July 3, 2014

John S. Howard
Director, FERC Hydro Compliance
FirstLight Power Resources/GDF Suez
Northfield Mountain Station
99 Millers Falls Road
Northfield, MA 01360
Re: Stakeholder comments on transects and detailed study sites for Study 3.1.2

Dear John,

I reviewed the “Relicensing Study 3.1.2. Northfield Mountain/Turners Falls Operations Impact on Existing Erosion and Potential Bank Instability: Selection of Detailed Study Sites” dated June 2014 (“Selection of Study Sites Report”), which presents the list of proposed calibration and representative locations for detailed study that will be used for Study No. 3.1.2. I attended the June 24, 2014 meeting with stakeholders to discuss this report. During this meeting, various stakeholders including myself requested additional information. Gomez and Sullivan sent us the requested information after the close of business on June 27, 2014. The purpose of this letter is to give you input on the proposed calibration and representative locations on behalf of the Connecticut River Watershed Council (CRC). We collaborated with the Franklin Regional Council of Governments (FRCOG) and other members of the Connecticut River Streambank Erosion Committee (CRSEC) on these comments and analyses.

Our analysis indicates that calibration transects and representative locations should either be moved or added to better represent the range of site conditions and influences that exist along the river.

Study Schedule

Stakeholders were asked to provide input before the Fourth of July holiday because field work is due to start immediately the following week. We would like to point out that the Revised Study Plan study schedule stated, “Fixed riverbank transects will be selected during the winter or early spring 2014.” We had anticipated being contacted about the transects during the winter or early spring; instead we have been asked to provide comments after having some key information for less than a week. As such, our comments included in this letter are “quick and dirty” given the allotted time. We would appreciate that the FirstLight Team honor stakeholder’s good faith efforts to provide thoughtful comments by giving us more time in the future.
Purpose of Comments

FirstLight and its consultants have provided location information on calibration locations, representative locations, and detailed site assessment locations. We will focus our comments on calibration locations and representative locations, but have some comments on detailed site assessments because that information was also given to us.

A calibration location is defined as a detailed study site established at an existing, permanent transect where data collection will occur to calibrate the BSTEM model. The Selection of Study Sites Report proposes 17 calibration locations. Representative locations are defined as detailed study sites established throughout the impoundment at locations that exhibit a representative range of riverbank features, characteristics and erosion conditions. The Selection of Study Sites Report proposes 15 representative locations. Detailed site assessments were part of the Full River Reconnaissance (FRR), already completed but report due in September, and a 2-page description was prepared for each one along with a site sketch. There are 36 locations throughout the impoundment and they were selected based on how representative they were of features or characteristics present in that area (see p. 4-1 of June 2014 report).

Role of restored sites

FirstLight and its predecessors have restored many river segments between 1996 and 2013. Turners Falls Impoundment Restoration Sites Maps 1 through 5 were sent to us on June 27. I compared those sites against the Proposed Representative & Calibration Locations for Detailed Study in Figures 7-2 through 7-6 of the “Selection of Study Sites Report” to make the following table. Note: the Army Corps of Engineers has engaged in bank stabilization efforts using tires and rip-rap that are not noted on the maps.

Representative and transect calibration locations that are located on restored sites
Representative Locations Transect Calibration Locations
90B: Upstream end of Shearer, Phase 1 9R: Campground Point, Phase 2
119B: Downstream end of Skalski, Phase 2 8B-R: Wallace/Watson, Phase 3
26: Urgiel upstream, Phase 2 6A-L: Skalski, Phase 2
10R: Urgiel upstream, Phase 2 6A-R: Flagg, Phase 1
29: Wickey, Phase 1 10R: Urgiel upstream, Phase 2
21: Kendall, Phase 2 5C-R: Bennett Meadow, Phase 2
2L: Bonnette Farm, preventative maintenance 3R: Kendall, Phase 2
2L: Bonnette Farm, preventative maintenance

Observations: 7 of 15 representative locations and 8 of 17 calibration locations are located at previously restored sites. Given the purpose of the BSTEM model, it seems odd to have such a high number of altered states used as calibration locations.

Stakeholder Recommendations: See FRCOG comment letter.
Analyses of the information provided

In order to assess whether or not the representative locations and calibration locations are adequately representative of conditions out on the river, we looked at several different characteristics and tried to group the study sites into those categories to see if anything is missing or over-represented.
1. Riverbank features and characteristics.

Table 7-3 in the Selection of Detailed Studies Report shows a matrix of riverbank features and characteristics vs. the proposed representative locations for detailed study. Such a matrix was not completed for calibration locations, and I did not complete one using Table 7-2.

Observations: There is only one representative site with a low upper riverbank height and one that has a medium height. Looking at Table 7-2, it appears that all calibration sites have “heavy” or some “moderate” upper riverbank vegetation, but no sparse vegetation.

Stakeholder Recommendations: Consider changing the sites so that there is a more representative mix of upper riverbank heights and vegetation.
2. Land use

A.
We had asked for transect locations in relation to land use maps. For some reason, we were given a map showing the detailed site assessment locations used in the FRR on the land use maps, even though input is not being requested on those locations. I therefore eyeballed the Proposed Representative & Calibration Locations for Detailed Study in Figures 7-2 through 7-6, plus one change as presented in Powerpoint at the June 26 meeting, against the land use maps given to us. Below is a matrix of locations split by study locations similar in nature to Table 7-2 in the Selection of Study Sites Report.

Land Use

Boat-based represent. location Land-based represent. location Existing Transect calibration location Existing transect calibration & representative location
Agriculture
(possible additional category separations shown below, but not provided to us) 87B 29*, 21* 8B-L, 6A-R*, 3R* 8B-L, 7L. 10L, 4L, 2L*
Agriculture with un-tilled veg buffer
Agriculture tilled to the edge
Agriculture with a single line of trees along river
Agriculture grazing land with veg buffer
Agriculture grazing land w/cattle edge ? ? ? ?
Barren (Not present in the study area)
Developed
Forest entire 500 ft swath 26* 9R* 10R*
Forest on an island 11I 6A-I
Forest with some developed land 12B, 6A-L*, 5C-R*, 11L
Forest with some agriculture land 119B 18 7R, 3L
Forest is narrow buffer near agriculture
Forest with some transportation
Forest with agriculture and transportation 75B
Non-Forested Wetland (Very little in study area and none along river’s edge)
Transportation (roads and bridge crossings)
* Indicates a restored site.

Observations: 10R and 26 seem duplicative. In stakeholder and FirsLight meetings, there has been talk about whether or not the type of agriculture at the top of bank affects erosion (tilling to the edge vs. hayfields or a tree buffer), and FirstLight has only provided information on “agriculture.” With more time, I might have been able to take a stab at splitting the agriculture types using MassGIS OLIVER coverages and Google Earth). Counting up the representative locations, there are 13 forested sites and 7agriculture sites. Forest with narrow buffer near agriculture is not well represented, whereas straight forest might be over-represented. Counting up the calibration locations, there are 8 agriculture sites and 9 forested sites.

Stakeholder Recommendations: Break down agriculture into more categories. Make sure that the sites with a narrow vs. wide forested buffers are adequately represented in the calibration sites and the representative sites. Representative locations may have too many forest sites and not enough agriculture sites – confirm, and if so, move a transect.
B.
Since we were sent the “Land Use & Detailed Site Assessments” maps 1-5 on June 27, I categorized the detailed site assessments in a similar way, as shown below.

Land Use Detailed Site Assessment Locations
Agriculture
(possible additional category separations shown below, but not provided to us) 6, 7 (restored or just downstream of restored), 25 (restored), 10, 11, 29 (restored), 21 (just downstream of restored), 19, 20
Agriculture with un-tilled vegetated buffer
Agriculture tilled to the edge
Agriculture with a single line of trees along river
Agriculture grazing land with vegetated buffer
Agriculture grazing land with cattle to the edge ?
Barren (Not present in the study area)
Developed
Forest entire 500 ft swath
Forest on an island
Forest with some developed land
Forest with some agriculture land
Forest is narrow buffer near agriculture
Forest with some transportation
Forest with agriculture and transportation 27, 26 (restored), 34, 2, 4, 3 (restored)
34, 33, 32,
28, 30
35, 24, 9, 14, 1
22 (restored), 12, 13, 36, 15, 16, 17, 18, 31
5
36
Non-Forested Wetland (Very little in study area and none along river’s edge)
Transportation (Roads and bridge crossings)

Observations: No site assessments on developed land – land is often not developed right at the river’s edge, but the section below the French King gorge has several parcels that fit into that category. There is perhaps an overabundance of forest with narrow buffer near agricultural land. Not sure if there is any hayfield agricultural land that was looked at. Forest on an island may be over-represented.

Stakeholder Recommendations: The FRR field work is already completed, but ideally there would have been some analysis of a site with developed land in the buffer, fewer sites on islands, and more agricultural sites that hadn’t already been restored.
3. River Morphology

Rivers naturally erode and aggrade and change course over time, and natural erosion or sediment accumulation is one aspect of Study 3.1.2. In this analysis, we looked at sites in relation to river morphology. The 2007 Field report noted that there were areas that experienced erosion not explained by shear velocities. Appendix 4 to the Field Report contained a Hydraulic Analysis conducted by Woodlot in 2007. It would be useful to identify sites that experience erosion in unexpected places, such as point bars or banks subject to relatively low shear stress.
A.
I categorized sites using the Proposed Representative & Calibration Locations for Detailed Study, Figures 7-2 through 7-6 plus one change as presented in Powerpoint at the June 26 meeting.

River segment Boat-based represent. location Land-based represent. location Existing Transect calibration location Existing transect calibration & representative location
Inside bend 9R*, 11I 7L*, 4L
Outside bend 12B, 75B, 119B* 6A-L*, 11L 7R
Straight 90B* 26*, 29*, 21*, 18 8B-R*, 6A-R*, 10L, 5C-R*, 3R*, 3L 8B-L, 6A-I, 10R*, 2L*
* Indicates a restored site.

Observation: I categorized quickly, and some of the categorization choices may be debatable. Counting up the representative locations, there are 2 inside bends, 4 outside bends, and 9 straight runs. Counting up the calibration locations, there are 4 inside bends, 3 outside bends, and 10 straight runs.

Stakeholder Recommendations: We would like to see more inside bends at representative locations. Move 2 straight run representative locations to inside bend locations. Also, move 2 calibration locations to an inside bend and an outside bend location.

B.
I categorized sites for the detailed site assessment locations based on the “Land Use & Detailed Site Assessments” maps 1-5.
River morphology Detailed Site Assessment Locations
Inside bend 27, 22, 11, 26 (restored), 12, 13, 21, 18, 31, 33
Outside bend 28, 35, 5, 10, 14, 15, 1, 2, 34, 30, 32
Straight 23, 24, 6, 7, 25, 9, 29 (restored), 36, 16, 17, 20, 19

Observations: Most of the outside bend sites are in that short segment between Vernon Dam and just downstream of Stebbins Island. Otherwise, outside bend sites are somewhat lacking.

Stakeholder Recommendations: The FRR field work is already completed, but ideally we would recommend moving 1-2 sites just below the Vernon Dam to other ouside-bend locations.
4. Hydraulic influences/geographic extent of fluctuation ranges

River fluctuation due to 1) the operation of the Vernon Dam and peaking facilities upstream, 2) the operation of the Turners Falls Dam, and 3) the operation of Northfield Mountain Pumped Storage are also causes of erosion in this section of the Connecticut River. We have not yet seen the results from the water level loggers in the river to truly understand where the river fluctuations vary, but we broke down the river into segments that we hypothesize might have different fluctuation ranges or heights. FirstLight’s February 2013 “Hydraulic Modeling Assessment of the TF Impoundment” identified hydraulic break points at the French King Gorge and just below Stebbins Island near the Vernon Dam. The remaining breakpoints are based on stakeholder knowledge of the river and/or intuition.

A.
I categorized sites using the Proposed Representative & Calibration Locations for Detailed Study, Figures 7-2 through 7-6 plus one change as presented in Powerpoint at the June 26 meeting.
River segment Boat-based represent. location Land-based represent. location Existing Transect calibration location Existing transect calibration & representative location
Barton Cove to FK Gorge/Bridge 12B 9R*
French King Gorge to Shearer 75B
Shearer to Route 10 bridge 90B*, 119B* 26* 8B-R*, 6A-R*, 6A-L*, 10L, 5C-R* 8B-L, 7R, 7L, 6A-I, 10R*
Route 10 bridge to state line 29* 4L
State line to just below Stebbins Island 21*, 18 3R*, 3L 2L*
Stebbins Island to Vernon Dam 11I, 11L
* Indicates a restored site.

Observation: Fluctuation from Northfield Mountain may be greatest near the tailrace, which is the segment between French King Gorge and Shearer. There is only one representative location (thanks to DEP’s comments) and no transect calibration locations in this area. Most of the sites are from Shearer to Route 10 Bridge.

Stakeholder Recommendations: Move 5 representative and 5 transect calibration locations to other areas of the river.

B.
I categorized sites for the detailed site assessment locations based on the “Land Use & Detailed Site Assessments” maps 1-5.

River segment Detailed Site Assessment Locations
Barton Cove to FK Gorge/Bridge 7, 28
French King Gorge to Shearer 22, 35, 5
Shearer to Route 10 bridge 23, 24, 6, 7, 25, 9, 26, 10, 11
Route 10 bridge to state line 12, 13, 29, 36, 14
State line to just below Stebbins Island 15, 21, 16, 17, 18, 19, 20, 1, 31
Stebbins Island to Vernon Dam 2,34, 4, 33, 30, 32, 3

Observations: Too many sites between state line and just below Stebbins Island. Lots of sites for teeny segment from Stebbins Island to Vernon Dam compared to other shorter segments. There are not enough sites in the two lowest segments.

Stakeholder Recommendations: The FRR field work is already completed, but ideally we would have recommended moving 3 sites between the Vernon Dam and the state line to the segments between Barton Cove and Shearer.

Other areas of comment

Section of river downstream of French King Gorge. At the June 24 meeting, we discussed at length the Selection of Detailed Study Sites page 6-4 where it states that the segment between the Turners Falls Dam and French King Gorge “is not of interest to the objectives of the study and therefore will not be investigated in detail.” CRC does not agree with the approach to largely leave this section out of Study 3.1.2.. The Lower Riverbank Sediment maps sent to us on June 27 indicate that the lower riverbank in the section of the Pool below French King Gorge is approximately 50% silt/sand composition rather than bedrock and rip rap. Also, there have been three phase 2 restoration projects in this section, and 6 preventative maintenance locations in this section. An unsolicited phone call to the CRC came in this week from a person concerned about retreating riverbank in this segment. This section has different fluctuation patterns than the rest of the river and deserves to be included in the study, albeit perhaps with fewer sites and taking into account the higher frequency of bedrock.

Upper and lower riverbank. We have not had time to digest the memorandum sent to us on June 30, 2014 defining the upper and lower riverbank and will send comments under separate cover at a later date.

Thank you for the opportunity to provide input on the field sites used for this study.

Sincerely,

Andrea F. Donlon
River Steward

Cc: Kimberly Noake MacPhee, FRCOG
Dave Foulis and Brian Harrington, MassDEP
Ken Hogan, FERC
Bill McDavitt, NOAA NMFS
Russ Cohen, MA Division of Ecological Restoration
CRSEC members: Tom Miner, Michael Bathory, John Bennett