 2008. Terrestrial ecosystem carbon
dynamics and climate feedbacks, Nature, 451(7176):289–292:
doi: 10.1038/nature06591.
Huang, S., S.A. Hager, K.Q. Halligan, I.S. Fairweather, A.K.
Swanson, and R.L. Crabtree, 2009. A comparison of individual
tree and forest plot height derived from lidar and InSAR,
Photogrammetric Engineering & Remote Sensing, 75(2):159–167.
Imhoff, M.L., 1995. Radar backscatter and biomass saturation:
Rami cations for global biomass inventory, IEEE Transactions
on Geoscience and Remote Sensing, 33(2):511–518: doi:
10.1109/36.377953.
Imhoff, M.L., and D.B. Gesch, 1990. The derivation of a sub-canopy
digital terrain model of a flooded forest using synthetic aperture
radar, Photogrammetric Engineering & Remote Sensing,
56(8):1155–1162.
Kasischke, E.S., and L.L. Bourgeau-Chavez, 1997. Monitoring South
Florida wetlands using ERS-1 SAR imagery, Photogrammetric
Engineering & Remote Sensing, 63(3):281–291.
Kiaei, M., and A. Samariha, 2011. Fiber dimensions, physical and
mechanical properties of five important hardwood plants, Indian
Journal of Science and Technology, 4(11):1460–1463.
Le Quere, C., M.R. Raupach, J.G. Canadell, G. Marland, L. Bopp,
P. Ciais, T.J. Conway, S.C. Doney, R.A. Feely, P. Foster, P.
Friedlingstein, K. Gurney, R.A. Houghton, J.I. House, C.
Huntingford, P.E. Levy, M.R. Lomas, J. Majkut, N. Metzl, J.P.
Ometto, G.P. Peters, I.C. Prentice, J.T. Randerson, S.W. Running,
J.L. Sarmiento, U. Schuster, S. Sitch, T. Takahashi, N. Viovy, G.R.
vander Werf, and F.I. Woodward, 2009. Trends in the sources
and sinks of carbon dioxide, Nature Geoscience, 2(12):831–836:
doi: 10.1038/ngeo689.
Lu, D., 2006. The potential and challenge of remote sensing-based
biomass estimation, International Journal of Remote Sensing,
27(7):1297–1328: doi: 10.1080/01431160500486732.
Lucas, R., J. Armston, R. Fairfax, R. Fensham, A. Accad, J.
Carreiras, 2010. An evaluation of the ALOS PALSAR L-band
backscatter-above ground biomass relationship Queensland,
Australia: Impacts of surface moisture condition and vegetation
structure, IEEE Journal of Selected Topics in Applied Earth
Observations and Remote Sensing, 3(4):576–593: doi: 10.1109/
JSTARS.2010.2086436.
Lucas, R.M., N. Cronin, A. Lee, M. Moghaddam, C. Witte, and
P. Tickle, 2006. Empirical relationships between AIRSAR
backscatter and LiDAR-derived forest biomass, Queensland,
Australia, Remote Sensing of Environment, 100(3):407–425: doi:
10.1016/j.rse.2005.10.019.
48
January 2016
PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING