Hundreds of HNBs and HVIs were used model above-ground
biomass of 4 leading world crops (rice, wheat, corn, alfalfa)
based on 2 years of detailed data acquired for these crops
in the irrigated agricultural fields of California by Marshall
and Thenkabail. The best biomass models explained greater
than 80% variability using highly selective sequential search
methods (SSM) involving two-band HVIs or multi-band HVIs
involving one to 3 HNBs. The key is also to select specific
narrowbands (~10 nm or less) from two or three distinct
portion of the spectrum: (a) green and near-infrared, (b) blue
and NIR, (c) near-infrared (NIR) and short-wave infrared
(SWIR), or (d) green, NIR, and SWIR. These specific HNBs
may change for crop to crop and even within crop. But,
what needs to be noted is that there are some very selective
HNBs and HVIs derived off them (see the Table 2 and 3 in
the highlight article of this issue) which consistently perform
highly across different crops and their varying characteristics.
The HNBs and HVIs vary because selecting one HNB versus
another often makes only a slight difference (e.g., 680 nm or
690 nm are highly correlated and perform about the same;
similarly 855 nm or 910 nm are highly correlated and perform
similarly; a point also noted in other reported studies). But, in
modelling a specific crop an HVI involving 855 nm and 680
nm (HVI855680) may perform marginally better than an HVI
involving 910 nm and 690 nm (HVI910690). This performance
may, at times differ for another crop. This does not mean that
we need to use both the indices. It will suffice to use a single
index (e.g., HVI855680) to model both crops because the two
HVIs are equally good (e.g., one index may have an R-square
of 0.85 with biomass and another 0.87; in which case we will
select the one with 0.87 and ignore the one with 0.85). The
study by Marshall and Thenkabail, re-affirms the fact that
there are redundant bands as well as there are specific HNBs
that are of highest importance to model specific biophysical
and biochemical characteristics of crops or vegetation. These
HNBs and HVIs perform significantly better than any known
broadband derived indices. Readers should refer to various
Tables and figures of the paper by Marshall and Thenkabail
for better understanding. Further, greater, comprehensive
understanding can be acquired by going through Table 2, 3,
and 4 as well as Figure 4a and 4b of the highlight article in
this special issue.
Hyperspectral remote sensing (or Imaging Spectroscopy) is
fast moving from an era of research into an era of applications.
Many spaceborne hyperspectral sensors (e.g., HypspIRI, UAV
based platforms, interest from private entities; see Thenkabail
et al., 2011) are planned in near future. This special issue
adds to maturing knowledge of hyperspectral remote sensing
in general, and hypersepctral remote sensing of vegetation
and agricultural crops in particular.
Credit to this special issue on “
Hyperspectral Remote
Sensing of Vegetation and Agricultural Crops
” goes to several
people. I would like to thank the authors for their outstanding
work. Each paper was reviewed by at least 3 reviewers.
Good reviewers are few but pivotal for success of any
quality journal. I am thankful to many good reviewers who
helped improve the quality of each paper. I am grateful for
the advice, support, and guidance of Dr. Russell Congalton,
Editor in Chief of PE&RS. Ms. Jeanie G. Congalton, PE&RS
manuscript coordinator, was always there with her insights.
Finally, I would like to thank the U. S. Geological Survey
(USGS), especially USGS Western Geographic Science Center
(WGSC) and its leadership, for all the opportunities and
encouragement that I have received over the years.
7.0 References:
Middleton, E.M.; Ungar, S.G.; Mandl, D.J.; Ong, L.; Frye, S.W.;
Campbell, P.E.; Landis, D.R.; Young, J.P.; Pollack, N.H.,
2013. “The Earth Observing One (EO-1) Satellite Mission:
Over a Decade in Space,”
Selected Topics in Applied
Earth Observations and Remote Sensing, IEEE Journal
of
, vol.6, no.2, pp.243,256, April 2013. doi: 10.1109/
JSTARS.2013.2249496
Thenkabail, P.S., Lyon, G.J., and Huete, A. 2011. Book
entitled: “Hyperspectral Remote Sensing of Vegetation”.
CRC Press- Taylor and Francis group, Boca Raton,
London, New York. Pp. 781 (80+ pages in color).
Reviews of this book:
/
isbn/9781439845370
Thenkabail, P.S., Mariotto, I., Gumma, M.K.,, Middleton,
E.M., Landis, and D.R., Huemmrich, F.K., 2013. Selection
of hyperspectral narrowbands (HNBs) and composition
of hyperspectral twoband vegetation indices (HVIs) for
biophysical characterization and discrimination of crop
types using field reflectance and Hyperion/EO-1 data.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED
EARTH OBSERVATIONS AND REMOTE SENSING,
Pp. 427-439, VOL. 6, NO. 2, APRIL 2013.doi: 10.1109/
JSTARS.2013.2252601.
Thenkabail, P.S., GangadharaRao, P., Biggs, T., Krishna, M.,
and Turral, H., 2007. Spectral Matching Techniques
to Determine Historical Land use/Land cover (LULC)
and Irrigated Areas using Time-series AVHRR
Pathfinder Datasets in the Krishna River Basin, India.
Photogrammetric Engineering and Remote Sensing. 73(9):
1029-1040.
Special Issue Editor
Dr. Prasad S. Thenkabail
Research Geographer
U. S. Geological Survey
Email:
PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING
August 2014
723
