content in vegetation could be highly variable. Sixteen sites distributed over upland,
sloped, and valley areas were randomly chosen in the undisturbed old field in Koffler
Science Reserve. Each study site was formed by two 50-m transects that were oriented
perpendicular to one another (forming a cross). Each transect was made up of 10 plots
50 cm ´ 50 cm with a 5-m distance from one plot to another. In each plot, a set of
grassland composition, biophysical, biochemical, and ground hyperspectral data was
measured using various instruments, field visual determination, and lab-based
approaches. The details are discussed below.
Samples were collected for the three dominant species in each chosen plot to be
used for lab-based spectral reflectance measurements and chlorophyll extraction. Leaf
spectral reflectance was measured for each species in a controlled laboratory environment using an analytical spectral device (ASD) FieldSpec 3 Max Portable spectroradiometer with a fiber optic. The illumination source was an ASD Pro Lamp (14.5 V,
50 W) that is adapted for indoor lab diffuse reflectance measurements over the region
350–2500 nm. The wavelength range for the collected reflectance data was 350–
2500 nm with a spectral sample of 1.4 nm in the 350–1050-nm range and 2 nm in the
1000–2500-nm range. For optimal spectral measurements, the fiber-optic sensor was
pointed opposite to the angle of the light source, which is approximately 45° pointing
downward, with the leaf sample being placed underneath the point of highest light
intensity. Five replicate measurements were taken for each leaf sample and were
averaged to suppress the measurement noise.
Canopy spectral reflectance was measured at each plot using the same instrument
that was used for lab-based leaf spectral reflectance measurement. The fiber-optic
sensor, with a field view of 25°, was hand held approximately 1 m above and
perpendicular to the ground. Calibration was conducted frequently with a calibrated
diffuse white reference panel provided by ASD Inc. during field sampling to minimize
measurement noise. To minimize atmospheric perturbations, spectral measurements
were taken during cloudless days between 10:00 AM and 2:00 PM. Canopy
composition data were collected using a 0.5 m ´ 0.5 m frame. The frame was lightly
tossed onto a random spot around the plot. Then the percentage top layer cover (grass,
forbs, shrubs, and standing dead), lower layer cover (litter, moss, lichen, rock, bare
ground), and species cover were visually determined within the frame and recorded.
Photographs were taken over each frame for more careful scrutiny back in the
laboratory.
Samples of the three dominant species were collected from every other plot for
species chlorophyll extraction. The samples were collected from every other plot
because field observation indicated that the difference in vegetation species and their
coverage within 5 m (the distance between two neighboring plots) are marginal. The
collected vegetation samples were kept on ice in a dark thermal cooler and then
transferred immediately to a −20°C freezer to prevent leaf pigments from deteriorating. In order to extract chlorophyll, each leaf sample was first placed in a microtube
containing a solution of 100% acetone and 30 mg of sodium bicarbonate. The tubes
were then placed into a mini–bead beater in order to vigorously break the leaf cells and
separate the chlorophyll content. Next, the homogenized solution was centrifuged to
separate the fibrous plant material from the supernatant (the chlorophyll content
DATA AND METHODS
129
sloped, and valley areas were randomly chosen in the undisturbed old field in Koffler
Science Reserve. Each study site was formed by two 50-m transects that were oriented
perpendicular to one another (forming a cross). Each transect was made up of 10 plots
50 cm ´ 50 cm with a 5-m distance from one plot to another. In each plot, a set of
grassland composition, biophysical, biochemical, and ground hyperspectral data was
measured using various instruments, field visual determination, and lab-based
approaches. The details are discussed below.
Samples were collected for the three dominant species in each chosen plot to be
used for lab-based spectral reflectance measurements and chlorophyll extraction. Leaf
spectral reflectance was measured for each species in a controlled laboratory environment using an analytical spectral device (ASD) FieldSpec 3 Max Portable spectroradiometer with a fiber optic. The illumination source was an ASD Pro Lamp (14.5 V,
50 W) that is adapted for indoor lab diffuse reflectance measurements over the region
350–2500 nm. The wavelength range for the collected reflectance data was 350–
2500 nm with a spectral sample of 1.4 nm in the 350–1050-nm range and 2 nm in the
1000–2500-nm range. For optimal spectral measurements, the fiber-optic sensor was
pointed opposite to the angle of the light source, which is approximately 45° pointing
downward, with the leaf sample being placed underneath the point of highest light
intensity. Five replicate measurements were taken for each leaf sample and were
averaged to suppress the measurement noise.
Canopy spectral reflectance was measured at each plot using the same instrument
that was used for lab-based leaf spectral reflectance measurement. The fiber-optic
sensor, with a field view of 25°, was hand held approximately 1 m above and
perpendicular to the ground. Calibration was conducted frequently with a calibrated
diffuse white reference panel provided by ASD Inc. during field sampling to minimize
measurement noise. To minimize atmospheric perturbations, spectral measurements
were taken during cloudless days between 10:00 AM and 2:00 PM. Canopy
composition data were collected using a 0.5 m ´ 0.5 m frame. The frame was lightly
tossed onto a random spot around the plot. Then the percentage top layer cover (grass,
forbs, shrubs, and standing dead), lower layer cover (litter, moss, lichen, rock, bare
ground), and species cover were visually determined within the frame and recorded.
Photographs were taken over each frame for more careful scrutiny back in the
laboratory.
Samples of the three dominant species were collected from every other plot for
species chlorophyll extraction. The samples were collected from every other plot
because field observation indicated that the difference in vegetation species and their
coverage within 5 m (the distance between two neighboring plots) are marginal. The
collected vegetation samples were kept on ice in a dark thermal cooler and then
transferred immediately to a −20°C freezer to prevent leaf pigments from deteriorating. In order to extract chlorophyll, each leaf sample was first placed in a microtube
containing a solution of 100% acetone and 30 mg of sodium bicarbonate. The tubes
were then placed into a mini–bead beater in order to vigorously break the leaf cells and
separate the chlorophyll content. Next, the homogenized solution was centrifuged to
separate the fibrous plant material from the supernatant (the chlorophyll content
DATA AND METHODS
129