RIVER.....VICKSBURG.....FLOOD ALERT
Posted: Thu May 23, 2002 10:06 am
And here are just a few of the problems with the data and effect/bennifit analysis put forth by the COE, as reported by the EPA
"The keystone of the Corps hydrology analyses is the elevation area curves depicted on Plates 4-7 to 4-10. The text (page 6-30) implies that data points from ten satellite images were used to generate these Plates. The Corps has not, however, documented any of the data points on these elevation-area curves. Additionally, the method used for "fitting" the data points to the curves was not given (page 6-31 only states that "a best fit curve routine was used"). While we assume that a composite elevation-area curve for the four reaches was developed and used in the analyses, this was not stated in the Draft EIS. This information is needed in order for the technical reviewer to verify the shape of the curve as currently assigned. Since all other analyses depend on the data generated from these curves, its documentation is important. These ten data points and the curves generated by these data points are the basis for the hydrological analyses, including the stage-frequency and stage area data (Table 6-9), stage-frequency curves (Plate 4-22- 4-23), elevation-area curves (Plates 4-7-4-10, elevation storage curves (Plate 4-11), and elevation-duration curves (Plates 4-24 - 4-25), as well as the hydrology analysis done for the economics evaluation. Minor differences in the shape of the curve, particularly given the total number of points (10) and the cluster of data points (nine are at 91.9 feet or below, one is at 100.3 feet), could result in significant differences in the reported number of flooded acres.
It is stated that the "satellite scenes were classified with an unsupervised classifier," (page 6-30). This apparently means that there was no validation of the GIS technique used by either photo-interpretation or ground-truthing.
There is a gap in data points from the satellite imagery between 91.9 feet (nine points at 91.9 feet and below) and 100.3 feet (one point at 100.3 feet). This gap results in lower confidence in assessing the less frequent flood stage events.
Ten satellite images were used, five of which were outside of the growing season when many of the wetland impacts and claimed cropland flood reduction are considered to be most critical to project assessment.
Accurate flood profiles can only be established by basinwide channel cross-section data. The Corps has only two cross-sections in the connecting channel between Steele Bayou and Little Sunflower), and no basinwide cross-section data.
The land use classification data used are out-dated (over ten years old).
Additionally, spatially explicit data were apparently not used in the environmental assessments; therefore, specific geographic locations of impacts cannot be determined.
There is insufficient detail about how the pump will be operated (see discussion in section B below).
The only hydrograph presented is for the l00-year flood (Plate 4-21). However, the pumping project will have more impacts on the more frequent flood events (such as the 2-year flood). Therefore, including the hydrographs for the more frequent events would allow for more confidence in results interpretation.
The Corps' analysis apparently assumes a static channel system throughout the project area with no changes due to channel filling by sedimentation. This would affect the analysis of storage curves and rate of delivery of water."
[ May 23, 2002: Message edited by: M.B.Waters ]
"The keystone of the Corps hydrology analyses is the elevation area curves depicted on Plates 4-7 to 4-10. The text (page 6-30) implies that data points from ten satellite images were used to generate these Plates. The Corps has not, however, documented any of the data points on these elevation-area curves. Additionally, the method used for "fitting" the data points to the curves was not given (page 6-31 only states that "a best fit curve routine was used"). While we assume that a composite elevation-area curve for the four reaches was developed and used in the analyses, this was not stated in the Draft EIS. This information is needed in order for the technical reviewer to verify the shape of the curve as currently assigned. Since all other analyses depend on the data generated from these curves, its documentation is important. These ten data points and the curves generated by these data points are the basis for the hydrological analyses, including the stage-frequency and stage area data (Table 6-9), stage-frequency curves (Plate 4-22- 4-23), elevation-area curves (Plates 4-7-4-10, elevation storage curves (Plate 4-11), and elevation-duration curves (Plates 4-24 - 4-25), as well as the hydrology analysis done for the economics evaluation. Minor differences in the shape of the curve, particularly given the total number of points (10) and the cluster of data points (nine are at 91.9 feet or below, one is at 100.3 feet), could result in significant differences in the reported number of flooded acres.
It is stated that the "satellite scenes were classified with an unsupervised classifier," (page 6-30). This apparently means that there was no validation of the GIS technique used by either photo-interpretation or ground-truthing.
There is a gap in data points from the satellite imagery between 91.9 feet (nine points at 91.9 feet and below) and 100.3 feet (one point at 100.3 feet). This gap results in lower confidence in assessing the less frequent flood stage events.
Ten satellite images were used, five of which were outside of the growing season when many of the wetland impacts and claimed cropland flood reduction are considered to be most critical to project assessment.
Accurate flood profiles can only be established by basinwide channel cross-section data. The Corps has only two cross-sections in the connecting channel between Steele Bayou and Little Sunflower), and no basinwide cross-section data.
The land use classification data used are out-dated (over ten years old).
Additionally, spatially explicit data were apparently not used in the environmental assessments; therefore, specific geographic locations of impacts cannot be determined.
There is insufficient detail about how the pump will be operated (see discussion in section B below).
The only hydrograph presented is for the l00-year flood (Plate 4-21). However, the pumping project will have more impacts on the more frequent flood events (such as the 2-year flood). Therefore, including the hydrographs for the more frequent events would allow for more confidence in results interpretation.
The Corps' analysis apparently assumes a static channel system throughout the project area with no changes due to channel filling by sedimentation. This would affect the analysis of storage curves and rate of delivery of water."
[ May 23, 2002: Message edited by: M.B.Waters ]