Difference between revisions of "GCDAMP FLAHG Page"

From Glen Canyon Dam AMP
Jump to: navigation, search
 
(9 intermediate revisions by the same user not shown)
Line 49: Line 49:
 
|class="MainPageBG" style="width:55%; border:1px solid #cef2e0; background:#f5faff; vertical-align:top; color:#000;"|
 
|class="MainPageBG" style="width:55%; border:1px solid #cef2e0; background:#f5faff; vertical-align:top; color:#000;"|
 
{|width="100%" cellpadding="2" cellspacing="5" style="vertical-align:top; background:#f5faff;"
 
{|width="100%" cellpadding="2" cellspacing="5" style="vertical-align:top; background:#f5faff;"
! <h2 style="margin:0; background:#cedff2; font-size:120%; font-weight:bold; border:1px solid #a3bfb1; text-align:left; color:#000; padding:0.2em 0.4em;"> Spring Disturbance Flows [http://gcdamp.com/images_gcdamp_com/3/30/A3_-_20201006_-_FLAHG_Predicted_Effects.v5.Oct5.pdf (excerpts from the Predicted Effects document)] </h2>
+
! <h2 style="margin:0; background:#cedff2; font-size:120%; font-weight:bold; border:1px solid #a3bfb1; text-align:left; color:#000; padding:0.2em 0.4em;"> Updates </h2>
 
|-
 
|-
 
|style="color:#000;"|
 
|style="color:#000;"|
 
To test both spring and fall HFEs, the HFE protocol proposed using two sediment accounting periods.
 
These sediment accounting periods are used to track the quantity of new Paria River sand available for
 
building beaches in Marble Canyon during an HFE, and HFEs are only triggered if the quantity of new
 
sand is large. This sediment accounting approach to triggering HFEs, coupled with state-of-the-art
 
sediment monitoring (Topping and Wright 2016), eliminates the possibility of unintentionally scouring
 
sediment resources from Marble Canyon during HFEs. When the HFE protocol was first proposed, it was
 
estimated that sediment-triggered fall HFEs would occur approximately two out of every three years and
 
sediment-triggered spring HFEs would occur approximately once every three years (Wright and Kennedy
 
2011). [http://gcdamp.com/images_gcdamp_com/3/30/A3_-_20201006_-_FLAHG_Predicted_Effects.v5.Oct5.pdf]
 
 
[[File:HFE windows.jpg|center|600px]]
 
 
Although Sediment-Triggered Spring HFEs and Proactive Spring HFEs are now possible, analysis of Paria
 
River discharge data indicates that Sediment-Triggered Spring HFEs may occur less frequently than
 
originally estimated (Grams and Topping 2018). Sediment accounting data available since the HFE
 
protocol was operationalized in 2012 bear this out. Specifically, since 2012 the sediment trigger for a fall
 
HFE has been reached 6 times (i.e., 2012-2016, 2018; no HFE occurred in 2015 owing to green sunfish)
 
while the sediment trigger for a spring HFE has never been reached (Grams and Topping 2020). Testing
 
of 5 fall HFEs over the past 8 years has benefitted sediment resources and reduced uncertainties
 
concerning sandbar response to HFEs in general (Figure 2). However, regular testing of fall HFEs since
 
2012 is also correlated with a growing population of brown trout in Lees Ferry (Runge and others 2018)
 
and critical uncertainties concerning the role of spring HFEs in achieving biological resource objectives
 
remain unanswered (Figure 2).
 
 
By May 2020, the FLAHG and GCMRC completed design of a conceptual hydrograph that included a high
 
spring release that was within power plant capacity. The FLAHG hydrograph capitalizes on a unique low
 
flow of 4,000 ft3 /s for 5 days, which is needed to conduct maintenance on the apron of Glen Canyon
 
Dam (see Figure 3). The FLAHG hydrograph proposes to follow this low flow disturbance with a high flow
 
disturbance that will culminate in a discharge of up to 25,000 ft3 /s for 82 hours. This combination of
 
desiccation at low flows followed by scour at high flows is hypothesized to disturb benthic habitats to a
 
much greater extent than either the low or high flows alone (Kennedy and others 2020). [http://gcdamp.com/images_gcdamp_com/3/30/A3_-_20201006_-_FLAHG_Predicted_Effects.v5.Oct5.pdf]
 
 
[[File:FLAHG_hydrograph.png|center|600px]]
 
 
Below we present predicted effects of the FLAHG hydrograph on the 11 LTEMP Resource Goals using the
 
Knowledge Assessment rubric from 2019
 
(http://gcdamp.com/index.php?title=Portal:GCDAMP_Knowlege_Assessments). For comparison, and to
 
anchor predictions concerning the FLAHG hydrograph, we also used the Knowledge Assessment
 
framework to predict effects of Spring and Fall HFEs on LTEMP Resource Goals. To simplify analysis of
 
hydrograph impacts, the FLAHG narrowed consideration of testing this hydrograph to sometime in
 
March based on the following two main reasons: 1) both the 1996 and 2008 Spring HFEs also occurred in
 
March, which will simplify evaluation and comparison of new biological data that might be collected
 
around a FLAHG hydrograph to earlier data from those spring HFEs, and 2) a March test of the FLAHG
 
hydrograph will minimize adverse impacts of the low flow to Recreational Experience by avoiding the
 
start of the commercial river trip motor season in April.
 
 
Knowledge Assessment groups often evaluated multiple specific measures to capture all the facets of a
 
given LTEMP goal. For example, the goal for Rainbow Trout Fishery is, “Achieve a healthy high-quality
 
recreational rainbow trout fishery in GCNRA and reduce or eliminate downstream trout migration
 
consistent with NPS fish management and ESA compliance.” To capture both facets of this goal, the
 
Knowledge Assessment team considered two specific measures: rainbow trout abundance in Lees Ferry,
 
and rainbow trout abundance at the Little Colorado River confluence. The predicted resource responses
 
to a given action often varied, depending on which specific measure was considered. We capture this
 
variation in predictions by presenting bookend, “lowest performing” and “highest performing”,
 
scenarios gleaned from the assessments and their differing specific measures. Note that the assessment
 
summaries and graphs are based on detailed assessments for each resource that were performed by
 
multiple subject matter experts (see section V. Acknowledgements for a complete list of participants).
 
Those detailed assessments are contained in a spreadsheet for each resource that accompany this
 
document. The detailed resource assessments were based on consideration of peer-reviewed literature,
 
modelling, and other quantitative science as well as more qualitative expert opinions, similar to previous
 
Knowledge Assessments. [http://gcdamp.com/images_gcdamp_com/3/30/A3_-_20201006_-_FLAHG_Predicted_Effects.v5.Oct5.pdf]
 
 
[[File:PredictedEffectsLowest.png|left|400px]]
 
 
 
[[File:PredictedEffectsHighest.png|right|400px]]
 
 
|-
 
! <h2 style="margin:0; background:#cedff2; font-size:120%; font-weight:bold; border:1px solid #a3b0bf; text-align:left; color:#000; padding:0.2em 0.4em;">[http://gcdamp.com/images_gcdamp_com/b/bd/A6a_-_2020.11.10_-_Project_O_-_AMWG_November_Version_-_Clean.pdf Project O]</h2>
 
|-
 
|style="color:#000;"|
 
 
==Project Element O.1. Does Disturbance Timing Affect Food Base Response?==
 
 
This element includes funding for tracking food base response to the FLAHG/GCRMC
 
hydrograph in Year 1 (Figure 1). Due to logistical constraints, we will focus our sampling efforts
 
in and around Lees Ferry. Specifically, we propose to sample aquatic insect drift intensively at
 
four time periods: just prior to the low flow associated with apron repair, during the low flow,
 
during the subsequent high flows after apron repair, and during the base flows immediately after
 
the high flow. Sampling would occur over the course of 1-2 weeks at 5-10 sites throughout Glen
 
Canyon and Upper Marble Canyon. Sites will be identical to our regular monitoring sites spaced
 
roughly equidistant from GCD (starting at River Mile [RM] -15) to the head of Badger Rapid
 
(RM 8), allowing us to contrast food web impacts above and below the Paria River confluence.
 
 
The objectives of this sampling are twofold:
 
#Quantify invertebrate export resulting from spring flow disturbance. Specifically, quantify the extent to which nonnative New Zealand mud snails are exported or suffer high mortality as a result of these flows, and the extent to which patterns of midge, blackfly, and Gammarus drift differ from baseline conditions in past spring seasons and during prior, fall HFEs.
 
#Quantify organic matter export resulting from drying during apron repair and subsequent flushing during high flows, which may have concomitant impacts on aquatic insect habitat and food resources.
 
 
==Project Element O.2. Bank Erosion, Bed Sedimentation, and Channel Change in Western Grand Canyon ==
 
 
In order to address the above research questions, we intend to study channel response to dam
 
operations in a short (~1 to 3 km) study reach (to be selected) downstream from Quartermaster
 
Canyon. We will work with the Hualapai Tribe to select a specific reach that is critical for boat
 
navigation. In the first part of our analysis, we propose to use available remote sensing data sets
 
to document historical changes in bank and river channel morphology. The second part of the
 
analysis will include collection of repeat surveys of the riverbed within the selected study reach
 
before, during, and following a dam-released flow pulse. The repeat surveys will allow
 
quantification of the magnitude and spatial distribution of channel morphological change
 
associated with the flow pulse and the return to normal dam operations. This analysis will be
 
conducted by using the field data to develop a streamflow and sediment transport model for the
 
study reach. The model will allow evaluating bed response in a predictive framework to
 
determine whether there are systematic changes in bed elevation caused by dam operations.
 
Because similar issues exist upstream along the deltas of the Colorado and San Juan arms of
 
Lake Powell, this research project also could provide guidance for management of other large
 
reservoirs in the Colorado River Basin.
 
 
==Project Element O.3. Aeolian Response to a Spring Pulse Flow==
 
 
We
 
propose to conduct research during the FLAHG spring flow at a combined archaeological
 
dunefield-sand bar monitoring site from Project D.1 where NPS is also considering conducting
 
riparian vegetation removal through the LTEMP vegetation management project, which could
 
increase the aeolian transport of sand from the sandbar to adjacent archaeological site. We
 
propose to leverage the FLAHG flow to measure sand drying rates, change in exposed subaerial
 
sand, and aeolian sediment transport potential during the extended low flow of 4,000 ft3/s
 
and subsequent high flow at the study site.
 
 
==Project Element O.4. Riparian Vegetation Physiological Response==
 
 
We propose to select a subset of species from those listed in Project C.2 (Figure 5), depending on
 
availability of those species at accessible river sites. The species selected will represent both
 
flood tolerant and drought tolerant species and will likely include arrowweed (Pluchea sericea),
 
coyote willow (Salix exigua), Emory’s baccharis (Baccharis emoryi), tamarisk (Tamarix spp.),
 
Emory’s sedge (Carex emoryi), and available Juncus spp. The daily measurements will include
 
those of the mesocosm experiments: stomatal conductance, stem water potential, relative
 
humidity, leaf turgor, and soil moisture. Measurements will be collected daily starting two days
 
prior to the experimental flows through two days after.
 
 
The target location for this work will be in and around Lees Ferry. If, however, weather
 
conditions result in a late start to the spring growing season and the target species are unlikely to
 
be fully leafed out and active at the time of the FLAHG flows, we will relocate to the area near
 
Phantom Ranch. Plants will be active in that river segment during the FLAHG flows but are less
 
desirable simply due to logistics.
 
 
==Project Element O.5. Mapping Aquatic Vegetation Response to a Spring Pulse Flow ==
 
 
 
We propose to use this unique opportunity (coupled with advances in Project Element E.2 in
 
TWP FY2021-23) to understand how a spring disturbance flow affects the dominant primary
 
producers in Glen Canyon, with a secondary objective of determining the scale at which we
 
might be able to do so. This would be designed as a before and after-impact study, with one trip
 
immediately prior to the low flow (e.g., late February or early March), one trip immediately after
 
the higher flow (e.g., late March or early April), and one trip in June to detect vegetation
 
response and recovery. Images from the June trip will be compared to baseline images from 2016
 
and 2019 that were taken in years lacking a spring disturbance flow. If we can detect a change in
 
aquatic vegetation cover and/or composition on a short-term scale (i.e., one season), then that
 
result will inform the frequency at which we should undergo aquatic vegetation surveys (Project
 
Element E.2). For example, if we can detect change within a season over multiple trips, this
 
method could be considered a sensitive tool for detecting vegetation community responses to
 
dam operations and would further our understanding of factors that drive primary production in
 
Glen Canyon (Project E).
 
 
==Project Element O.6. Brown Trout Early Life Stage Response to a Spring Pulse Flow ==
 
 
While a low steady flow timed during peak emergence could improve short-term swim-up and
 
growth conditions for brown trout fry, we anticipate an energetic cost for newly emerged brown
 
trout fry during the spring disturbance flow. Therefore, we plan on collecting data during the
 
year of the spring disturbance flow and comparing results to a non-flow year using the methods
 
outlined in Project Element H.3, which will improve understanding of how spring flow
 
configurations may affect brown trout in their early life history stages. Results will be compared
 
to age-1 brown trout catch from the TRGD project in fall of Year 2 for comparison, which would
 
be an indicator of brown trout recruitment strength following the spring disturbance flow
 
(Project Element H.2).
 
 
==Project Element O.7. Native Fish Movement in Response to a Spring Pulse Flow==
 
 
We propose the use of sonic tags to track the responses of humpback
 
chub and flannelmouth sucker to the proposed spring disturbance flow. While rare, adult
 
razorback sucker will be included in this study if captured or detected on the remote Submersible
 
Ultrasonic Receiver (SUR) network, since the species also spawns in spring and may respond to
 
simulated flood hydrograph (USFWS, 2018). We propose targeting two study sites, one within
 
the JCM-west reach to economically align with ongoing studies (Project Elements G.5, G.6), and
 
one in the Lake Mead formation area below ~RM 235 that is accessible via up-runs from Pearce
 
Ferry. We propose to sonic tag approximately 35 adult fish per site, and USFWS will sonic tag
 
another 35 fish as a match, for a total of ~70. Approximately half the tags will be inserted into
 
adult flannelmouth sucker, the other half into humpback chub. If we capture any adult razorback
 
sucker, they will receive priority over the other two species since they are rare in the system. The
 
effort will benefit from an array of ~27 remote SURs already in place within Grand Canyon
 
distributed from the LCR to Pearce Ferry to passively track native fish movement. We will also
 
actively track fish at both sites as time and resources allow, combined with analysis of general
 
movement patterns at the JCM-west vs. JCM-east sites to refine mark-recapture modeling
 
(Project Elements G.3, G.6).
 
 
==Project Element O.8. Do Disturbance Flows Significantly Impact Recreational Experience?==
 
 
Surveys will be conducted to obtain information on recreationists’ preferences and economic
 
values associated with flow attributes specific to a spring disturbance flow experiment.
 
Consistent with past research of angler and whitewater boater’s flow preferences, the surveys
 
will be designed to elicit economic values using choice experiment instruments in addition to
 
investigation into other quantitative and qualitative metrics of recreationists’ preferences and
 
perspectives. Participants will be intercepted immediately prior, during, and following the spring
 
disturbance flow, differing from past recreational surveys (Bishop and others, 1987; Neher and
 
others, 2017), and respondents will either be interviewed on-site or sent a mail survey packet,
 
with a follow-up protocol for non-responders.
 
 
==Project Element O.9. Are There Opportunities to Meet Hydropower and Energy Goals with Spring Disturbance Flows? (funded in N.1).==
 
 
This project element is addressed in Project N. The objective of Project N is to identify,
 
coordinate, and collaborate on design, monitoring, and research opportunities associated with all
 
operational experiments at GCD to meet hydropower and energy resource objectives, as stated in
 
the LTEMP ROD (U.S. Department of the Interior, 2016b). The possibility of higher spring flow
 
experiments will be addressed in Project N. Funding for this element is included in the Project
 
N.1 budget.
 
 
==Project Element O.10. Sandbar and Campsite Response to Spring Disturbance Flow (funded in B.1). ==
 
 
Because deposition at sandbars and associated campsite area increase is expected to be much
 
lower in response to the ~25,000 ft3
 
/s or lower pulse flow than occurs during sediment-enriched
 
fall HFEs, extensive field measurements of sandbars and campsites before and after the pulse
 
flow are not planned. Instead, evaluation of the sandbar and campsite response to the pulse flows
 
will rely on daily images from the network of remote cameras that is maintained as part of
 
project B.1. Funding for this element is included in the Project B.1 budget.
 
 
==Project Element O.11. Decision Analysis ==
 
 
This project element will utilize the multi-criteria decision and value of information analysis that
 
was undertaken in the decision analysis to support development of the GCD LTEMP (Runge and
 
others, 2015). The fundamental resource goals and performance metrics will be utilized to
 
instruct the proposed monitoring and research in the individual project elements and in the
 
allocation of funding within and across project elements. A workshop in Year 2 will occur
 
following the implementation of the FLAHG hydrograph. The workshop will provide an
 
opportunity to summarize the FLAHG hydrograph results, evaluate trade-offs identified with the
 
spring disturbance flow, and present an overview of the decision process with respect to
 
prioritization of funding for monitoring and research related to this and other potential future
 
spring flow experiments.
 
 
==Budget Justification==
 
 
Funding in Year 1 for all project elements will be sought through the Experimental Management
 
Fund (C.5 Experimental Management Fund; see TWP FY2021-23) except for Project Element
 
O.11. Note that Reclamation retains decision-making authority for the allocation of funds from
 
the C.5 Experimental Management Fund. Also, requests to support Project O through the
 
Experimental Management Fund should be considered in context with other requests from the
 
Experimental Management Fund (i.e. including, but not limited to Projects A.4, B.6.1-5, and
 
J.3). Additionally, consideration of funding for Project O elements should be done in accordance
 
with the recommendation developed by the BAHG on October 8, 2020.
 
 
In Year 1 Project Element O.11 will seek funding from TWP carryover funds from prior years,
 
or through annual review of the TWP, or through other Reclamation considerations. Likewise,
 
in Year 2, funding for O.1 and O.2 will be sought from TWP carryover funds from prior years, or
 
through annual review of the TWP, or through other Reclamation considerations. Opportunities
 
to leverage external resources and support from Program partners will be considered and
 
explored by GCMRC and Reclamation for Year 2 funding. It should be noted that funding for a
 
third year of data analysis and modeling is required for Project Element O.2 in order for it to be
 
successfully completed; however, at this time a funding source has not been identified.
 
Three project elements have funding requests in Year 2; 1) $146,563 for O.1 to quantify food
 
base response to spring disturbance flows, 2) $161,959 for O.2 to identify whether dam
 
operations exacerbate or mitigate boat navigation challenges associated with bed-sediment
 
accumulation in the western Grand Canyon, and 3) $61,359 for O.11 to conduct decision
 
analysis. Year 2 funding totals include salary for short-term field technicians, travel and training,
 
operating expenses, and logistics. The remaining project elements (O.3-O.10) seek funding only
 
in Year 1. The proposed funding for these elements includes cooperator support, travel and
 
training, operating expenses, logistics, and salary for short-term field technician support.It should
 
be noted that most funding for GCMRC salaries involved in project elements O.3-O.10 is already
 
included in related project elements in Projects A through N in the TWP FY2021-23.
 
 
  
  
Line 323: Line 66:
 
|style="color:#000;"|
 
|style="color:#000;"|
  
 +
*[http://gcdamp.com/index.php?title=Spring_Disturbance_Flow Spring Disturbance Flow page]
 
*[http://gcdamp.com/index.php?title=The_HFE_Page The HFE page]
 
*[http://gcdamp.com/index.php?title=The_HFE_Page The HFE page]
 +
*[http://gcdamp.com/index.php?title=FOOD_BASE The Foodbase page]
 
*[http://gcdamp.com/index.php?title=Portal:GCDAMP_DOI_Direction DOI Direction page]
 
*[http://gcdamp.com/index.php?title=Portal:GCDAMP_DOI_Direction DOI Direction page]
*[http://gcdamp.com/index.php?title=GCDAMP_Budget GCDAMP Budget page]
 
 
*[http://gcdamp.com/index.php?title=GCDAMP_BAHG_Page BAHG page]
 
*[http://gcdamp.com/index.php?title=GCDAMP_BAHG_Page BAHG page]
 
*[http://gcdamp.com/index.php?title=FY21-23_Triennial_Budget_and_Workplan FY21-23 Triennial Budget and Workplan]
 
*[http://gcdamp.com/index.php?title=FY21-23_Triennial_Budget_and_Workplan FY21-23 Triennial Budget and Workplan]
  
 
|-
 
|-
! <h2 style="margin:0; background:#cedff2; font-size:120%; font-weight:bold; border:1px solid #a3b0bf; text-align:left; color:#000; padding:0.2em 0.4em;">Presentations and Papers</h2>
+
! <h2 style="margin:0; background:#cedff2; font-size:120%; font-weight:bold; border:1px solid #a3b0bf; text-align:left; color:#000; padding:0.2em 0.4em;"> Documents </h2>
 
|-
 
|-
 
|style="color:#000;"|
 
|style="color:#000;"|
  
*[https://www.usbr.gov/uc/progact/amp/twg/2020-10-15-twg-meeting/20201015-FLAHGHydrograph-508-UCRO.pdf Presentation on Spring Disturbances and the FLAHG Hydrograph ]
 
  
*[[Media:A1_-_20191220_-_FLAHG_Charge.pdf| FLAHG Charge]]
+
|-
*[[Media:A3_-_20201006_-_FLAHG_Predicted_Effects.v5.Oct5.pdf| FLAHG Predicted Effects document]]
+
! <h2 style="margin:0; background:#cedff2; font-size:120%; font-weight:bold; border:1px solid #a3b0bf; text-align:left; color:#000; padding:0.2em 0.4em;"> Presentations and Papers </h2>
*[[Media:A6a_-_2020.11.10_-_Project_O_-_AMWG_November_Version_-_Clean.pdf| Project O - AMWG November Version]]
+
|-
*[[Media:A6b_-_20201110_-_Project_O_-_AMWG_November_Version_-_Track_Changes.pdf| Project O - AMWG November Version - Track Changes]]
+
|style="color:#000;"|
*[[Media:A4_-_20201008_-_Final_BAHG_recommendation_Project_O_20201008.pdf| Final BAHG recommendation Project O]]
+
 
 +
'''2023'''
 +
*[https://www.usbr.gov/uc/progact/amp/amwg/2023-08-17-amwg-meeting/20230817-FLAHGProposalAmendHFEProtocolOtherConsiderations-508-UCRO.pdf FLAHG – Proposal to Amend the HFE Protocol & Other Considerations]
 +
*[https://www.usbr.gov/uc/progact/amp/twg/2023-06-15-twg-meeting/20230615-FLAHGProposalAmendHFEProtocol-508-UCRO.pdf FLAHG – Proposal to Amend the HFE Protocol]
 +
 
 +
'''2020'''
 
*[[Media:A5_-_20201015_-_TWGMotionsActionItemsDRAFT.pdf| TWG Motions Action Items]]
 
*[[Media:A5_-_20201015_-_TWGMotionsActionItemsDRAFT.pdf| TWG Motions Action Items]]
 
*[[Media:A2_-_20200820_-_AMWGMotionsActionItemsFINAL.pdf| AMWG Motions Action Items FINAL]]
 
*[[Media:A2_-_20200820_-_AMWGMotionsActionItemsFINAL.pdf| AMWG Motions Action Items FINAL]]
 
*[[Media:A0 - 2020.11.17 - Draft AMWG Agenda - NOV20.pdf| Draft AMWG Agenda - NOV20]]
 
*[[Media:A0 - 2020.11.17 - Draft AMWG Agenda - NOV20.pdf| Draft AMWG Agenda - NOV20]]
 
'''2020'''
 
 
*[https://www.usbr.gov/uc/progact/amp/amwg/2020-11-17-amwg-meeting/20201117-FlowAdHocGroupCharge-508-UCRO.pdf Flow Ad Hoc Group Charge ]
 
*[https://www.usbr.gov/uc/progact/amp/amwg/2020-11-17-amwg-meeting/20201117-FlowAdHocGroupCharge-508-UCRO.pdf Flow Ad Hoc Group Charge ]
*[https://www.usbr.gov/uc/progact/amp/amwg/2020-11-17-amwg-meeting/20201117-PredictedEffectsSpringDisturbanceFlowLTEMPResources-508-UCRO.pdf Predicted Effects of a Spring Disturbance Flow on LTEMP Resources Link is to a PDF file – Oct. 6, 2020]
 
 
*[https://www.usbr.gov/uc/progact/amp/amwg/2020-11-17-amwg-meeting/20201117-BudgetAdHocGroupRecommendationProjectO-508-UCRO.pdf Budget Ad Hoc Group Recommendation for Project O Link is to a PDF file – Oct. 8, 2020]
 
*[https://www.usbr.gov/uc/progact/amp/amwg/2020-11-17-amwg-meeting/20201117-BudgetAdHocGroupRecommendationProjectO-508-UCRO.pdf Budget Ad Hoc Group Recommendation for Project O Link is to a PDF file – Oct. 8, 2020]
 
*[https://www.usbr.gov/uc/progact/amp/amwg/2020-11-17-amwg-meeting/20201117-DraftProject%20O-EvaluatingResourceResponseSpringDisturbanceFlows-508-UCRO.pdf Draft - Project O: Is Timing Really Everything? Evaluating Resource Response to Spring Disturbance Flows Link is to a PDF file – Nov. 10, 2020]
 
*[https://www.usbr.gov/uc/progact/amp/amwg/2020-11-17-amwg-meeting/20201117-DraftProject%20O-EvaluatingResourceResponseSpringDisturbanceFlows-508-UCRO.pdf Draft - Project O: Is Timing Really Everything? Evaluating Resource Response to Spring Disturbance Flows Link is to a PDF file – Nov. 10, 2020]
Line 363: Line 108:
 
|style="color:#000;"|
 
|style="color:#000;"|
  
==AMWG Action Item (May 2019)==
 
It was suggested that the TWG take up consideration of the remaining “HFE Assessment” action item, which reads, “A next step would be for GCMRC to identify experimental flow options that would consider high valued resources of concern to the GCDAMP (i.e., recreational beaches, aquatic food base, rainbow trout fishery, hydropower, humpback chub and other native fish, and cultural resource), fill critical data gaps, and reduce scientific uncertainties.”  The AMWG did not object to the remaining action item passing from GCMRC to the TWG.
 
 
==DOI Direction (August 2019)==
 
Dr. Petty Guidance Memo – Page 3 – “In response to stakeholder input at recent AMWG meetings, the feasibility of conducting Spring High Flow Experiments (HFE), along with modeling for improvements and efficiencies that benefit resources including natural, cultural, recreational, and hydropower should be explored.  As a potential starting point, I encourage you to consider opportunities to conduct higher spring releases within power plant capacity, along with spring HFEs that may be triggered under the current LTEMP Protocol”
 
 
==TWG motion on the 2021-2023 Triennial Budget and Work Plan (June 2020)==
 
The TWG recommends that the AMWG recommend for approval to the Secretary of Interior the Triennial Work Plan and Budget FY 2021-2023 as provided to the TWG on June 23, 2020 and as requested to be revised by the TWG during their meeting on June 23 and 24, 2020. <br>
 
 
Revisions requested by the TWG on June 23 and 24, 2020:
 
*Include the GCMRC B.4 work element in the budget ($58,000 for first 2 years and $64,000 for year 3).
 
*Remove and/or reduce GCMRC D.2 (approximately $39,000 in year 1, $36,000 in year 2, and $54,000 in year 3) and GCMRC D.3 (approximately $28,000 in year 1, $29,000 in year 2, and $0 in year 3).
 
*Include Havasu Creek and LCR-mouth gage in GCMRC A.1 at 17,000/year.
 
*Please change GCMRC Project N verbiage (Pg 294) from “For example, modeling a change in ramp rates to maintain or improve the hydropower and recreational resource objectives is a possible application of GCMRC Project N.” to: “For example, modeling a change in ramp rates to improve the hydropower resource objective is a possible application of Project N.”
 
*In accordance with direction provided by the AMWG as described in the FLAHG charge, include a project and/or project element to support the FLAHG charge, and provide funding if necessary.
 
*Remove Reclamation B.4, TWG Chair reimbursement (25,000 for FY 2021)
 
*Propose AGFD and GCMRC look to integrate work efforts to allow for an additional TRGD site to be monitored. Cost estimate for going from 1 TRGD to 2 TRGD sites is approximately 67,000.
 
*Prioritize the use of available, unprogrammed and unspent funds from FY 2020, 2021 and 2022 towards funding GCMRC G.6 (JCM-West) in 2023.
 
 
== AMWG motion on the 2021-2023 Triennial Budget and Work Plan (August 2020)==
 
AMWG recommends to the Secretary of the Interior the Glen Canyon Dam Adaptive Management Program Triennial Budget and Work Plan—Fiscal Years 2021-2023 (July 29, 2020 draft), subject to the following:
 
*Removal from GCMRC Project N the following verbiage:
 
"Past research into changes in regional energy costs attributed to alteration of GCD operations have shown that no changes occur in hourly prices (U.S. Department of Interior, 2016b). Specifically, hourly energy prices at the regional hub important to GCD (Palo Verde) remain approximately the same with variation in production of energy at GCD. In addition, the analysis of experiments is a short-run analysis, assuming that demand for energy is inelastic (demand does not change with small changes in prices) and surplus power capacity exists. Therefore, changes in $/MW and $/MWh are accurate representations of the changes in consumer and producer surplus when evaluating minor, short-run changes in GCD operations. However, long- run changes in the energy sector may lead to a different economic outcome and a more complete modeling approach would be required. The evaluation of GCD operation and long-run changes in the electricity sector such as the integration of renewable energy, repurposing of federal hydropower resources, or power system capacity expansion would require a significant increase in research scope."
 
 
*Consideration of GCMRC Project O is deferred, but will be included in the 2021-2023 Triennial Budget and Work Plan as a proposal to be considered for the Reclamation C.5 Experimental Management Fund, pending revisions to be made by GCMRC and the Bureau of Reclamation and review by the Technical Work Group. After consideration and if recommended by AMWG, a springtime disturbance flow will be planned to occur in coordination with Glen Canyon Dam apron repairs, to ensure sufficient time to integrate the information and learning about the importance of springtime high flows into the 2021-2023 TWP, subject to an evaluation of the resource conditions described in the LTEMP ROD. <br>
 
 
*AMWG acknowledges and appreciates the effort to develop Project O in response to elements of the TWG Recommendation for the 2021-2023 Triennial Budget and Work Plan, consistent with guidance from the Secretary’s Designee (memo issued August 14, 2019), and in support of the Flow Ad Hoc Group charge. GCMRC is commended for their effort. <br>
 
 
*AMWG members will submit written comments to GCMRC and Reclamation on Project O no later than Friday, September 4, 2020. GCMRC and Reclamation will make revisions based on comments received and will submit the revised Project O plan for TWG consideration by Wednesday, October 7, 2020, for discussion at the October 2020 TWG meeting. AMWG directs the TWG to review the revised Project O and to forward a revised Project O recommendation for AMWG consideration no later than Friday, October 30, 2020. The AMWG will act on the TWG recommendation no later than Friday, November 20, 2020.
 
 
==FLAHG Recommendation to the TWG (September 2020) ==
 
The FLAHG, working with GCMRC, has identified a higher spring release within power plant capacity hydrograph that meets the requirements set forth by the LTEMP ROD and considers high value resources of concern to the GCDAMP (including recreational beaches, aquatic food base, rainbow trout fishery, hydropower, humpback chub and other native fish, cultural resources, and vegetation), fills critical data gaps, and helps to reduce scientific uncertainties. The FLAHG submits this hydrograph to the TWG for their consideration.
 
 
==Finalization of the Predicted Effects to Resources Document (October 2020)==
 
 
==BAHG Recommendation to the TWG on Project O (October 2020)==
 
The BAHG expresses support for a Project O that evaluates the effects of the proposed FLAHG hydrograph. The BAHG recommends to the TWG the following elements outlined in Project O (July 2020 draft) be prioritized in the following rank order for funding a first year of work effort: <br>
 
 
Tier 1 Projects (Weighted mean rankings from BAHG rating exercise: 9.3 and 8.6). The BAHG identified that these project elements are the most important for understanding the effects of the proposed FLAHG hydrograph.
 
*Project Element O.1. Does Disturbance Timing Affect Food Base Response?
 
*Project Element O.5. Mapping Aquatic Vegetation Response to a Spring Pulse Flow   
 
 
Tier 2 Projects (Weighted mean rankings from BAHG rating exercise: 6.8, 5.7, and 3.9). The BAHG identified that these project elements are very important for understanding the effects of the proposed FLAHG hydrograph.
 
*Project Element O.6. Brown Trout Early Life Stage Response to a Spring Pulse Flow
 
*Project Element O.2. Bank Erosion, Bed Sedimentation, and Channel Change in Western Grand Canyon
 
*Project Element O.7. Native Fish Movement in Response to a Spring Pulse Flow
 
 
Tier 3 Projects (Weighted mean rankings from BAHG rating exercise: ≤ 0). The BAHG identified that these project elements are important for understanding effects of the proposed FLAHG hydrograph.
 
*Project Element O.11. Decision Analysis
 
*Project Element O.4. Riparian Vegetation Physiological Response
 
*Project Element O.8. Do Disturbance Flows Significantly Impact Recreational Experience?
 
*Project Element O.3. Aeolian Response to a Spring Pulse Flow
 
 
The BAHG acknowledges that the following projects are already funded in the TWP as Projects N.1 and B.1, thus a recommendation on rank priority is unnecessary:
 
*Project Element O.9. Are There Opportunities to Meet Hydropower and Energy Goals with Spring Disturbance Flows? (funded in N.1).
 
*Project Element O.10. Sandbar and Campsite Response to Spring Disturbance Flow (funded in B.1). 
 
 
In our deliberations of Project O, the BAHG recommends that the Experimental Fund (Reclamation C.5) should not be used to support the following items and, to the extent that a proposed project element includes an item, that item should be removed from the project element: 
 
*Multi-year commitments because the decision to use the Experimental Fund is made on a year-by-year basis;
 
*Monitoring for experiments or activities that occur with a level of regularity or certainty that would lend themselves to be planned for and funded through the TWP because this is counter to the intent of the Experimental Fund; and
 
*Salaries for positions lasting more than one year (i.e. anything more than a one-year term position or contract) because this may lead to unreasonable expectations of work security. 
 
 
The BAHG also recommends that prioritizing elements in Project O for funding through the Experimental Fund should be made in context with other requests from the Experimental Fund and vice versa.
 
 
[[Image:ProjectOranking.png|600px]]
 
 
==TWG motion on the Flow Ad Hoc Group (FLAHG) Hydrograph (October 2020)==
 
The TWG recommends that the AMWG recommend to the Secretary of the Interior, to implement, when conditions warrant and consistent with the LTEMP protocol for implementing flow experiments, the spring disturbance flow hydrograph developed by the FLAHG in coordination with the GCMRC, as described in the FLAHG Predicted Effects document and associated presentations to the TWG on October 14, 2020.
 
 
==TWG motion on Project O (October 2020)==
 
The TWG recommends that the AMWG recommend to the Secretary of the Interior GCMRC’s “Project O” proposal, as provided to the TWG on October 7, 2020, for inclusion in the 2021-2023 Triennial Budget and Workplan, but with the following revisions and areas of emphasis:
 
*Reclamation retains decision-making authority for the allocation of funds from the C.5 Experimental Management Fund.
 
*Requests to support Project O through the Experimental Management Fund should be considered in context with other requests from the Experimental Management Fund (i.e. including, but not limited to Projects A.4, B.6.1-5, and J.3) and vice versa.
 
*Consideration of funding for Project O elements should be in accordance with the recommendation developed by the BAHG on October 8, 2020.
 
*Elements O.1 and O.2 – Funding requests for Year 2 should come from TWP carryover funds from prior years or through annual review of the TWP.
 
*Element O.11 – Funding requests for Year 1 and Year 2 could come from TWP carryover funds from prior years or through annual review of the TWP or other Reclamation considerations.
 
*References to specific years (e.g. FY2021) will be replaced with general references to Year 1 and Year 2 to accommodate uncertainty regarding implementation timing.
 
*Opportunities to leverage external resources and support from Program partners will be considered and explored by GCMRC and Reclamation.
 
 
==AMWG motion on the Flow Ad Hoc Group (FLAHG) Hydrograph (November 2020)==
 
The AMWG recommends to the Secretary of the Interior, to implement, when conditions warrant and consistent with the LTEMP protocol for implementing flow experiments, the spring disturbance flow hydrograph developed by the FLAHG in coordination with the GCMRC, as described in the FLAHG Predicted Effects document (dated October 6, 2020) and associated presentations to the AMWG on November 17, 2020.
 
 
==AMWG motion on Project O (November 2020)==
 
The AMWG recommends to the Secretary of the Interior GCMRC’s “Project O” proposal, as provided to the AMWG on November 10, 2020, for inclusion in the 2021-2023 Triennial Budget and Workplan, with the following areas of emphasis:
 
*Reclamation retains decision-making authority for the allocation of funds from the C.5 Experimental Management Fund.
 
*Requests to support Project O through the Experimental Management Fund should be considered in context with other requests from the Experimental Management Fund (i.e. including, but not limited to Projects A.4, B.6.1-5, andJ.3) and vice versa.
 
*Consideration of funding for Project O elements should be in accordance with the recommendation developed by the BAHG on October 8, 2020.
 
*Elements O.1 and O.2 – Funding requests for Year 2 should come from TWP carryover funds from prior years or through annual review of the TWP.
 
*Element O.11 – Funding requests for Year 1 and Year 2 could come from TWP carryover funds from prior years or through annual review of the TWP or other Reclamation considerations.
 
*Opportunities to leverage external resources and support from Program partners will be considered and explored by GCMRC and Reclamation.
 
  
 
|}
 
|}

Latest revision as of 10:19, 31 October 2023


Flahg (1).jpg

Flow Experiments Ad Hoc Group (FLAHG)

ESTABLISHED JANUARY 15, 2020, Status: Active

Charge

In accordance with direction provided by the AMWG at its August 18, 2018 meeting, and the Secretary Designees August 14, 2019 guidance to BOR and GCMRC, the FLAHG is charged with working with GCMRC to evaluate opportunities for conducting higher spring releases that may benefit high value resources of concern to the GCDAMP (recreational beaches, aquatic food base, rainbow trout fishery, hydropower, humpback chub and other native fish, cultural resources, and vegetation), fill critical data gaps, and reduce scientific uncertainties. As a starting point, the FLAHG shall consider the benefits of and opportunities for conducting higher spring releases within power plant capacity. The FLAHG and GCMRC will report their initial findings to the TWG in April 2020 so that the TWG and GCMRC can report their conclusions to the AMWG in May 2020.

The FLAHG is also charged with working with GCMRC to develop and propose a project element in the FY 2021-2023 Triennial Budget and Work Plan to identify other spring high flow options that may be allowed under the current LTEMP ROD (as determined by DOI) that may benefit high valued resources of concern to the GCDAMP (defined above), fill critical data gaps, and reduce scientific uncertainties.

Members

Peggy Roefer (chair), Jeff Arnold, Jan Balsom, Cliff Barrett, Rob Billerbeck, David Brown, Peter Bungart, Kelly Burke, Shane Capron, Winkie Crook, Craig Ellsworth, Charlie Ferrantelli, Clarence Fullard, Michelle Garrison, Paul Harms, Brian Healy, Leslie James, John Jordan, Vineetha Kartha, Ted Kennedy, Jakob Maase, Ryan Mann, Craig McGinnis, Jessica Neuwerth, Clayton Palmer, Bill Persons, Richard Powskey, Ben Reeder, Peggy Roefer, Scott Rogers, Seth Shanahan, Larry Stevens, Jim Strogen, Lee Traynham, Steve Wolff, Kirk Young.

Invited technical advisors: Helen Fairley, Heather Patno, Mike Moran, Scott VanderKooi

--
--
--

Updates



Links

Documents


Presentations and Papers

2023

2020

Process