Attachment_1_-_Pumpkin_Creek_Restoration_Design_Plan_1.pdf

LOW-TECH PROCESS-BASED

RESTORATION ON PUMPKIN CREEK

Planning and Design for Phase One

Prepared by:

Jordan Gilbert, Department of Watershed Sciences, 5210 Old Main Hill, Logan, UT 84322-5210

In association with:

BLM Montana-Dakotas (Miles City Field Office)

February 2022

P U

M P

K

IN

C R

E E

K R

E S

T O

R A

T

IO

N D

E S

IG

N

Department of Watershed Sciences, 5210 Old Main Hill, Logan, UT 84322-5210

Recommended Citation:

Gilbert, J.T., Morris, C. E., and Stuart, C. M. LOW-TECH PROCESS-BASED RESTORATION ON PUMPKIN CREEK:

Planning and Design for Phase One. Prepared for BLM Montana/Dakotas. Utah State University Wheaton Ecogeomorphology & Topographic Analysis Lab. Logan, UT. 53 Pages.

This work is licensed under a Creative Commons Attribution 4.0 International License.

http://creativecommons.org/licenses/by/4.0/

M P

K

IN

C R

E E

K R

E S

T O

R A

T

IO

N D

E S

CONTENTS

List of Figures

Summary

Introduction

Project location and context

Preliminary project goals and objectives

Planning design and approach

Low-tech process-based restoration

Rationale for design

Resource assessment

Valley setting

Geomorphic condition

Channel characteristics

Hydrology

Riparian assessment

Risk assessment

Potential future condition

Restoration indicators

Restoration design

Temporal design

Spatial design – reach delineation

Structural elements and complex design

Complex design

Adaptive management

Construction plan and logistics

Material sourcing, types, and fill volume estimates

Site access, material staging, and fueling/equipment storage

Implementation

References

Appendix A – Principles of riverscape health and restoration

Riverscapes priniciples

Restoration principles

Appendix B – Structure types and locations

Appendix C – PALS and BDA construction methods, structure types, and schematics

Appendix D – Riverscapes tools

M P

K

IN

C R

E E

K R

E S

T O

R A

T

IO

N D

E S

Riverscapes Context

Beaver Restoration Assessment Tool (BRAT)

LIST OF FIGURES

Figure 1 - Location of restoration project on Pumpkin Creek Figure 2 - Land ownership within the Pumpkin Creek watershed. The panel on the right is zoomed in over the area the BLM owns where this restoration design will be applied Figure 3 - Outline depicting an adaptation of NRCS’s Conservation Planning Process used to guide the restoration planning and design process (from Wheaton et al. 2019) Figure 4 - Pumpkin Creek and its valley bottom along the portion of riverscape owned by the BLM Figure 5 - Delineation of Pumpkin Creek’s valley bottom into geomorphic units. The square outlines the area where the project will begin in conjunction with a restoration workshop Figure 6 - The valley bottom of the restoration workshop area (square in figure 5) of Pumpkin Creek delineated into geomorphic units underlain by a lidar derived hillshade raster. Two reaches have some distinct inset floodplains, while the remainder is incised with no floodplain Figure 7 - Reaches within the project area with reach breaks based on geomorphic condition overlain on a 10 m DEM.

The inset photos illustrate reach conditions at different parts of the stream network Figure 8 - The USGS peak flow record for Pumpkin Creek just downstream of the project area demonstrates the high year to year variability in peak stream flow Figure 9 - This drone imagery illustrates the typical riparian composition of Pumpkin Creek. Herbaceous vegetation is found in the low-lying areas near the channel with occasional cottonwood and upland vegetation like sagebrush on the inactive floodplain Figure 10 - Infrastructure within the project area valley bottom that should be considered in implementing restoration Figure 11 - Stream evolution model (SEM) proposed by Cluer and Thorne (2014) illustrating approximate stages and pathways associated with recovery to stage 0. Restoration in Pumpkin Creek is intended to accelerate recovery trajectories Figure 12 - Locations of the 16 complexes identified along Pumpkin Creek in the project area and their primary objective Figure 13 - Design for complexes 1 – 4 Figure 14 - Design for complexes 5 – 8 Figure 15 - Design for complexes 9 – 12 Figure 16 - Design for complexes 13 - 16 Figure 17 - Adaptive management for monitoring and ongoing restoration of LTPBR complexes. Many of the concepts illustrated are also applicable at the scale of an individual structure, or at the scale of an entire project. From Chapter 6 of Wheaton et al., 2019 Figure 18 - Access roads and potential material staging areas within the downstream part project area Figure 19 - Access roads and potential material staging areas within the upstream part project area Figure 20 - Schematic of a bank-attached PALS Figure 21 - Schematic of a mid-channel PALS Figure 22 - Schematic of a channel-spanning PALS Figure 23 - Schematic of frames for securing wood pieces in PALS Figure 24 - Schematic of post placement for single row and staggered, double-row placements Figure 25 - Schematic illustration proper dam crest orientation Figure 26 - Schematic of a postless BDA

M P

K

IN

C R

E E

K R

E S

T O

R A

T

IO

N D

E S

Figure 27 - Instructions and schematic for building a postless BDA Figure 28 - Schematic of a BDA supported by a single row of posts Figure 29 - Schematic of a BDA supported by staggered, double rows of posts Figure 30 - Schematic with instructions on building a post-supported BDA Figure 31 - Outputs of the BRAT capacity model for the project area Figure 32 - Outputs of the BRAT restoration and conservation management model for the project area

M P

K

IN

C R

E E

K R

E S

T O

R A

T

IO

N D

E S

SUMMARY

Pumpkin Creek is a perennial tributary of the Tongue River, located in southwest Montana near Miles City. It is a prairie stream, draining ~700 mi2 of hilly, prairie landscape within Montana. The stream is in a degraded condition as a result of historic land use. Specifically, historical extirpation of beaver and livestock grazing practices caused a reduction in riparian vegetation cover, which led to increased erosion potential and subsequent channel incision. The little remaining riparian vegetation consists of senescing cottonwood trees that are primarily situated on the inactive floodplain surfaces. Overall, Pumpkin Creek within the project area is now an incised, single threaded, meandering stream. In-stream channel complexity is very low, and the stream is structurally starved (i.e., absences of wood, beaver dams and/or dense vegetation). Lateral connectivity under normal flood flows is limited to some small inset floodplains where channel widening has occurred.

The proposed restoration design is intended to improve riverscape health by promoting incision recovery in incised reaches by accelerating the geomorphic processes of widening and aggradation, and to increase aggradation and resultant lateral connectivity in reaches that have naturally begun to recover from previous incision. These objectives will be accomplished using low-tech, process-based restoration structures, including beaver dam analogs (BDAs) and post-assisted log structures (PALS). This restoration design outlines an initial project phase intended to treat approximately 8 miles of stream within a parcel of land managed by the Bureau of Land Management (BLM). We designed 16 complexes with 10-20 structures each. While we anticipate these structures to produce results rapidly after high flows, additional restoration phases will likely be necessary to continue to push the stream along the recovery trajectory.

M P

K

IN

C R

E E

K R

E S

T O

R A

T

IO

N D

E S

INTRODUCTION

The Bureau of Land Management (BLM) Montana/Dakotas is seeking to improve riverscape health across the lands it manages using low-tech, process-based restoration (LTPBR) methods. Eventually, each of the field offices will identify locations for, and implement LTPBR projects on riverscapes throughout the management area. As part of this effort, the Miles City field office is conducting a pilot project with assistance from Utah State University and the Riverscapes Consortium on a section of Pumpkin Creek.

Pumpkin Creek is a first order tributary of the Tongue River located approximately 11 miles south of Miles City, Montana. It is a perennial prairie stream, located entirely within the Northwestern Great Plains ecoregion (level 3 EPA Ecoregions). This region is characterized by erratic precipitation. Stream base flows are therefore relatively low, and peak flows are driven by spring rains and are highly variable. Like much of the region, the valley bottom of Pumpkin Creek was extensively grazed by cattle, resulting in a reduction in riparian vegetation cover. Due to this reduction, there is currently almost no large wood in the channel. Beaver dams are absent from the watershed as a result of historic extirpation and a lack of current resources to support them. Without these structural elements, the channel is greatly simplified, and in most locations consists of a deeply incised single-threaded channel lacking geomorphic diversity.

To guide restoration, we assume that the reference condition for a stream in this region would consist of a multi-threaded channel with high lateral connectivity across the valley bottom, driven in large part by structural forcing (Cluer & Thorne, 2014). Structural forcing in these types of systems would be driven by abundant riparian vegetation, in this case primarily cottonwood, as well as by beaver dams. Restoration here aims to move the riverscape along a trajectory of recovery toward this reference condition. This report outlines a restoration design for the initial phase, which will treat the eight miles of lower Pumpkin Creek located on BLM land, as well as laying out the process for carrying out additional restoration phases. The design follows planning, implementation, and project management guidelines identified by the Natural Resources Conservation Service’s (NRCS) Conservation Planning Process built within an adaptive management framework. This report provides an overview of the project location, restoration goals and objectives, an assessment of resources, a restoration design that includes estimated structure types, quantities, and locations, an assessment of potential risks to infrastructure, and an overview of adaptive management for the project.

PROJECT LOCATION AND CONTEXT

The Pumpkin Creek watershed is a sub-watershed within the Lower Tongue HUC8 watershed. The watershed encompasses approximately 700 mi2 of rolling plains with occasional buttes and badlands. Much of the basin is rangeland with mixed grass prairie cover, with additional areas of sagebrush steppe and some ponderosa forest cover at the highest elevations. The highest elevation in the basin is 4,412 feet (1,345 m), and the elevation at the basin outlet is 2,440 feet (744 m). The basin receives an annual average of 14 inches (356 mm) of precipitation. The restoration project site is located on the downstream end, approximately 4.5 miles above its confluence with the Tongue River (Figure 1).

M P

K

IN

C R

E E

K R

E S

T O

R A

T

IO

N D

E S

Figure 1 - Location of restoration project on Pumpkin Creek.

The BLM owns approximately four miles of Pumpkin Creek’s valley bottom, containing nearly 8.5 miles of channel (Figure 2). This creates the opportunity to implement restoration along a large portion of stream, addressing the degradation at a scale at which riverscape health can be meaningfully improved. Additionally, the restoration of important processes such as wood recruitment may eventually result in improvements to riverscape health outside of the project bounds as well. A pilot phase associated with a planned restoration workshop will begin by constructing complexes on the downstream end of the project area.

M P

K

IN

C R

E E

K R

E S

T O

R A

T

IO

N D

E S

Figure 2 - Land ownership within the Pumpkin Creek watershed. The panel on the right is zoomed in over the area the BLM owns where this restoration design will be applied.

PRELIMINARY PROJECT GOALS AND OBJECTIVES

The overall goal of restoration on Pumpkin Creek is to promote natural, self-sustaining fluvial processes that create and maintain a healthy, resilient riverscape. Doing so will increase habitat quality for the native fish species as well as terrestrial species that depend on riparian habitats. Some specific goals associated with the initial phase of this project that are necessary to achieve the broader goal here are:

o Increase the proportion of active valley bottom o Increase lateral connectivity o Expand riparian cover along the channel and within active floodplain o Increase channel/habitat complexity

Later in the planning process here these objectives are revisited within the context of indicators that can be used to evaluate restoration effectiveness at addressing these goals.

PLANNING DESIGN AND APPROACH

The lower Pumpkin Creek riverscape restoration design should follow an adaptive management framework that has three phases: 1) Collection and Analysis (focused on planning), 2) Decision Support (design), and 3) Application and Evaluation (implementation, monitoring, and additional phases as needed; Figure 3). In this report, the planning process includes components specific to riverscape restoration that are consistent with LTPBR designs and practices with the overall intent of presenting the preliminary restoration goals and objectives of the project, conducting resource assessment, risk, and recovery assessment, using those results to refine/recast the goals and objectives of the conceptual design, and arrive at measurable indicators to evaluate progress toward objectives (Wheaton et al. 2019).

M P

K

IN

C R

E E

K R

E S

T O

R A

T

IO

N D

E S

Figure 3 - Outline depicting an adaptation of NRCS’s Conservation Planning Process used to guide the restoration planning and design process (from Wheaton et al.

2019).

Low-tech process-based restoration

LTPBR is based on a set of riverscape and restoration principles that are applied based on the characteristics and limitations set by individual riverscapes (Appendix A). The first question we seek to answer before developing a LTPBR design is “is the riverscape structurally starved?” Structural starvation (i.e., the absence of wood, beaver dams and/or dense vegetation) in riverscapes is one of the most common impairments affecting riverscape health. Generally, a structurally starved riverscape drains quickly, has limited lateral connectivity, is more prone to incision, and has simple and homogenous habitat. By contrast, a riverscape with natural amounts of structure creates flow obstructions that lead to structurally-forced hydraulic diversity and geomorphic diversity, creating a more resilient riverscape that provides diverse habitat and a suite of ecosystem services.

M P

K

IN

C R

E E

K R

E S

T O

R A

T

IO

N D

E S

LTPBR approaches use the addition of structural elements to mimic, promote, and sustain natural riverscape processes. Rather than trying to create a specific channel form, implementation of LTPBR relies on stream power (or beaver) to “do the work.” LTPBR explicitly acknowledges that one treatment of structural elements is unlikely to reverse long periods of management impacts and that successful restoration will likely require multiple treatments (i.e., phases).

Therefore, LTPBR designs include phases, and work best when projects are monitored in order to determine when new phases or maintenance are required. The following design is presented within an adaptive management framework to incorporate monitoring and phased implementation in a transparent and structured plan (Figure 3).

Rationale for design

The project area on Pumpkin Creek is a suitable candidate for LTPBR for various reasons. First, the project area land is owned by the BLM and overseen by the Montana-Dakotas region, who has the vision of restoring health to riverscapes throughout the lands they manage. The stream is a suitable size for use of LTPBR structures, and has unregulated flow, which allows for a variable hydrograph with a variety of flood flows which can perform the geomorphic work associated with restoration. There are also limited infrastructure concerns within the project area.

RESOURCE ASSESSMENT

Valley setting

Pumpkin Creek is a single threaded, laterally partly confined to unconfined stream that flows through a valley ranging from roughly 200 – 800 m wide (Figure 4). The average valley gradient is approximately 0.25%. We delineated the valley bottom, comprised of the active channel(s) and active and inactive floodplain, because it represents the maximum extent that can be influenced by stream restoration and provides information regarding current conditions. We delineated the valley bottom using a combination of digital elevation models (DEMs): one based on LiDAR (Light Detection and Ranging) available from the Montana Lidar Viewer webmap with 1 m resolution, which covers the lower part of the reach, and a 10 m National Elevation Dataset DEM from the USGS. We also used several sources of nationally available aerial imagery to delineate active versus inactive floodplain.

Delineating the valley bottom and identifying the active channel(s) and active and inactive floodplain components can be used to 1) identify potential risk associated with restoration and 2) aid in developing realistic goals for restoration.

Figure 4 - Pumpkin Creek and its valley bottom along the portion of riverscape owned by the BLM.

M P

K

IN

C R

E E

K R

E S

T O

R A

T

IO

N D

E S

We classified the valley bottom into active channel, active floodplain, and inactive floodplain (Figures 5 and 6). The active channel is the area reworked by fluvial processes by 1-2-year recurrence interval flows (i.e., typical flood). The active floodplain is the area that is inundated during 5-10-year recurrence interval flows and generally capable of supporting riparian vegetation. The inactive floodplain is the area that is currently not influenced by flows except during rare high flow events. Inactive floodplain areas tend to support upland species such as sagebrush, while active floodplain areas tended to support riparian species such as willow. In Pumpkin Creek, 10% of the valley bottom is comprised of the active channel and floodplain, and 90% is inactive floodplain. As the vast majority of Pumpkin Creek’s floodplain is inactive, there is significant potential for restoration to convert inactive floodplain areas to active floodplain, capable of supporting riparian vegetation.

Figure 5 - Delineation of Pumpkin Creek’s valley bottom into geomorphic units. The square outlines the area where the project will begin in conjunction with a restoration workshop.

Figure 6 - The valley bottom of the restoration workshop area (square in figure 5) of Pumpkin Creek delineated into geomorphic units underlain by a lidar derived hillshade raster.

Two reaches have some distinct inset floodplains, while the remainder is incised with no floodplain.

M P

K

IN

C R

E E

K R

E S

T O

R A

T

IO

N D

E S

Geomorphic condition

We classified the project area into nine reaches based on their geomorphic condition (Figure 7). Reaches classified as “intact” would be characterized by multiple channels, greater geomorphic complexity, and more extensive riparian areas. “Good” condition reaches are characterized by a lack of channel incision, and therefore capable of supporting more abundant riparian vegetation. “Moderate” condition reaches are characterized by channel incision that limits channel-floodplain connectivity, but may have an inset floodplain capable of supporting some riparian vegetation.

“Poor” condition reaches are characterized by high levels of incision, resulting in complete disconnection between the channel and floodplain. No reaches in the project area were classified as good or intact. Five reaches are in moderate geomorphic condition, and four (longer reaches) are in poor condition.

Figure 7 - Reaches within the project area with reach breaks based on geomorphic condition overlain on a 10 m DEM.

The inset photos illustrate reach conditions at different parts of the stream network.

M P

K

IN

C R

E E

K R

E S

T O

R A

T

IO

N D

E S

Channel characteristics

Throughout the project area, Pumpkin Creek is deeply incised, though the areas in moderate condition have begun widening and have developed some inset floodplains. The banks are generally steep and erodible, with little vegetation cover. Overall, there is very little riparian vegetation. The channel substrate consists mostly of fine material derived from the highly erodible shales, siltstones and sandstones of the basin. Occasional pockets of gravel are primarily deposited on bar surfaces. The lack of riparian vegetation leaves the channel structurally starved; within the project area there are no large wood accumulations in the channel.

Hydrology

The basin that Pumpkin Creek drains is semi-arid prairie, receiving only 14 inches of precipitation per year, most of which occurs in spring and summer. Baseflow is typically 0 - 5 cubic feet per second (cfs). Peak flow typically occurs in spring; however, it occasionally occurs during summer or early fall as a result of strong thunderstorms. The magnitude of annul flow is highly variable from year to year (Figure 8). The USGS has a gage near the downstream end of the stream that has been operational for two periods of time. The largest flood on record occurred in 2011 (~ 7,900 cfs), inundating the valley bottom despite the deep channel incision, which generally disconnects the floodplain from the channel. The lowest peak flow on record was 36 cfs in 1980. Most years, peak flow is sufficient for structural forcing from LTPBR channel elements (PALS and BDAs) to initiate geomorphic processes of erosion and deposition.

Figure 8 - The USGS peak flow record for Pumpkin Creek just downstream of the project area demonstrates the high year to year variability in peak stream flow.

Riparian assessment

Riparian vegetation along the channel and on inset floodplains is predominantly herbaceous. Sparse remnant cottonwoods are found in some areas on the inactive floodplain surface (Figure 9). As flooding of these inactive surfaces is very rare, these trees are rarely watered by overbank flows, and fresh sediment deposits for recruitment of new trees are essentially nonexistent. Additionally, the distance to groundwater has increased due to channel incision.

Because of this, the existing trees/stands are senescent. For the processes that LTPBR seeks to jump start to become self-sustaining, natural sources of wood are necessary. Therefore, one of the primary objectives of the first phases of restoration is to increase the area of inset floodplains and to expand the extent of woody riparian vegetation within these floodplains. The inactive floodplain is covered in grasses and sage brush typically characteristic of uplands.

M P

K

IN

C R

E E

K R

E S

T O

R A

T

IO

N D

E S

Figure 9 - This drone imagery illustrates the typical riparian composition of Pumpkin Creek.

Herbaceous vegetation is found in the low-lying areas near the channel with occasional cottonwood and upland vegetation like sagebrush on the inactive floodplain.

Outputs of the Beaver Restoration Assessment Tool (BRAT) illustrate the riparian resource limitation. Current estimates for the project area to support dam building activity show rare (1 dam/km) to occasional (1-5 dams/km) capacity, whereas estimates of historic capacity range from occasional (1-5 dams/km) up to frequent (5-15 dams/km). Based on this departure, the baseline restoration guidelines produced by the model suggest that this part of Pumpkin Creek is suitable for “long-term investment” restoration, with some small areas of “quick return” possibly where inset floodplains are more developed (Appendix D).

Figure 10 - Infrastructure within the project area valley bottom that should be considered in implementing restoration

Risk assessment

Within the valley there are limited infrastructure concerns. Highway 59 runs parallel to the reach to its south, but remains on terraces outside of the valley bottom, except for a small portion at the upstream end. Some areas in the inactive floodplain are used by the BLM to plant grains to provide pheasant habitat in an agreement with Pheasants Forever. There is one culvert recently placed by BLM under an unimproved road with the

M P

K

IN

C R

E E

K R

E S

T O

R A

T

IO

N D

E S intent of flooding a floodplain wetland are during overbank flows. There are, additionally, several unimproved roads and old earthen berms in the inactive floodplain. These constraints these features create, however, are limited, as the BLM will be engaged in the adaptive management process. This will allow them to remove or alter structures as needed if they threaten infrastructure they wish to maintain, or allow structural forced changes to affect infrastructure where it is of less concern.

Potential future condition

Prior to human alteration, many riverscapes such as Pumpkin Creek were characterized by multiple channels and high channel-floodplain connectivity, and were, as a result, more resilient to disturbance. The stream evolution model presented by Cluer and Thorne (2014) describes valley bottoms characterized by multiple channels and high channel-floodplain connectivity as “Stage 0”, and describes how the hydrologic, hydraulic, substrate, geomorphic, and ecological benefits of this stage are greater than other stages in the stream evolution cycle (Figure 11). This concept, when applied to unconfined valley reaches like Pumpkin Creek provides an overarching target for restoration and potential pathways of recovery.

Without active structural additions it would likely take Pumpkin Creek centuries to naturally recover to something near a Stage 0 condition. With targeted restoration actions, there is potential to access much of the valley bottom throughout the project area based on the absence of infrastructure. However, because of its highly degraded condition, achieving full recovery, i.e., the stage at which restoration objectives have been achieved and the processes that maintain riverscape health are self-sustaining, multiple stages of restoration will likely be required. We do however, anticipate that the initial restoration actions associated with this design will produce immediate benefits and kick start the system toward a trajectory of recovery.

Figure 11 - Stream evolution model (SEM) proposed by Cluer and Thorne (2014) illustrating approximate stages and pathways associated with recovery to stage 0. Restoration in Pumpkin Creek is intended to accelerate recovery trajectories.

M P

K

IN

C R

E E

K R

E S

T O

R A

T

IO

N D

E S

Restoration indicators

To track success in achieving project objectives, a series of restoration targets and indicator metrics are recommended for evaluating the effectiveness of restoration. For each indicator, estimates of current and potential (i.e., target) values have been developed that correspond to broad recovery timelines (Table 1). All metrics are intended to be summarized through monitoring efforts using methods such as those described within the LTPBR Implementation and Monitoring Protocol (Weber et al., 2020). These methods allow quantification of indicator metrics via orthoimagery acquisition using a consumer level drone, or through measurements taken during rapid field surveys. Metrics should reflect progress toward the specific project objectives, such that they provide information that can be used within the adaptive management approach to ensure that restoration actions are leading to the desired outcomes. Here we chose several indicators that relate to the stated project objectives (Table 1).

Restoration Indicator Metrics

Percent active valley bottom – the percent of the total valley bottom that is inundated under 2-5-year return interval floods. This is expected to increase as restoration actions widen incised channels and increase lateral connectivity with currently active and inactive floodplain surfaces.

Percent inundated at base flow – the percent of the valley bottom that is inundated at baseflow (or percent that is considered active channel). This is anticipated to increase as structural elements pond water, widening the channel, and as secondary channels form and are activated due to erosion and activation of floodplains.

Percent inundated at typical flood – this indicator is similar to percent active valley bottom, however, some of the restoration actions included in this plan are aimed at inundating portions of previously inactive floodplain at more typical (1-2-year return interval) floods.

Woody riparian vegetation extent – the percent of the valley bottom covered by woody riparian vegetation. This is currently ~1%, mostly on inactive floodplains. An increase in extent will likely require targeted riparian plantings in association with the LTPBR actions planned here.

Jam density – the number of large wood jams per kilometer of stream. There are currently no jams in the stream.

Initially, the increase in this number will simply represent additions as a direct result of restoration actions. However, it is anticipated that over the long term as woody riparian vegetation recruits and matures, the process of wood recruitment and accumulation in the channel will become self-sustaining, increasing this metric without the need for intervention.

Dam density – the number of beaver dams or BDAs per kilometer of stream. Like jam density, this metric will directly reflect intervention by restoration actions until the processes become self-sustaining (i.e., until woody riparian cover increases and beaver are translocated and can maintain dam complexes).

Pool frequency – number per 100 m of in-channel concave geomorphic units (Wheaton et al., 2015; e.g., pools) created by erosion, and/or damming. Expected to increase in response to structural treatments. Pool habitat is typically important refuge for fish species.

Bar frequency – number of per 100 m of in-channel concave geomorphic units (Wheaton et al., 2015; e.g., point bars, mid-channel bars). Expected to increase as a result of the structural intervention due to increased hydraulic diversity.

M P

K

IN

C R

E E

K R

E S

T O

R A

T

IO

N D

E S

Table 1 – Current and target indicator metrics and their link to specific project objectives for the project area. Current values are estimates from field recon and imagery, and may be updated with additional information. Ranges in metrics reflect uncertainty in the timeline and outcomes from restoration treatment. These ranges, too, can be adjusted as the adaptive management process informs ongoing treatments as phases beyond the pilot phase are implemented.

Status Target Metrics

Indicator Current As-Built Short-Term 2 - 5 years

Medium - Term 5 - 10 years

Long-Term 10+ years

Objective 1: Increase proportion of active valley bottom

% active valley bottom 8 - 12% 8 - 15% 10 - 20% 15 - 40% 20 - 60%

Objective 2: Increase lateral connectivity

% inundated at base flow 3% 4 - 7% 4 - 8% 5 - 10% 5 - 15%

% inundated at typical flood 8% 8 - 15% 10 - 20% 10 - 30% 20 - 50%

Objective 3: Expand riparian cover

Woody riparian vegetation extent

~1% 2 - 10% (with plantings)

5 - 15% (with plantings)

10 - 25% (plantings + natural)

20 - 50%

Objective 4: Increase habitat complexity

Jam density (jams/km) 0 / km 15 - 25 / km

15 - 25 / km 15 - 30 / km 15 - 30 / km

Dam density (dams/km) 0 / km 3 - 8 / km 3 - 8 / km 3 - 10 / km 3 - 15 / km

Pool frequency (pools/100m) ? ? 1 - 3 / 100m 2 - 6 / 100m 4 - 8 / 100m

Bar frequency (bars/100m) 0 - 2 / 100m

0 - 2 / 100m

1 - 4 / 100m 3 - 8 / 100m 4 - 10 / 100m

Number of active channels 1 1 1 - 3 2 - 4 2 - 5

RESTORATION DESIGN

The LTPBR restoration design consists of the following components used to guide the implementation of treatments over time:

Temporal Design – The temporal design is used to guide initial and subsequent implementation phases (i.e., temporally punctuated structural treatments inclusive of new structures, maintenance, and structure enhancement).

Spatial Design – Reach Delineation – Restoration reach delineation based on factors such as valley setting and geomorphic condition. The delineation of reaches is used to set specific objectives and adjust restoration expectations according to limitations set by the riverscape.

Structural Elements and Complex Design – Description of structure types and their organization, distribution, and function within structure complexes (i.e., groups of multiple structures).

M P

K

IN

C R

E E

K R

E S

T O

R A

T

IO

N D

E S

Temporal design

Temporal design should take into consideration both the expectations for flood events of a given magnitude, as well as rates of vegetative, geomorphic, and hydrologic recovery. The restoration design, therefore, takes a phased approach to implementation in order to facilitate the adaptive management process. We recommend starting with a pilot phase in selected reaches (which in this case can begin with the restoration workshop), followed by implementation in the entire project area (Phase 1). A second structural treatment (Phase 2) would follow after at least 1-2 typical (e.g., 2-year return interval) flow events. A third treatment phase would take place after several moderate floods and at least one large flow (>5-year return interval). Additional phases could be added based on progress towards restoration targets and/or establishing self-sustaining processes. A phased approach is also beneficial in that it enables implementers to work out initial logistics at a smaller scale and subsequently scale up restoration more efficiently, while in the meantime training and building a local workforce. The phased approach also fits an iterative process that can be applied to multiple ongoing restoration projects over large spatial scales. Importantly, the specific timing of additional treatments, while likely to correspond to the timeframes listed above are in practice driven by adaptive management, and progress towards meeting restoration objectives.

Table 2. Estimated time table for phased implementation in Pumpkin Creek. Structure estimates are approximations. The number of new structures and those that need maintenance ins subsequent phases should be assessed through the adaptive management process. Structure estimates should be updated as the project is expanded beyond the pilot phase, and designs for these additional areas are completed.

Phase Year(s) Restoration Actions Structure Estimate

▪ Pilot restoration in lower reaches associated with workshop

▪ Implement restoration throughout the rest of the project area

▪ Structure maintenance and additions in project area

▪ Any additional restoration measures (e.g., riparian planting) in project area

New: 194 – 276

New and maintained: 194

- 276

2 - 5

▪ Evaluate Phase 1 restoration

▪ Structure maintenance and additions within project area

▪ Additional restoration measures (e.g., riparian plantings) throughout project area

5 - 10

▪ Evaluate Phase 2 restoration

▪ Structure maintenance and additions within project area

▪ Additional restoration measures (e.g., riparian plantings) throughout project area (if necessary)

Additional

10+

▪ Evaluate the establishment of self-sustaining processes

▪ Potential beaver reintroduction

Spatial design – reach delineation

As part of the resource assessment, the project area was broken into reaches based on geomorphic condition. Due to this varying condition, restoration design objectives vary between reaches. This distinction between reaches guides the process for designing complexes, which are groups of structures designed to accomplish specific restoration objectives. The spatial arrangement of different reaches provides important contextual information, because to meet certain objectives in downstream reaches (e.g., aggradation), specific actions upstream may be required (e.g., promoting bank erosion to mobilize sediment that can be captured in structures downstream). In Pumpkin Creek, poor condition reaches are highly incised and narrow. Restoration objectives for complexes in these reaches are intended to accelerate stream evolution by promoting processes of channel widening and aggradation. In moderate condition

M P

K

IN

C R

E E

K R

E S

T O

R A

T

IO

N D

E S reaches, riverscapes are further along within the stream evolution model; the channel is wider and inset floodplains are present. Restoration objectives in these complexes is therefore focused on continued aggradation and overbank flooding to increase lateral connectivity with floodplains.

Structural elements and complex design

Structural Elements

Structural elements proposed in the design include Beaver Dam Analogues (BDAs) and Post-Assisted Log Structures (PALS). These structure types can be constructed using a variety of locally sourced materials where available (from adjacent floodplains and hillslopes or nearby forest management activities) and installed using manual labor that will result in minimal to no impact to existing riparian vegetation and habitat. Appendix C provides details on BDA and PALS construction methods, different structure types, how different structure types should be used to promote specific responses, and design schematics.

Post-Assisted Log Structures (PALS)

PALS are composed of woody material of various sizes secured with untreated wooden posts driven into the substrate and positioned to mimic natural wood accumulations. PALS are generally designed to increase geomorphic diversity, force lateral channel migration, force overbank flows, encourage widening, and encourage aggradation and channel avulsion (Figures 20-23; Appendix C). However, PALS can also be built on the floodplain and disconnected side-channels in anticipation of floodplains being reactivated. There are three basic types of PALS: bank-attached, mid-channel, and channel spanning. Bank-attached PALS are used to widen channels, recruit sediment, promote scour pools, and build bank-attached bars. Mid-channel PALS are used to split flows, build mid-channel bars, scour pools, and recruit sediment. Channel-spanning PALS are used to force aggradation, promote overbank flow during high flow, and promote plunge and dam pools. Different types of PALS are often used in combination with beaver dam analogues to produce a variety of localized geomorphic effects. PALS are typically built in high densities (3-5 PALS/100m) such that if a PALS is blown out, woody material is likely to be captured by downstream structures (i.e., safety in numbers restoration principle; Appendix A).

Beaver Dam Analogues (BDAs)

BDAs mimic the form and function of natural beaver dams. BDAs are temporary, permeable structures built with or without posts using a combination of locally available woody material and sediment (Figures 25-30; Appendix C). The design and implementation of BDAs is a simple and cost-effective method to restore the processes that are responsible for physically complex channel and floodplain habitat. They can be used to support existing populations of beaver by increasing the stability of existing dams, create immediate deep-water habitat for beaver translocation, or used to promote many of the same processes driven by natural beaver dams such as increased channel-floodplain connectivity during both high and low flow conditions, increased groundwater recharge, expansion of riparian vegetation and wetland areas, increased hydraulic diversity including deep-slow water habitat, and incision recovery through channel widening and aggradation.

Complementary Restoration Priorities

While it is not specifically addressed in this restoration design document, woody riparian plantings will likely be a crucial part of restoration success in Pumpkin Creek as existing riparian coverage is extremely limited. In order for the processes that the restoration treatments are seeking to mimic and restore to become self-sustaining, a natural source of woody material must be present in the future. The combination of restoration treatments and plantings will create diverse conditions for woody riparian vegetation to recruit and expand by attenuating flows, creating depositional environments, and elevating water tables.

M P

K

IN

C R

E E

K R

E S

T O

R A

T

IO

N D

E S

Figure 12 - Locations of the 16 complexes identified along Pumpkin Creek in the project area and their primary objective.

Complex design

While individual structures (PALS and BDAs) may have local influence, they are unlikely to achieve project restoration objectives unless they are coordinated in a larger reach scale effort.

Thus, individual structures are designed to work together in complexes. A complex may be composed of a single structure type or a mix of structure types, and be composed of as few as two structures or as many as 10s of structures. Individual PALS and BDAs that are part of a complex help to increase the stability of any given structure within the complex. We have identified a series of 16 complexes designed to meet multiple objectives. The objectives for each complex are aligned with the overall restoration objectives for the project. Figure 12 provides an overview of complex locations along the project area. Table 3 provides a list of objectives for each complex and describes how they will be accomplished using various structure types and includes counts of structures. Following the table, additional figures of individual complexes show detail on types and locations of structures within the complexes.

Table 3 - Complex objectives, structure count and type. Structure counts are presented as a range to accommodate field-based changes to the design.

Complex no.

Primary Objective

Secondary Objective

Structure Count

Structure Types Design Narrative

1 Channel aggradation

Expand inset floodplains

15 - 20 PALS: bank-attached, channel-spanning, mid channel; BDAs:

primary and secondary

Bank-attached and mid channel PALS will drive bank erosion and channel widening; channel-spanning PALS and BDAs will encourage deposition of eroded sediment to aggrade the channel and build floodplains

2 Channel aggradation

Increase lateral connectivity

15 - 20 PALS: bank-attached, channel-spanning, mid channel; BDAs:

primary and secondary

Bank-attached and mid channel PALS will drive bank erosion and channel widening; channel-spanning PALS and BDAs will encourage deposition of eroded sediment to aggrade the channel and BDAs will direct flows onto existing floodplain surfaces.

M P

K

IN

C R

E E

K R

E S

T O

R A

T

IO

N D

E S

3 Channel widening

Develop inset floodplains

15 - 20 PALS: bank-attached, channel-spanning

The combination of bank-attached and channel-spanning PALS will "blast" the banks to drive erosion and widening, while channel spanning PALS will encourage aggradation of the eroded sediment.

4 Channel widening

Develop inset floodplains

17 - 22 PALS: bank-attached, channel-spanning, mid channel

The combination of bank-attached and channel-spanning PALS will "blast" the banks to drive erosion and widening, while channel spanning PALS will encourage aggradation of the eroded sediment.

5 Channel aggradation

Increase lateral connectivity

15 - 20 PALS: bank-attached, channel-spanning, mid channel; BDAs:

primary and secondary

BDAs in the upstream part of the complex will force flows onto low lying parts of the currently inactive floodplain; PALS in the downstream section will continue channel widening and development of inset floodplains.

6 Channel aggradation

Channel Widening

15-20 PALS: bank-attached, channel-spanning, mid-channel; BDAs:

primary and secondary

PALS will be positioned in order to enhance erosion at the bends in this reach, widening the channel and mobilizing sediment that can be deposited in the BDAs located interspersed between PALS.

7 Channel widening

Develop inset floodplains

8-12 PALS: bank-attached, channel-spanning

The combination of bank-attached and channel-spanning PALS will "blast" the banks to drive erosion and widening, while channel spanning PALS will encourage aggradation of the eroded sediment.

8 Channel widening

Develop inset floodplains

10-15 PALS: bank-attached, channel-spanning

The combination of bank-attached and channel-spanning PALS will "blast" the banks to drive erosion and widening, while channel spanning PALS will encourage aggradation of the eroded sediment.

9 Channel aggradation

Channel widening

10-15 PALS: bank-attached, channel-spanning;

BDAs: primary and secondary

Bank-attached PALS will enhance erosion at the bends in this reach, widening the channel and mobilizing sediment that can be deposited behind channel-spanning PALS and BDAs.

10 Channel aggradation

Channel widening

12-18 PALS: bank-attached, channel-spanning;

BDAs: primary and secondary

Bank-attached PALS in the upstream end of the reach will promoted erosion of banks, providing sediment for deposition in a series of BDAs and channel-spanning PALS below.

11 Channel widening

Channel aggradation

8-12 PALS: bank-attached, channel-spanning;

BDAs: primary and secondary

Bank-attached and channel-spanning PALS at a series of bends on the upstream end of the reach will promote bank erosion and channel widening, and provide sediment for deposition in a series of BDAs at the downstream end.

12 Channel aggradation

Increase lateral connectivity

12-18 PALS: bank-attached, channel-spanning, mid-channel; BDAs:

primary and secondary

Channel-spanning PALS and BDAs in the middle of this reach are designed to aggrade the channel and produce overbank flows which will direct water through a newly placed culvert to flood a floodplain wetland. Downstream PALS will provide sediment for downstream complexes.

13 Channel widening

Channel aggradation

10-15 PALS: bank-attached, channel-spanning;

BDAs: primary and secondary

The focus of structures in this complex is to enhance bank erosion to widen the incised channel; there is a pair of BDAs in encourage channel aggradation onto a low-lying bar.

M P

K

IN

C R

E E

K R

E S

T O

R A

T

IO

N D

E S

14 Channel aggradation

Expand inset floodplains

12-18 PALS: bank-attached, channel-spanning;

BDAs: primary and secondary

This complex is focused on using BDAs to aggrade the channel and create more connectivity with inset floodplains. PALS are placed at bends to enhance erosion and mobilize sediment.

15 Channel aggradation

Expand inset floodplains

8-13 PALS: bank-attached, channel-spanning;

BDAs: primary and secondary

This complex is focused on using BDAs to aggrade the channel and create more connectivity with inset floodplains. PALS are placed at bends to enhance erosion and mobilize sediment.

16 Channel aggradation

Expand inset floodplains

12-18 PALS: bank-attached, channel-spanning, mid-channel; BDAs:

primary and secondary

Mid-channel PALS will promote bar growth, and together with channel-spanning PALS continue development of inset floodplains. BDAs and channel-spanning PALS will aggrade the channel to increase connectivity with inset floodplains.

Figure 13 - Design for complexes 1 – 4

M P

K

IN

C R

E E

K R

E S

T O

R A

T

IO

N D

E S

Figure 14 - Design for complexes 5 – 8

M P

K

IN

C R

E E

K R

E S

T O

R A

T

IO

N D

E S

Figure 15 - Design for complexes 9 – 12

M P

K

IN

C R

E E

K R

E S

T O

R A

T

IO

N D

E S

Figure 16 - Design for complexes 13 - 16

See Appendix B for a table with individual structure types and locations.

ADAPTIVE MANAGEMENT

LTPBR is more appropriately thought of as an ongoing process of restoration and management rather than a ‘one-and-done’ effort. Here we discuss how adaptive management can be used to guide future phases of restoration. We use the term ‘phases’ here to refer to any restoration action taken, rather than when a specific restoration objective has been met. Adaptive management plays a major role in 1) evaluating the response to restoration through monitoring and 2) determining how the response to restoration guides future restoration actions (Figure 13). LTPBR projects can be evaluated at multiple scales, ranging from the scale of the individual structure to the entire project area. Here we

M P

K

IN

C R

E E

K R

E S

T O

R A

T

IO

N D

E S focus on the complex and project scale rather than the scale of individual structures, since project objectives are not met at the structure scale.

Figure 17 - Adaptive management for monitoring and ongoing restoration of LTPBR complexes. Many of the concepts illustrated are also applicable at the scale of an individual structure, or at the scale of an entire project. From Chapter 6 of Wheaton et al., 2019.

Common maintenance or phased restoration actions which necessarily occur at the scale of individual structures within a complex include:

▪ Lateral extension of structures through adding wood

▪ Increase structure height through adding wood

▪ Plugging gaps through adding more wood

▪ Adding posts to existing structures

▪ Repair minor breaches

▪ Building new structures

▪ Removing structures if causing harm

M P

K

IN

C R

E E

K R

E S

T O

R A

T

IO

N D

E S

The specific actions taken at an individual structure or location depend on the specific complex objectives and the specific structure objective within that complex.

An additional consideration in LTPR projects is that streams may have different pathways to recovery, or recovery trajectories, for a given starting condition. Incised streams may recover by going through a widening phase, leading to aggradation and eventual reconnection or by immediate aggradation and reconnection. It may be impossible to know what recovery trajectory is most likely for a given project area, or a specific reach within a project area. The goal of adaptive management is to be able to guide future management actions in the face of this uncertainty. Returning to Figure 11, there are essentially two trajectories of recovery: clockwise and counter-clockwise. Adaptive management can give guidance as to which trajectory a project is taking, and thus help direct restoration actions

Counter-clockwise Recovery Trajectory

In this recovery trajectory, the incised stream (e.g., Stage 3 in Figure 11) recovers by establishing processes that promote aggradation, which results in the establishment of channel-floodplain connectivity. Once established, subsequent phases of restoration can further increase the duration, frequency and extent of connectivity, pushing the stream from Stage 1 conditions into Stage 0 conditions in practice, it may take multiple treatments to fully reconnect highly incised stream to its floodplain.