Megan L. Frayer
This project used data collected by the Advanced Very High Resolution Radiometer (AVHRR) on water reflectance for the Florida Bay region over the 12-year period from July 1985 to September 1997, and field data on light attenuation and changes in bottom cover.
Stumpf, Richard P.
The full article is available via journal subscription or single article purchase. The first page of the article may be viewed on the website below.
Frayer, M. L.; Durako, M. J.; Brock, J. C.
Ellerberg, H.
Pennock, J. R.
The full article is available via journal subscription or single article purchase. The abstract may be viewed on the website below.
Pennock, J. R.
The full article is available via journal subscription or single article purchase. The abstract may be viewed on the website below by selecting the volume and issue number.
Bittaker, H. F.
The full article is available via journal subscription or single article purchase. The abstract may be viewed on the website by selecting the volume and issue number.
Chen, J.
The full article is available via journal subscription or single article purchase. The abstract may be viewed on the website below.
Chen, J.
The full article is available via journal subscription or single article purchase. The abstract may be viewed on the website below.
Monthly average image values were corrected month-by-month for a systematic bias for greater offsets with earlier scenes at the Gulf Stream site (Stumpf et al. 1997). In addition to reflectance calculations, cloud-contaminated pixels were flagged using a combination of techniques: thresholds on sea-surface temperature and channel 2, and spatial variations on channel 2 and a thermal band.
While each AVHRR sensor is calibrated prior to launch, the lack of onboard calibration raised concerns for post-launch calibration and sensor deterioration. The problem was resolved when a relative calibration of the sensors on the afternoon satellites was developed (Rao and Chen, 1995 & 1996)
The failure of NOAA-13 on launch in 1993 left NOAA-11 as the primary satellite in 1994. The orbit of NOAA-11 had precessed to late afternoon precluding quality imagery in winter 1994 due to low sun angles. NOAA-11 failed in September 1994 and an afternoon satellite was not available till the launch of NOAA-14 in late December.
In the dataset the typical month has 3-9 nominally cloud-free scenes prior to a1994 and 8-16 scenes starting in 1995. Winter 1994 had only four acceptable scenes owing to the extremely late overpass time of NOAA-11 (1600 local standard time). Scenes with solar angles >70 degrees were discarded. Gaps occur in late summer months for 1988-1989. From the available images, monthly average images were determined using the cloud-free pixels. Then the mean winter reflectance from December to March and the mean summer reflectance June to September were determined.
Field ecologists in the project area document light attenuation using scalar radiometers that measure photo-synthetically-active radiation (PAR, 400-700nm). We used a Licor 4pi PAR sensor for profiles of scalar irradiance at 0.25-m increments from 0.25 m (or 0.5m) depth to the bottom.
SEAGRASSES Bottom coverage of Thalassia testudium in Rankin Lake, Johnson Key Basin, and Rabbit Key Basin was determined using a modified Braun-Blanquet cover-abundance method (Mueller-Dombois and Ellerberg, 1974). Coverage for 1991 was obtained by randomly sampling 10 locations per basin from a 0.5-nautical (nmi) grid. Coverage for 1994-1996 was obtained by systematic random sampling within 30-35 tessellated hexagons per basin. All data were obtained by sampling four 0.25 meter square quadrats per station (north, east, south and west of the vessel). The density values correspond to the following modified Braun-Blanquet cover-abundance scale: 5=cover of more than 75% of the quadrat; 4=50-75% cover; 3=25-50% cover; 2=5-25% cover, 1=numerous stems with less than 5% cover or scattered with up to 5% cover; 0.5=few stems with small cover; 0.1=solitary, with small cover, and 0=not present. Frequency of occurrence and density information for each species within a particular basin was calculated using the formulas: Frequency = no. of occupied quad/total no. of quads and Density = sum of B-B scale values/total no. of quads. The areal extent of each cover was estimated by kriging the mean station data for each basin using Surfer software package Golden Software. With the nominal 1-km spacing of the samples, each 1-m sq sample site fell in a different pixel. Reflectances for all pixels with sites in a basin were averaged; the combination of kriging and averaging dampens out the scaling problems between the 1-m sq samples with the individual 1.2 km sq pixels.
Any use of trade, product, or firm names is for descriptive purposes only and does not constitute endorsement by the U.S. Government
U.S. Department of the Interior, U.S. Geological Survey
Comments and suggestions? Contact: Heather
Henkel - Webmaster
Generated by mp version 2.8.18 on Mon May 16 10:13:05 2011