A REPORT ON THE ECONOMICS OF FOREST RESTORATION
IN THE SIERRA NEVADA
 
December 1998 revised February 2002
 
by Dominic Roques
Roques Wildland Resources
roqueswild@vdn.com
805 550 7915
and
Tom Gaman
East-West Forestry Associates
415 669 7100
 
tgaman@forestdata.com
 
or
www.forestdata.com/sierra.htm
 

Background.
 
This report was prepared for and funded by a California non-profit which wishes to remain anonymous on our website.  Recognizing the importance and timeliness of this information, that group has authorized its release or consideration by the various groups who may find it useful. The report consists of the enclosed report, attached tables, and a number of watershed maps, showing land ownership, use, urbanization, fire probability, and impacts of mapped roads on mapped riparian systems in the 26 major Sierra watershed.   The information herein is solely the responsibility and the work of its authors and does not reflect an agreement with the opinion, policy, or position of any organization.

Tables and maps not found can be supplied on request.
 
 
 


TABLE OF CONTENTS

ABSTRACT
 
ISSUES SUMMARY

Summary Report with Maps in PDF format

PART I.   FOREST RESTORATION NEEDS ASSESSMENT

    1. Introduction and Objectives
    2. Assessing Forest Restoration Needs in the Sierra Nevada
        The Spatial Analysis
            River Basins and Subregions
        Experimental Design
            Land Protection Status
            Urbanization & Human Settlement
          Road Impacted Waterways
          Fire Return Intervals
             Forests
          Restoration Goals

    3. Forest Restoration Needs
        Land Protection Status
        Vegetation
        Impacts of Urbanization, Roads, Fire Suppression  
        Forest Restoration needs

PART II.   ECONOMIC EFFECTS OF FOREST RESTORATION
 
    4. The Sierra Nevada Economy
        Overview of population, income, poverty, employment
        Resource vs. amenity-based ecomonies
 
    5. The Economic Enterprise of Forest Restoration
        Costs, Accumulating Benefits and Types of Activities

    6. Financing Forest Restoration
        Existing and New Sources of Funding
        Market Based Approaches

Tables

Table 2-1: Twenty-Six Primary River Basins of the Sierra Nevada>

Table 2-1: Twenty-Six Primary River Basins of the Sierra Nevada

Table 2-2: GIS Data Analysis Procedure

Table 2-3: Vegetation Classifications for Data Analysis and Map Displays (Holland, 1986)

Table 2-4: Residential Density Classifications

Table 3-1: Land Ownership in the Sierra Nevada

Table 3-2: Acres of Forests in Sierra Nevada River Basins

Table 3-3: Settled Forestland

Table 3-4: Percentage Cover on Two Land-use Types Associated with Human Settlement

Table 3-5: Decreases in Canopy Cover at Varying Lot Densities

Table 3-6: Forest Roads in Riparian Zones in Selected Hydrologic Sub-Areas

Table 3-7: Forest Roads in Riparian Areas by Land Protection Status

Table 3-8: Acres by Fire Return Interval less than 250 years

Table 3-9: Acres by Fire Return and Land Protection Status

Table 3-10: Acres by Fire Return and Dominant Forest Type

Table 3-11: Summary of Forest Restoration Needs in Sierra Nevada

Table 4-1: Projected Population Growth for Sierra Nevada Counties

Table 4-2: Poverty Rates in Sierra Nevada Timber Counties

Table 4-3: Small Business in the Sierra Nevada

Table 4-4: Travel and Tourism-related Employment in the Sierra Nevada

Table 4-5: Annual Ecosystem Use Values for the Sierra Nevada

Table 4-6: Annual Hydroelectric Production and Value for Sierra Nevada River Basins

Table 5-1: Estimated Costs of Fuels Treatments

Table 5-2: Estimated Costs of Roads Treatments

Table 5-3: Estimated Costs of Immediate Fuel Management Needs

Table 5-4: Estimated Costs of Addressing Impacts from Roads in Riparian Areas

Table 5-5: Wages for Selected Occupations in Natural Resources Management

Table 5-6: Average Socioeconomic and Community Capacity Scores by Region

Figures

Maps


PART I. FOREST RESTORATION NEEDS ASSESSMENT
 
 

1. INTRODUCTION AND OBJECTIVES
 

The principal goal of this study is to explore the economic implications of pursuing restoration of Sierra Nevada forests. A spatial analysis that identifies and quantifies areas likely to be candidates for restoration is required for the economic assessment to be founded on factual information about forest conditions throughout this highly varied landscape.
 

We have identified three primary stressors on forest ecosystems which have impacted Sierra Nevada forests to the point where restoration is needed. These stressors include: road-building in forest riparian environments, suppression of natural fire regimes, and human settlement. Fire suppression and roading in forests are significant stressors associated with timber harvesting, as is the removal of trees. The effects of tree removal on the structure and composition of forests are not directly addressed here because accurate and timely information is not available for the whole region. Information on the current management of forest lands is included however, and is used to infer where harvesting has occurred.
 

Forest restoration is the focus of this study and should be distinguished from the broader term Òwatershed restoration.Ó A broad range of impacts and needs affects entire watersheds in the Sierra Nevada. Watershed restoration emphasizes the linkage between upstream and downstream conditions in aquatic environments and must examine many issues not addressed by this more focused analysis of forest restoration.
 

The spatial analysis conducted for this study required the assembly of a Geographic Information System (GIS) constructed from digital files obtained through the Sierra Nevada Ecosystem Project (SNEP) and other sources. This information supports the first spatially-explicit estimates of regional forest restoration needs ever reported and will significantly advance conservation management, planning and research efforts underway throughout the region.
 

Using the spatial data, we examined the costs of forest restoration, potential sources of funding for restoration, and the potential for restoration to stimulate local economies. These issues are addressed in the context of key Sierra Nevada economic indicators. The indicators reveal a region in transition toward greater economic diversification. The quality of life in the region supports the trend toward diversity by attracting the social capital that supports enterprise. Forest restoration will be essential to preserving the quality of life in the region.
 

2. ASSESSING FOREST RESTORATION NEEDS
 

When we examine large landscapes with an eye toward solving environmental problems, the watershed presents itself as a uniquely valuable unit on which to base a wide array of analyses. The simple reason being that the impacts of human activity on the land flow downhill and inevitably become apparent in the quality of aquatic and riverine environments. The integrity of forest ecosystems is without question a fundamental determinant of watershed health, since forests buffer the effects of natural and human-caused disturbances, regulate water flow, and protect water quality. Large watershedsÑthe river basins of the Sierra NevadaÑare used here to report forest restoration needs in the Sierra Nevada.
 

Human culture has delineated boundaries across the landscape which reflect the values of early inhabitants and the social and political structure of contemporary society. Similarly, other terrestrial mammals have ranges and habitat often irrespective of watershed boundaries. From the ecological perspective the watershed is a compelling way to approach environmental understanding and protection. But it is critical that we recognize human communities as artifacts of ecology as well, and that we are careful not to ignore community boundaries based on a perceived ecological imperative. In promoting a river basin perspective on Sierran forest restoration, this report recognizes the link between forest protection and water quality and aquatic ecosystem health.
 

This section of the report describes the approach we took to develop a spatially explicit assessment of forest restoration needs in the Sierra Nevada. The experimental design shows what specific questions we asked, what data we used, and how we manipulated them. The goals of forest restoration as they relate to a regional analysis such as this are also discussed.
 

2.1 Discussion of Spatial Analysis
 

The spatial analysis phase of this project was designed to objectively identify forests in need of restoration and offer an alternative to the anecdotal, popular, and disparate views on regional forest restoration priorities. There is a recognized need to expand the information sources used for conservation management, planning, and research beyond site-level characteristics to provide a consistent means for placing localized information into regional and range-wide contexts (Jennings, 1997). This analysis yields findings which complement, rather than substitute for, local assessments for forest restoration.
 

A spatially explicit and partially complete environmental information system for the entire Sierra Nevada was assembled for the Sierra Nevada Ecosystem Project (SNEP). Our analysis benefited from the SNEP effort and observed the boundaries established by that project for the Sierra Nevada Ecoregion (SNEP, 1996).
 

We constructed a Geographic Information System (GIS) from digital files obtained through the Sierra Nevada Ecosystem Project (SNEP) and other sources. A river basin template was selected for organizing the GIS and reporting findings. Subregions were further defined to group river basin findings. These subregions reflect both the natural and social character of different parts of the Sierra Nevada.
 

The principal management activities which impact forest ecosystems, include timber harvesting, fire suppression, road construction, and residential development. Our purpose was to identify the location and character of forests altered by these activities. The experimental design for characterizing these forests is based on the availability of range-wide data produced by SNEP scientists and other researchers. For example, forest areas in need of restoration from grazing impacts are not identified in this study, since no system-wide assessment of forest grazing effects has been performed. This analysis does not seek to identify areas to be targeted for protection, as does GAP analysis, but rather to describe forests that can be enhanced or preserved through active restoration.
 

Map: River Basins of the Sierra Nevada
 

2.1.1 River Basins and Hydrologic Subareas
 

The river basins we selected for analysis were derived from Hydrologic Areas and Hydrologic Sub-Areas in the CalWater classification system currently in use by public resource agencies (Map 1). CalWater is a set of standardized watershed boundaries, nested into larger previously standardized watersheds, meeting standardized delineation criteria (Brandow, 1994). Along the western side of the Pacific Crest, the river basins of the Sierra Nevada form the headwaters of the Sacramento-San Joaquin River Delta. South of the San Joaquin Basin, Sierran rivers feed the Tulare Lake Hydrologic Region which is an enclosed basin. The North and South Lahontan Hydrologic Regions drain the east side of the Sierra Nevada into extensive internally drained valleys in California and Nevada.
 

Only portions of the basins within the Sierra Nevada Ecoregion (SNEP, 1996) and within California are considered in this report. Several basins on the east side, including the Tahoe Basin are truncated by the State line. Hydrologic Sub Areas (HSAs), identified on maps with a five-digit number, are subwatersheds within each river basin, and are used to more precisely characterize restoration needs within a river basin.
 

The twenty six primary river basins of the Sierra Nevada, broken into five subregions in this report, range in size from 274,344 to 1,751,590 acres (Table 2-1). Eighteen Sierra counties intersect these river basins as illustrated in Map 2.
 

2.1.2 Subregions
 

Subregions were delineated based in part on the diverse natural character of the Sierra Nevada as well as on cultural features which give even greater distinction to more and less populated areas.
 

North

The North includes five river basins belonging to the Feather (North Fork, East Branch North Fork, Middle Fork), Sacramento (Mill, Big Chico, and Butte Creeks), and Susan (Honey and Eagle Lakes) Rivers. Lassen, Plumas and portions of Tehama and Butte Counties are located in this heavily forested region of the Sierra Nevada.
 

North Central

The North Central subregion includes the Yuba, Bear, American, and Consumnes River Basins west of the Pacific Crest, and the Truckee River and Lake Tahoe Basins on the east side. The Bear and North Fork American Rivers are combined in this analysis. Sierra, Nevada, Placer, El Dorado Counties, and a portion of Yuba County comprise the political landscape of the North Central Region. Heavily populated (more than 70% of all Sierra Residents) and possessing world renown rivers and lakes, this subregion experiences the greatest forest stress.
 

Map 2: Counties and River Basins of the Sierra Nevada
 
 
 
 

Table 2-1: Twenty-Six Primary River Basins of the Sierra Nevada
 

SUBREGION ACRES
NORTH
Mill-Big Chico-Butte Creeks 658,454
North Fork Feather 783,417
East Branch North Fork Feather 656,980
Middle Fork Feather 871,789
Honey-Eagle Lakes 1,422,459
4,393,100
NORTH CENTRAL
Yuba 842,731
North Fork American-Bear River 928,961
South Fork American-Consumnes 946,790
Truckee 274,344
Lake Tahoe Basin (CA only) 325,337
3,318,164
SOUTH CENTRAL
Mokelumne 506,179
Stanislaus-Calaveras 873,314
Tuolumne 1,055,720
Merced 699,089
3,134,301
SOUTH
Chowchilla-Fresno 597,713
San Joaquin 1,098,744
Kings 1,159,038
Kaweah 552,532
Tule 639,779
Kern 1,751,590
5,799,396
EAST SIDE
Carson 290,058
Walker 584,272
Mono Basin 430,589
Crowley Lake 1,188,401
Owens Lake 876,319
Mojave 609,566
3,979,206
 
TOTAL ALL RIVER BASINS 20,624,166
Source: CalWater, Version 2.0.
 

South Central

The South Central Subregion comprises the Mokelumne, Calaveras, Stanislaus, Tuolumne and Merced River Basins. Amador, Calaveras, Tuolumne, and Mariposa Counties are located in this subregion. While not as populated as the North Central Subregion, many of the same impacts to forests are found here, but more than a quarter of the subregion is protected in Yosemite National Park.
 

Southwestern

The largest subregion, Southwestern comprises six river basins, including the Chowchilla and Fresno, San Joaquin, Kings, Kaweah, Tule, and Kern. This rugged, more arid subregion is less forested and less populated than other west slope subregions. A significant portion of the subregion is in the Sequoia-Kings Canyon National Parks and adjacent wilderness areas.
 

East

Dropping abruptly from the Pacific Crest to the valleys of the Eastern Sierra, the East Subregion includes the California portions of the Carson River, Walker River, and Mono Basins, and all of the Crowley Lake, Owens Lake, and Mojave Basin. Alpine, Mono, Inyo and portions of Kern Counties constitute the political landscape of this subregion. Sparsely forested but possessing superlative alpine landscapes, the East Subregion confronts fewer pressures on its forests than the forests of other subregions.
 

2.2 Experimental Design and Data Quality
 

Table 2-2 describes the GIS data analysis procedure followed for this report. It identifies the function of data, the specific digital coverages used, how we manipulated or treated the data, the key map products, and the tabular data prepared by analyzing the digital coverages. Throughout this procedure we followed the conventions of GIS development and were constrained by the analytical flexibly inherent in ARCInfo and ARCView software for PC.
 

Table 2-2: GIS Data Analysis Procedure
 
data Function digital Coverage Treatment of data MAPS TABLES
Select forest areas At-Risk to management with Sierra Nevada 1. Land Mgmt/Ownership (Davis, Stoms, 1996); 

2. GAP Vegetation (Holland, 1986) 

3. Calveg (Parker, Mathias, 1977) 

4. SNEP Boundary  

5. CalWater Version 2 (CDF) 

6. USGS 7.5 min. Topo Index (USGS, 1993)

7. Maintain original 5 Land Mgmt/Ownership classes 

8. Aggregate forest veg into forest types for map (further aggregated into 6 classes for data analysis) 

9. Remove all large bodies of water using CALVEG water layer.  

10. Make consistent with SNEP boundary 

11. Select, modify, superimpose Hydrologic Sub Area (HSA) and River Basin boundaries 

12. Overlay USGS 7.5 minute topo index

River Basins; Land Protection Status; 

Forest Vegetation

Acres in each river basin; 

Acres in Land Protection Classes; 

Acres in Forests by basin

 
HUMAN SETTLEMENT
 
Evaluate forests affected by human settlement 1. Residential Density, 1990 (Duane, 1996) 

2. Forest Type

1. Reclassify by original 11 density classes (0 to >640- units/sq. mi) 

2. Select polygons for higher densities (20 to >640 units/sq. mi.) 

3. Intersect with forest types  

4. Assign coefficients for density, based on forest type (McBride, et al, 1996)

Residential Density, 1990 (acres/dwelling unit) Acres of forest type affected by human settlement 

(canopy loss)

 
ROAD-IMPACTED WATERWAYS
 
Identify riparian areas potentially affected by roads 1. Streams (USGS, 1993) 

2. Rivers (USGS, 1993) 

3. Lakes (USFS, 1:24K) 

4. Roads (USGS, 1993) 

5. CalWater Version 2 (CDF) 

6. Forest Type (Holland, 1986) 

7. Land Protection Status (Davis, Stoms, 1996)

8. Fix waterways buffer at 46 m (150 ft) 

9. Overlay roads onto buffered waterways 

10. Calculate road and lengths where adjacency occurs 

11. Overlay HSAs  

12. Overlay land protection status 

13. Overlay Forest Type

Roads within 150 feet of Waterways Miles of Road in Riparian Areas; Land Protection Status of Miles of Road affecting Riparian Areas; Forest type and Road effects.
 
FIRE SUPPRESSION
 
Identify candidate areas for high priority fuel management 1. Fire Return Intervals (Sapsis, et al, 1996) 

2. Forest Type (Holland) 

3. Land Protection Status (Davis, Stoms, 1996) 

4. CalWater Version 2 (CDF)

5. Overlay with Land Protection Status 

6. Overlay with Forest Vegetation 

7. Conduct analysis only for FRI 1-100 and 100-250 years in Basins and HSAs

Fire Return Interval Acres in FRI 1-100 and 100-250; FRI acres in dominant forest types; FRI acres by Land Protection Status
 
FOREST ASSETS
 
Identify where assets occur relative to restoration needs 1. LS/OG (Franklin, Fites-Kaufmann, 1996) 

2. Ecologically Significant Areas (Millar, et al, 1996) 

3. Sequoia groves (SNEP; USFS) 

4. CalWater Version 2 (CDF)

Giant Sequoia Groves; 

LS/OG ; 

Significant Natural Areas

 

2.2.1 Land Protection Status and Vegetation Base Coverages
 

Land Management And Ownership

Land management and ownership for the Sierra Nevada were described in SNEP based on work completed by Frank Davis and David Stoms in a collaboration between SNEP and the National Biological Service Gap Analysis Project (NBS GAP). Their classification is available for 73 percent of the SNEP core area. The databases used in the SNEP/NBS GAP analysis comprise the most spatially detailed land management maps ever assembled for the region as a whole (Davis, Frank W., Stoms, David M., 1996). These data were not modified for this report. Areas described in this report which are not covered by the Davis and Stoms analysis were assigned classifications based on land ownership which provides a more coarse description of management activity. Those areas with less detailed management information include the Mill/Big Chico/Butte Creeks Basin, a portion of the North Fork Feather River Basin, the Honey-Eagle Lakes Basin, and almost the entire East side below the Carson River Basin.
 

Vegetation in the Sierra Nevada Ecoregion

Vegetation types for the Sierra Nevada Ecoregion are derived from Holland (1986) which classifies vegetation based on overstory described by one to three species, each contributing more than 20% of the relative canopy. These species are classified into natural plant community types used by the California Department of Fish and Game, Natural Heritage Division. We aggregated Holland forest vegetation to simplify map displays and data analysis. We used greater detail (less aggregation) for the maps than for data analysis (Table 2-3). Vegetation maps are not included in this report because of budget limitations.
 

Vegetation for areas outside of the Sierra Nevada Ecoregion was derived from Calveg (Parker and Mathias, 1977) as provided in a current CDF digital file. We made some assumptions in translating the Calveg types into the Holland types, but we do not expect these assumptions to appreciably affect the results of our analysis.
 

2.2.2 Human Settlement Data
 

These data are 1990 census-based and are for residences onlyÑcommercial and industrial uses would increase the developed area significantly. Thus, the map provides a proxy for where urbanization is occurring in the Sierra Nevada. Housing density was calculated by dividing 100% housing count by the land area of the census blocks. We eliminated the low density classes (<20 dwelling units/square mile). The majority (89%) of housing throughout the Sierra Nevada occurs in densities of 20 units/square mile and greater (Duane, 1996). We chose a threshold value of 1du/32ac to distinguish wildlands from urban-intermix areas. Thus, the five classes used in this report represent only the densities found in urban and urban-intermix areas (Table 2-4). To our knowledge, the forest impacts of areas developed at lower densities have not been quantified by previous authors and are therefore not addressed in this report.
 

Densities are accurate to within 1,000 square meters. Census block boundaries should be within 51 meters of their actual position. Individual census blocks are aggregated into five classes starting with 20-40 dwelling units/square mile and ending with over 640 du/sq.mi. Errors are likely to exist in the largest and most heterogeneous census blocks. This is because census blocks are not evenly distributed across the region. For example, one census block may include areas with two very different densities, yet they are averaged across the block. The occurrence of these errors is low and most census block boundaries should contain relatively homogeneous units (Duane, 1996).
 

Table 2-3: Vegetation Classifications for Data Analysis and Map Displays (Holland, 1986)
 

Vegetation Categories for Data Analysis Vegetation Categories for Map
Oak Woodland Oak Woodland
Oregon Oak Woodland Oregon Oak Woodland
Black Oak Woodland Black Oak Woodland
Valley Oak Woodland Valley Oak Woodland
Blue Oak Woodland Blue Oak Woodland
Interior Live Oak Woodland Interior Live Oak Woodland
Broadleaf Forest Broad Leaf Forest
Canyon Live Oak Forest Canyon Live Oak Forest
Interior Live Oak Forest Interior Live Oak Forest
Black Oak Forest Black Oak Forest
Tan-Oak Forest Tan-Oak Forest
Aspen Forest Aspen Forest
Mixed Conifer Mixed Conifer
Ultramafic Mixed Coniferous Forest Ultramafic Mixed Coniferous Forest
Sierran Mixed Coniferous Forest Sierran Mixed Coniferous Forest
Lower Cismontane Mixed Conifer-Oak Forest Lower Cismontane Mixed Conifer-Oak Forest
Upper Cismontane Mixed Conifer-Oak Forest Upper Cismontane Mixed Conifer-Oak Forest
Westside Ponderosa Pine Conifer Woodland
Westside Ponderosa Pine Forest Open Foothill Pine Woodland
Serpentine Foothill Pine-Chaparral Woodland
Conifer Woodland Non-Serpentine Foothill Pine Woodland
Open Foothill Pine Woodland Foothill Pine-Oak Woodland
Serpentine Foothill Pine-Chaparral Woodland Juniper-Oak Cismontane Woodland
Non-Serpentine Foothill Pine Woodland Pinyon Pine-Oak Woodland
Foothill Pine-Oak Woodland Cismontane Juniper Woodland
Juniper-Oak Cismontane Woodland Oak-Pinyon Woodland
Pinyon Pine-Oak Woodland Northern Juniper Woodland
Cismontane Juniper Woodland Great Basin Pinyon-Juniper Woodland
Oak-Pinyon Woodland Great Basin Pinyon Woodland
Northern Juniper Woodland Great Basin Juniper Woodland and Scrub
Great Basin Pinyon-Juniper Woodland Mojavean Pinyon and Juniper Woodland
Great Basin Pinyon Woodland Mojavean Pinyon Woodland
Great Basin Juniper Woodland and Scrub Mojavean Juniper Woodland and Scrub
Mojavean Pinyon and Juniper Woodland
Mojavean Pinyon Woodland Ponderosa Pine
Mojavean Juniper Woodland and Scrub Eastside Ponderosa Pine Forest
Westside Ponderosa Pine Forest
Conifer
Knobcone Pine Forest Lodgepole Pine
Eastside Ponderosa Pine Forest Lodgepole Pine Forest
Sierran White Fir Forest
Big Tree Forest Jeffrey Pine
Modoc White Fir Forest Jeffrey Pine Forest
Jeffrey Pine Forest
Red Fir-Western White Pine Forest Jeffrey/Fir
Jeffrey Pine-Fir Forest Jeffrey Pine-Fir Forest
Red Fir Forest
Lodgepole Pine Forest White Fir
Whitebark Pine-Mountain Hemlock Forest Sierran White Fir Forest
Whitebark Pine-Lodgepole Pine Forest
Foxtail Pine Forest Red Fir
Bristlecone Pine Forest Red Fir Forest
Whitebark Pine Forest
Limber Pine Forest Red Fir/White Pine
Red Fir-Western White Pine Forest
Knobcone Pine
Knobcone Pine Forest
Big Tree
Big Tree Forest
Whitebark Pine Group 
Whitebark Pine-Mountain Hemlock Forest
Whitebark Pine-Lodgepole Pine Forest 
Whitebark Pine Forest
Foxtail/Bristlecone/Limber Pine
Foxtail Pine Forest
Bristlecone Pine Forest
Limber Pine Forest
 
 
 

Table 2-4: Residential Density Classifications
 

Density in sq. miles >> equivalent to >> Density in acres
20-40 du/sq.mi. (32 ac/du -16 ac/du)
40-80 du/sq.mi. (16 ac/du -8 ac/du)
80-160 du/sq.mi. (8 ac/du -4 ac/du)
160-640 du/sq.mi. (4 ac/du -1 ac/du)
>640 du/sq.mi. (<1 ac/du)
 

2.2.3 Road-Impacted Waterway Data
 

These data were derived from United States Geological Survey (USGS) 100,000 Scale Digital Line Graph (transportation and hydrology) data available from USGS, Menlo Park. The roads were subdivided into groups based on their DLG road labels (USGS, 1993).

· Major Highways: Interstate highways, U.S. Routes, State Routes, County Routes

· Primary and Secondary Routes: (undivided, divided by centerline, divided lanes separated, one way other than divided highway, and class 3 roads and streets)

· Undifferentiated Forest Roads: Class 4 roads and streets; Four-Wheel-Drive Trails.
 

We converted all the roads data to Albers projection and overlaid them onto the SNEP River Basin map and the Hydrologic Sub Areas (HSAs). Then we created a spatial buffer of 150 feet (46 meters) each side of the centerline of a lake, creek or river and analyzed them to determine which roads entered the riparian buffers. This analysis was then overlaid onto forest vegetation in the Sierra Nevada to examine the extent of the problem in forested areas only.
 

The resulting map indexes the possible extent of damage to forest riparian areas caused by roads. Where there is a high density of riparian intrusion by roads, we would expect a smaller network of skid trails to occur on lands managed for timber harvest, as well as Òghost roadsÓ known to exist but not on maps.
 

2.2.4 Data on Fire Return Intervals for Large Fires
 

These data show the expected annual frequency of large (300+ acres) fires on a grid of 10-acre cells as determined by Sapsis and others (1996) for the SNEP report. The areas depicted in the two classes with highest return frequency (1-100 years and 100-250 years) are areas that would be high priority candidates for the reintroduction of fire and other fuel management efforts.
 

The primary data from which these maps were derived are fire history records (39,986 fire records from the period 1981-93). They describe ignition history ratio of large fires to ignitions, and estimates of mean fire size within nine strata. Strata are based on life form (grass, brush, timber, red fir), National Weather Service fire weather zones, and population density class. The resulting fire data and map:

Òare perhaps more reliable for describing current and near term future risk. ...a paucity of fire incidence over substantial areas in the data period could translate into unreasonably low fire frequency estimates for these areas. One should assume that actual fire frequency in areas where fires were scarce in the data period may be higher than the map indicates...We are confident that the relative frequency and regional trends evinced by the map are a reflection of actual likelihood of large fire, and are hence useful information in interpreting risk across the study area.

--Sapsis, et al, 1996.
 

2.2.5 Data on Forest Assets
 

The Sierra Nevada Ecosystem Project was charged with identifying specific high value attributes of Sierra Nevada forests. Teams of scientists were assembled to provide assessments of Late Successional/Old Growth (LS/OG) forests, Giant Sequoia groves, and Ecologically Significant Natural Areas. The assessments were compiled as digital coverages and associated data bases for these important and rare forest features. We secured the digital information on these three types of forest assets from the Alexandria Web Site which serves as a clearinghouse for SNEP data. The maps we produced represent no manipulation of source data other than to overlay our hydrologic boundaries onto the original coverages.
 

2.3 Forest Restoration Goals
 

This assessment of forest restoration needs identifies impacts to Sierra Nevada forests that can be reasonably approached through local restoration strategies. These strategies could include road repair or obliteration, thinning of forests to permit the safe reintroduction of fire, and using fuel management techniques in settled areas that offer maximum protection of biodiversity. An assessment of restoration needs at the scale of actual interventions requires space- and time-specific restoration goals. For example, restoring the role of fire in a 500-acre catchment requires that actual fuel loads be determined, and that the natural fire return interval for that catchment be understood and targeted in pre-treatments (e.g. thinning from below), prescribed burns, and managed wildfires. Similarly, to restore the complexity and diversity of riparian habitat impacted by a forest road, the specific stream flows, needed to erode banks and deposit point bars which in turn permit lateral migration of meandering channels, would need to be estimated and provided. And finally, where the goal of restoring the forest to its condition prior to the arrival of European-American settlers may be appropriate in certain locations, such a target state is no longer feasible in a west slope forest with an extensive human community. From these examples it is clear that both spatial and temporal scales of a finer grain than possible in this regional analysis determine the range of actual treatments.
 

The assessment of forest restoration needs for the entire 20 million acres and 26 major river basins of the Sierra Nevada ecosystem requires that restoration goals be more general. At the regional scale, we define restoration as Òcoordinated actions designed to return an impacted ecosystem to a prior, more natural target stateÓ (Hrubes, 1997), and leave to those who implement restoration at the local scale, the designation of the specific target state and the natural processes to emphasize (e.g. burning, reducing ground fuels, natural pathogens). Restoration needs at the regional scale are then stated more generally in terms of location and quantity of acres of forest or miles of stream that require finer level analysis before specific treatments are selected. This regional analysis serves to focus the work of developing those specific strategies on the areas that need it most.
 

3. FOREST RESTORATION NEEDS BY RIVER BASIN
 

The following analysis of forest restoration needs in the Sierra Nevada specifically examines the effects of human settlement, roads located in riparian areas, and fire suppression. In each case we quantify the area affected based on information available for the whole region. The structure and forests in the Sierra Nevada have been altered by these activities as well as a broad range of other activities from timber harvesting to intensive recreation. The data presented here are therefore not a full accounting of forest impacts in the region. They do however represent the principal problem areas that can reasonably be addressed through the work of forest restoration, and in that sense can be taken as a basis for both prioritizing and for estimating the costs of this work.
 

3.1 Land Protection Status
 

Approximately 65 percent of the Sierra Nevada is under public ownership. The USDA Forest Service alone manages 42 percent of the entire Sierra Nevada ecosystem (Table 3-1). The type of management within each one of these ownerships determines to a great extent the condition of the forests and watersheds in the region. For example, the Bureau of Land Management and USDA Forest Service manage for a variety of uses that range from full wilderness protection to timber production and the condition of these lands reflect this.
 

Table 3-1: Land Ownership in the Sierra Nevada
 

LAND OWNERSHIP ACRES PERCENT OF TOTAL
Private 6,705,506 33%
State 190,594 1%
USDA Forest Service 8,672,735 42%
Bureau of Land Management 2,242,491 11%
National Park Service 1,644,225 8%
Other Public 521,208 3
Lakes 453,105 2%
Source: USDA, USFS.
 

Land Protection Status provides a more meaningful description of land management than simple land ownership can convey. Each class of Land Protection Status profiles both ownership and management and provides a crude measure of risk of development or resource over-exploitation (Davis and Stoms, 1996). Classes distinguish land based on permitted use and assume that the most pervasive land uses affecting the status and trends of terrestrial biodiversity in the Sierra Nevada are grazing, fire suppression, timber harvest, and urban, residential, and agricultural development. The existing management affecting forests of the Sierra Nevada ultimately dictates the options available for forest restoration. For example, restoration on private unprotected forest lands will likely require some mix of incentives and education to promote voluntary action by landowners, whereas restoration on public lands will require in some cases a shift in policy or an augmentation of funds to undertake it, or both.
 

The pattern of protection and ownership mirrors topography somewhat, with greater levels of protection and federal ownership at higher elevation, grading into more vulnerable unprotected private lands at lower elevation (Maps 3a-3e). Central and southern portions of the range have more protected lands than the north. Forests of the Yuba, Truckee, and North Fork American River Basins have a more complex arrangement of protection status, in part due to the checkerboard of public and private lands that remain as a legacy of the disposition of federal lands to transcontinental railroads (Map 4).
 

The five Land Protection Status classes are as follows (Davis and Stoms, 1996):

Class 1 Full Protection

Public or private land formally designated for conservation of native biodiversity and within which timber harvests are precluded. Natural disturbance events are generally allowed to proceed without interference or are mimicked through management. The areas may include national parks, national monuments, ungrazed lands within USDA Forest Service wilderness areas, research natural areas, and wild and scenic rivers, Blue Ridge National Wildlife Refuge, The Nature Conservancy preserves, and state parks and ecological reserves.

Class 2 Mostly Protected--no grazing

National forest land that is generally managed for its natural values but is not formally designated for conservation of native biodiversity. Development and grazing are excluded, and timber harvest is generally excluded because it conflicts with other multiple-use objectives. Wildfires are generally suppressed. The distribution of recreational activities on Class 2 lands is unknown, but a small fraction of the land is developed for recreational facilities.

Class 3 Mostly Protected--some grazing

Public land that is generally managed for its natural values, is treated in existing management plans as unsuitable for timber harvest, and may be grazed. Wildfires may be actively suppressed. Examples include grazing allotments within USFS wilderness areas, grazing allotments on national forest lands classified as unsuitable for timber harvest, the San Joaquin Experimental Range, Bureau of Land Management (BLM) areas of critical environmental concern, and BLM wilderness areas.

Class 4 Other Public Lands

Public lands not included in Classes 1-3, mainly multiple use federal lands managed by the Bureau of Indian Affairs (BIA), Bureau of Reclamation, BLM, and USFS. National forest lands in this category include areas that are classified in existing plans as suitable for timber harvest. These USFS areas can also be within existing grazing allotments. Wildfires are actively suppressed.
 
 

Class 5 Private and Unprotected

Private lands other than those in Class 1. In the absence of more detailed zoning data, we assume that these lands are potentially available for development, timber harvest, and grazing and that wildfires are actively suppressed.
 

3.2 Forest Vegetation
 

Regional forest vegetation maps convey an incomplete picture of a regionÕs forest in that they do not describe the actual ground covered by trees. Anyone who has visited the Sierra Nevada is aware of the great variability in tree density there. Maps assign a forest type to a vast landscape which may actually be treeless in areas up to several acres in size. This is an essential fact to consider when forest vegetation maps are used to quantify the extent of a problem such as the need for forest restoration.
 

The six forest types used in this report include: west-side ponderosa pine, mixed conifer, conifer woodland, other conifers, broadleaf forests, and oak woodlands (see section 2.2.1 for complete descriptions of forest types). Their distribution is shown in Table 3-2.
 

Table 3-2: Acres of Forests in Sierra Nevada River Basins
 

Westside Ponderosa Pine Other Conifer Broadleaf Forest Oak Woodland Mixed Conifer Conifer Woodland TOTAL Forest Vegetation
 
ACRES 1,000s ACRES
 
North Mill/Big Chico/Butte Creek 0 0 0 219,538 313,613 2,359 536
No. Fork Feather 47,351 94,160 22,468 11,465 457,767 11,581 645
E. Br. No.Fk.Feather 48,005 235,059 935 3,074 230,309 7,238 525
Middle Fork Feather 57,769 297,888 25,801 940 232,351 13,782 629
Honey-Eagle Lake 686 187,074 0 12,980 213,661 93,515 508
North Central Yuba 37,012 175,416 74,578 35,254 369,143 35,039 726
Bear River/ North Fork American 76,199 115,570 114,616 46,140 330,626 93,817 777
S.Fork American/Consumnes 190,236 115,466 39,898 64,583 259,910 109,669 780
Truckee 0 208,909 0 2,145 2,356 3,182 217
Lake Tahoe Basin  0 110,290 11 0 4,940 907 116
South Central Mokelumne 109,907 130,821 45,699 18,667 104,882 19,912 430
Stanislaus/Calaveras 89,515 202,055 27,965 103,886 169,879 131,172 724
Tuolumne 118,102 278,163 56,189 84,306 118,617 85,461 741
Merced 46,449 209,795 74,685 49,921 87,563 38,072 506
Southwest Chowchilla-Fresno 13,921 0 86,101 260,224 51,669 81,244 493
San Joaquin 95,802 384,598 39,539 65,530 167,974 65,252 819
Kings 60,418 380,831 48,754 159,754 128,130 46,240 824
Kaweah 21,446 81,261 48,712 243,845 69,972 560 466
Tule 27,820 36,256 32,391 338,256 66,301 27,664 529
Kern 75,548 483,344 33,194 110,519 121,613 443,526 1,268
East Carson 0 181,030 2,188 0 447 38,055 222
Walker 0 90,069 645 0 55,497 231,417 378
Mono Basin 0 45,731 560 0 15,136 84,922 146
Crowley Lake 0 201,885 3,307 0 1,602 237,789 445
Owens Lake 0 54,286 0 0 0 101,290 156
Mojave 0 8,405 1,883 11,197 4,735 185,505 212
13,815
Source: GAP Veg and UCB FTP site, based on Holland (1996), and Calveg, provided by CDF, based on Mathias, Parker (1977).

Note: See Section 2.2.1 for definition of forest types.
 

3.3 Impacts on Sierra Nevada Forests
 

3.3.1 Human Settlement
 

Human settlement in forest ecosystems results in a variety of effects on wildlife habitat, hydrology and fire behavior. In Sierra Nevada forests human settlement has resulted in a decrease in crown canopy cover, a reduction in tree density, and an introduction of exotic tree species (McBride, Russell, and Kloss, 1996). The decrease in crown canopy cover is examined here to infer effects on fire hazard, hydrology, and wildlife habitat value that could be addressed through forest restoration.
 

Human settlement affects fire protection costs and losses by changing fire risk, fire hazard, and exposure of high value forest assets. For example, increases in population, automobile traffic, and recreation come with increases in the frequency of human-caused fires. Also, settlement changes vegetation, in turn changing the behavior of fire. For example, ladder fuelsÑsmall trees and brush which carry fire into the canopyÑare often eliminated, lots are thinned to improve access and views, and large, woody ground fuels are removed in higher use areas. Added roads can improve access for fire suppression resources, but they can also host more roadside fires.
 

Higher density settlements increase fire ignition frequency. Regression analysis of ignition frequency and population by the California Department of Forestry and Fire Protection Fire and Resources Assessment Program (CDF FRAP) found a 189% increase in annual fire starts per thousand acres when residential densities went from 50-acre parcels to one-acre parcels (FRAP Website, 1998). Conversely, most fire behavior theory predicts an increase in fire hazard with increasing crown canopy cover. Thus the fragmentation occurring in the forest canopy of the Sierra Nevada as a result of human settlement could actually lead to a reduction in fire hazard if development extends over a large enough area (McBride, Russell and Kloss, 1996).
 

Runoff of precipitation from settled forest areas is greater than runoff from undeveloped forests, since interception of precipitation by tree canopies is lower in developed areas and more precipitation reaches the ground faster. Fewer trees results in less duff and woody debris on the forest floor, reducing the absorptive capacity of the land. Human settlements also introduce impervious surfaces like roofs, driveways and streets that eliminate or greatly reduce infiltration of precipitation, further altering the natural drainage of a forest.
 

Loss of canopy contributes to the observed decline in wildlife species diversity along gradients of increasing urbanization, while some well-adapted urban species increase in abundance. Where canopy losses occur, understory vegetation is altered as well. Introduced species, induced dominance of understory vegetation by opportunistic shade intolerant species, or simply lots void of brush are alterations seen commonly in settled areas (McBride, Russell and Kloss, 1996).
 

Residential Density

Throughout the Sierra Nevada approximately 32% of the land is outside of residential parcels, 62% is in parcels with densities below a wildland threshold of 1du/32ac, and 6% is parceled and settled at urban densities from less than 1 du/ac up to 1du/32 ac (Duane, 1996). This analysis focuses only on forest lands where densities are above the wildland threshold where forest impacts are known to be most significant. Approximately 87% of the regionÕs population lives within the areas settled at densities above 1du/32.
 

The most heavily settled river basins have over 90% of their population living at urban densities typically on 15% or less of the total land area. The Truckee River basin represents the extreme case in which 98% of the population is living on about 16% of the land area. The Walker River is the other extreme with 44% of its population living on 1% of the watershed.
 

Housing densities generally reflect the location of major urban centers in the Central Valley and the highways that link them to the Sierra Nevada (Map 5). The most dense areas are found in the Sierra Nevada foothills in Amador, El Dorado, Calaveras, Placer, and Nevada counties. Lake Tahoe Basin and Mammoth Lakes also reflect the higher density of Sierran recreational centers (Duane, 1996).
 

River basins with a large portion of their forested lands settled at urban densities (greater than 1du/32ac), include the Truckee (forest lands in urban densities occur in 14% of the basin), Chowchilla-Fresno (13%), Carson (12%), Mokelumne and South Fork American (11%), North Fork American/Bear River (10%) and the Yuba (8%) (Table 3-3, Figure 3-1). The South Fork American/Consumnes basin has the most forest land (over 100 thousand acres) settled in the higher density classes (Map 6). The larger river basins (over one million acres), including the Tuolumne, San Joaquin, Kings, and Kern have relatively small portions of their forests dedicated to residential use (between 0.2 and 3.7%).
 
 
 
 
 
 
 
 

Table 3-3: Acres of Sierran Forest Settled at Densities Greater than

One Dwelling Unit/32 Acres, 1990
 

Region River Basin Acres Of Settled Forest (Acres)
North Mill-Big Chico-Butte Creeks 21,221 
North Fork Feather 15,298 
E. Branch Of N. Fork Feather 9,429 
Middle Fork Feather 20,817 
Honey-Eagle Lake 4,311 
North Central Yuba 75,272 
N. Fork American-Bear River 134,042 
S. Fk. American-Consumnes 145,629 
Truckee River 38,027 
Lake Tahoe Basin (CA only) 15,276 
South Central Mokelumne 58,976 
Stanislaus-Calaveras 61,404 
Tuolumne 59,419 
Merced 16,012 
Southwest Chowchilla-Fresno 79,239 
San Joaquin 32,847 
Kings 20,697 
Kaweah 13,763 
Tule 16,825 
Kern 18,316 
East Side Carson 504 
Walker 21 
Mono 1,080 
Crowley Lake 11,127 
Source: Duane 1996, based on 1990 Census; GAP Vegetation (Holland, 1986).

Figure 3-1: Settlement of Sierra Nevada Forests, 1990
 
 

Source: Vegetation: GAP Vegetation and UCB FTP Site (based on Holland, 1986), and Calveg (based on Parker and Mathias, 1977) provided by CDF, 1997. Residential: Duane, 1996 based on 1990 Census.
 

Forest Structure Affected

The measurement of canopy loss in settled areas was undertaken for SNEP in a study of woodlands and forests occurring in portions of Sacramento, El Dorado, Amador, Nevada, and Calaveras Counties (Table 3-4). The characteristics of the forests and woodlands in these counties are typical of those farther north and south, and are similar in direction to those reported for areas of human settlement in Jeffrey pine forests in the Lake Tahoe Basin (McBride, Russell, and Kloss, 1996). The SNEP study compared canopy cover on developed lots and undeveloped lots and found measureably less canopy in developed lots. The study did not address the fact that canopy in undeveloped lots may have been affected by fire suppression. If undeveloped sites have higher canopy cover because fire has not entered them in recent times, then the measured differences between developed and undeveloped parcels would exaggerate the effects of residential development on canopy loss.
 

Table 3-4: Percentage Cover on Undeveloped and Developed Property Associated with Human Settlement (adapted from McBride, Russell and Kloss, 1996)
 

Percentage of Ground Covered by Tree Canopy
 
Parcel Size= <1 acre 3-5 acres 10-20 acres Average Cover
Foothill Woodland Developed 43% 70% 52%a
Undeveloped 69% 90% 74% 78%
Lost Coverb 35% 8% 26%
. Ponderosa Pine Developed 62% - -
Undeveloped 90% - -
Lost Cover  28%
Mixed Conifer Developed 64% - -
Undeveloped 92% - -
Lost Cover  28%
Notes

Foothill Woodland: Elevation 500-2,500 ft.; dominated by blue oak (Quercus douglasii); other common trees include maul oak (Q. chrysolepis), interior live oak (Q. wislizenii), and foothill pine (Pinus sabiniana)

Ponderosa Pine: Elevation 2,000-2,500 ft. in central Sierra Nevada; dominated by ponderosa pine (Pinus ponderosa); common trees include California black oak (Q. kelloggi) and incense cedar (Calocedrus decurrens) at higher elevations

Mixed conifer: Elevation 2,500-6,000 ft. in central Sierra Nevada; trees include ponderosa pine, incense cedar, white fir (Abies concolor), Douglas fir (Pseudotsuga menziesii), and sugar pine (P. lambertiana); California black oak is common.

a 52% is average of cover loss on area immediately around structures (houses, farm buildings, sheds) and the portion of property not adjacent to structures

b =(Average cover of undeveloped lots) - (cover on developed lot)
 

Based on these measured values we extrapolated cover losses in settled areas of oak woodland, ponderosa pine, and mixed conifer vegetation types (Table 3-5). Conifer woodland, while extensively settled in the northern and central Sierra Nevada, was not examined, since no empirical data are available about the effects of residential development in this forest type. The extrapolated cover losses were then used to quantify average acres of canopy remaining in areas developed at five residential densities (See Appendix Table A-1: Total Acres in Residential Development, and Average Acres Under Forest Canopy Prior to, and Remaining After Development).
 

Table 3-5: Extrapolated Decreases in Percentage of Ground Covered by Tree Canopy at Various Lot Densities
 

Decrease in Percentage of Ground Covered by Canopy
 
Parcel Size= <1 acre 1-4 acres 4-8 acres 8-16 acres 16-32 acres
Foothill Woodland 30% 29% 22% 15% 8%
Ponderosa Pine 28% 22% 16% 10% 2%
Mixed conifer 28% 22% 16% 10% 2%
 

Approximately 62,700 acres of forest canopy are estimated to have been removed in these three forest types throughout the Sierra Nevada. Mixed conifer forests bare the brunt of residential development in the Sierra Nevada. Over 50 percent of settled forest lands (over 250,000 acres excluding Tahoe and Big Chico-Mill-Butte basins) are mixed conifer which occurs in a broad band between approximately 2,500 feet and 6,000 feet elevation in the central portion of its range. The proportion of mixed conifer supporting housing is greatest north of the Tuolumne River Basin. Over 29,000 acres of mixed conifer are settled at densities above 1du/32 acres in the South Fork American and Consumnes Basins alone. This has resulted in an estimated 23,366 acres of canopy loss. Over 51,000 acres are estimated to have been lost in the North Fork American and Bear River Basins (Figure 3-2).
 

Development of the westside ponderosa pine forests is greatest in the American, Bear, Consumnes, Tuolumne and Mokelumne River Basins where almost 85,000 acres of this forest type are settled with densities greater than 1 du/32ac. An estimated 18,000 acres of ponderosa pine canopy loss has occurred range-wide. The Truckee River Basin has lost an estimated 3,572 acres and the South Fork American approximately 3,000 acres. The Tahoe Basin has also seen substantial impacts (approximately 1,836 acres) in ponderosa pine canopy loss just on the California side. Broadleaf forests are settled principally in Central and Southern portions of the Sierra with the Chowchilla/Fresno Basin having the most acres of housing in this forest type (36,141 acres). Five southern river basins are the only ones in the Sierra where the majority of housing at urban densities occurs in non-conifer forest types; these include, the Merced, Chowchilla/Fresno, Kings, Kaweah, and the Tule River Basins. A prevalence of oak woodland explains this trend except in the Merced and the Chowchilla/Fresno where the broadleaf forest supports about 27% and 48% of the denser housing in forested areas.
 

The non-coniferous forests of the Southwest Basins have been affected the most by human settlement. Oak woodlands support the higher residential densities on more than 158,000 acres (not including Tahoe and Big Chico-Mill-Butte Basins) throughout the range. We estimate that approximately 21,273 acres of canopy loss has occurred in oak woodlands throughout the Sierra Nevada. The American, Bear, and Consumnes River Basins have over 48,000 acres of Oak Woodlands developed at densities greater than 1 du/32 acres, resulting in almost 8,000 acres of canopy loss in these basins (Figure 3-3). Oak woodland canopy loss is also significant in the Chowchilla-Fresno (approximately 2,877 acres lost), the Kaweah (approximately 1,298 acres lost) and the Tule (approximately 1,234 acres lost) Basins.
 

Approximately 4,500 acres of broadleaf forest are developed at densities of 1/16-1/32 du/ac in the dispersed development around Bootjack, in the Chowchilla-Fresno Basin (Figure 3-3). In the Fresno River watershed alone, almost 20,000 acres of oak woodland around Oakhurst and areas to the immediate south and west along highway 49 are no longer wildlands, having been developed at densities greater than 1du/32ac. The FresnoÕs watershed also has the greatest impacts in conifer and mixed conifer within the basin.
 

3.3.2 Road-Impacted Waterways
 

Forest riparian systems are adversely affected by road construction and maintenance. Roads cause the direct loss of acreage of riparian areas, the direct loss of large trees, reduced structural complexity of riparian and aquatic environments, reduced supply of large woody debris to aquatic systems, reduced base flows with increased peak flows in streams and rivers, gully development and accelerated downstream sedimentation.
 

The type, distribution, and total miles of roads that have impacted waterways in Sierra Nevada forests will never be entirely known because of the regionÕs size. Nevertheless, we undertook to estimate the overall scale of the problem using data available for the whole region. Where there is a high density of riparian intrusion by roads included in our source data (USGS 1:100,000 Scale Digital Line Graph Maps), we infer the existence of a smaller network of skid trails on lands managed for timber harvest, as well as other Òghost roadsÓ known to exist but not on maps. The coarseness of the source data limits the use of our results to that of indexing where further analysis of road problems is required within the regionÕs 26 major river basins. When combined with local knowledge of the condition of road networks, the results will aid in prioritizing future analysis.
 

Figure 3-2
 

Figure 3-3
 

Where roads are less than 150 feet from a waterway we infer their potential to impact the waterway. The use of a fixed 150-foot buffer for an entire ecosystem is highly simplistic since road effects are so variable. At a finer scale of analysis it would be appropriate to employ variable widths based on: community area (the area which provides for the living requirements of those organisms dependent for their survival on the special conditions of the riparian area); energy area (the area that supplies organic material and attenuates the affects of solar radiation); and an index of slope distance around the aquatic system equivalent, for example, to the height of a mature tree in that location; and possibly other measurable risk factors (Kondolf, Kattelmann, Embury and Erman, 1996).
 

Additionally, the effects of a road on a stream can extend a considerable distance downstream from the road. The estimates presented here focus on the source of the problemÑthe roadÑand do not attempt to quantify the full extent of riparian impacts caused by the road over time. Restoration efforts that eliminate the source of the problem are the essential first step in restoring the area impacted. The potential for a full recovery of the area is also dependent on the inherent conditions on the site (geology, soils, slope, and climate) and historical and current land use disturbance upslope of the site (e.g., impervious cover, reduced vegetation). If natural recovery processes do not occur at a rate deemed acceptable, further intervention can be pursued. Such intervention could require that many issues unrelated to forest management be addressed (e.g., flow releases from reservoirs, grazing in forest meadows).
 

Distribution of Waterways Affected by Roads

Figure 3-4 shows the occurrence of roads in forest riparian areas is most common in North Central Basins, particularly the North Fork American/Bear, South Fork American/Consumnes, and Stanisluas/Calaveras River Basins. These data are screened for the six Sierra Nevada forest types and do not include roads in lands classified as non-forest vegetation. Forest roads (includes minor streets but is principally composed of roads and trails in wildlands) are the most likely to occur in sensitive riparian areas due largely to their sheer abundance. However major roads (e.g. interstate highways) and primary routes (e.g. undivided and divided paved roads and streets) are significant in the North Central Basins particularly.
 

Figure 3-4: Major, Primary, and Forest Roads in Riparian Zones by Basin!
 

Restoration needs associated with roads in the Sierra Nevada are greatest on roads built for harvesting timber. The installed timber road base was not located, constructed, nor has it been maintained, with adequate attention to protecting riparian and aquatic environments. On the other hand, roads and streets constructed for other purposes (e.g., select recreational routes, trans-Sierra routes, utility service roads, fire roads, and roads and streets associated with human settlement) are typically maintained for continued use and are managed in a manner to minimize impacts beyond initial construction impacts.
 

The Feather River Basin has the most miles of mapped forest roads within the 150-foot buffer of all the basins of the Sierra Nevada (Map 7a, Table 3-6). Also heavily roaded are the North Fork American/Bear and South Fork American/Consumnes Basins (Map 7b) (See Appendix Table A-2 for breakout by HSA). The Mount Harkness (North Fork Feather River Basin), South Fork American, and North Fork Consumnes Hydrologic Sub Areas each have more than 80 miles of road entering riparian areas
 

The South Central and Southwest subregion HSAs have approximately 450 and 470 miles of forest roads in riparian areas, respectively (Map 7c, 7d). The upper Mokelumne has the most miles of riparian roads of all HSAs in these subregions (approximately 85 miles). However, the riparian areas in the Calaveras, Clavey, and Tuolumne River basins suffer significant incursions by forest roads as well (Table A-2). The riparian areas of the more arid and less forested East Subregion have fewer miles of roads than other subregions (Map 7e).
 

Table 3-6: Forest Roads Within 150 feet of Riparian Areas in Select Hydrologic Sub-Areas
 

RIVER BASIN HSA MILES
North Fork Feather 518.40 Mount Harkness  94.7
518.60 N. Fk. W. Br. Feather 17.4
518.90 19.8
131.9
E. Branch No. Fk. Feather 518.51 11.2
518.52 38.1
518.53 25.6
518.54 26.6
518.55 19.5
518.56 11.4
132.3
Middle Fork Feather 518.20 Little Grass Valley (S.Fk. Feather) 16.7
518.30 Frenchman Lake (M. Fk. Feather) 13.4
518.32 29.3
518.33 39.1
518.35 32.4
518.80 7.6
138.6
North Fork American 514.20 Auburn  3.9
514.45 1.9
514.49 31.9
514.50 N Fk. American (Blue Canyon) 32.2
514.60 N. Fk American (Hell Hole) 5.5
514.90 Snow Mountain  7.5
Bear River 516.20 15.1
516.30 Upper Bear (Rollins Reservoir) 59.3
157.3
S. Fk. American/Consumnes 514.30 S. Fk. American (Union Valley) 82.5
514.40 M. Fk. American (Kyburz) 31.2
514.80 1.9
532.20 North Fork Consumnes  97.7
213.3
Source: Source: 150-Foot buffer applied to Roads and Hydrology: USGS 100K Digital Line Graph, 1993; Vegetation: GAP Vegetation and UCB FTP Site, Holland, 1986, and Calveg (Parker and Mathias, 1977) provided by California Department of Forestry and Fire Protection, 1997.
 

Land Protection Status of Roads Affecting Waterways

An examination of the distribution of potential road problems relative to forest protection status offers insight into the likely condition of the roads, as well as the options available for restoration. With 61 percent of the regionÕs total miles of riparian roads occurring in private and unprotected lands, restoration strategies will necessarily involve a full compliment of approaches that advance the interests of private landowners (Table 3-7). Such approaches will vary according to stakeholder interests which include everything from industrial timber management to quality of life for residents. Approximately 23 percent of riparian roads are currently under management permitting timber harvest and other multiple uses by the USDA Forest Service and other federal land management agencies (Figure 3.5). The condition of roads under these different regimes varies considerably. For example, USDA Forest Service roads experience intense use beyond timber harvesting for recreation and other uses. Their condition is often worse than roads in private areas where access is controlled.
 

Table 3-7: Forest Roads in Riparian Areas by Land Protection Status (Miles)
 

REGION RIVER BASIN Class 1: Full Protection Class 2: Mostly Protected  

(no grazing) 

Class 3: Mostly Protected  

(some grazing) 

Class 4: Other Public Land  Class 5: Private and Unprotected TOTALS
 
North Mill/Big Chico/Butte 8.2