8/2/2017

Cornell Local Roads Program

County of Dutchess, New York

Summer 2017

Robert Schumacher and George DeWitt

Dutchess County Department of Public Works:

Engineering Division

626 Dutchess Turnpike

Poughkeepsie, NY 12603

1

Table of Contents

Executive Summary…………………………………………………………………………………………………………………. 2

What is Cornell Local Roads Program (CLRP)?…………………………………………………………………………….. 2

Training …………………………………………………………………………………………………………………………………. 2

Results ………………………………………………………………………………………………………………………………….. 2

Process and Project Parameters ……………………………………………………………………………………………… 3

Surveying Roads ……………………………………………………………………………………………………………………... 3

Inventory Data …………………………………………………………………………………………………………………..…… 3

Data Collection ……………………………………………………………………………………………………………………….. 4

Condition Data …………………………………………………………………………………………………………………..…… 4

CAMP-RS Software ………………………………………………………………………………………………………………….. 9

Potential Repairs …………………………………………………………………………………………………………………… 10

Preventative Road Maintenance …………………………………………………………………………………………….. 10

Decision Tree ……………………………………………………………………………………………………………………….. 10

Repair Categories …………………………………………………………………………………………………………….…… 10

Results…….……………………………………………………………………………………………………………………………… 12

County Analysis...…………………………………………………………………………………………………………….……. 12

Local Analysis...…………………………………………………………………………………………………………….………. 13

Functional Class Analysis..………………………………………………………………………………………………….…… 14

Error Analysis….....…………………………………………………………………………………………………………….…… 15

Observations.………………………………………………………………………………………………………………………... 15

Conclusion …………………………………………………………………………………………………………………………….. 15

Appendices ……………………………………………………………………………………………………………………………. 17

Appendix A: Map of Dutchess County ……………………………………………………………………………………… 17

Appendix B: Map of Progress by Summer…………………………………………………………………………………. 18

Appendix C: Sample Pavement Condition Survey ……………………………………………………………………... 19

Appendix D: Sample Inventory Sheet …………………………………………………………………………………...…. 20

Appendix E: PCI Comparison of County Roads by Municipality …….……………………………………………… 21

Appendix F: PCI County Roads Map…………………………………………………….……………………………………….. 22

Appendix G: PCI Comparison of Local Roads by Municipality ……………………………………………………… 23

Appendix H: PCI Local Roads Map……………………………………………………………………………………………. 24

Appendix I: Change in miles of road by municipality………………………………………………………………….. 25

2

Executive Summary:

What is Cornell Local Roads Program (CLRP)?

The Cornell Local Roads Program (CLRP) is a training program instituted by the State of

New York to provide local municipalities with the training needed to properly manage and

maintain the roads under their jurisdiction. CLRP has designed a computer program (CAMP-RS)

to digitize the road condition data on a robust platform. The Pavement Internship Training, that

CLRP offers, is designed to train college students in the CAMP-RS software and provide them

with the tools needed to evaluate the roads within their municipality. During the ten week

internship program, interns are expected to develop an inventory, conduct road evaluations,

and prepare a final report of the management plan for the local roads and streets.

Training

Training was held over the course of 3 days on Cornell’s campus in Ithaca, New York.

The training consisted of education on road design, construction, and maintenance. It also

included lessons on asphalt paving principles, pavement maintenance options, and how to

operate the CAMP-RS software.

Results

During the summer of 2017, the second phase of the CLRP survey process was

completed. A total of 2,304 sections, equaling 1,046 miles of road, were surveyed. This in

conjunction with the 3,176 sections (945 miles) evaluated in 2016, provides a full and complete

assessment of all local asphalt roads within Dutchess County.

In 2016, 10 towns, 3 villages, and 2 cities were provided with a road inventory list

containing all roads within the municipality and the conditions of these roads. In 2017, an

additional 10 towns and 3 villages were provided with this data. The map in Appendix B shows

which municipalities were completed each summer. The data provided to these municipalities

omits the county roads within their geographic borders. These reports also provide information

on how the Pavement Condition Index (PCI) was calculated for each road, and what those

ratings mean.

3

Process and Project Parameters:

Dutchess County has 2,050.3 miles of local roads, ranging from unpaved seasonal roads

to paved principal arterials. The original goal of this internship program was to evaluate all

asphalt roads in the entire county over 3 summers. This goal was successfully met after 2

summers. George DeWitt and Anna Kern worked on this project the first summer. George

DeWitt and Robert Schumacher worked on this project the second summer. Due to the large number of roads being surveyed, an effective method of evaluating the

roads needed to be implemented. The standard process of surveying roads was to use printed

survey sheets to record the road information and condition while in the field, and then input

this data into the database upon returning to the office. Since there were two full time interns

assigned to this project, a laptop was used by the passenger to record the data, while the driver

and passenger worked together to identify distresses in the roads. The road inventory was

imported into the laptop before going to survey, and the condition data was entered into the

database as each road was inspected. This method saved paper and a significant amount of

time.

Surveying Roads:

Inventory Data

The road inventory was taken from the New York State Department of Transportation

Local Highway Inventory - 2015 Local Roads and Streets, for Dutchess County. This list included

all county, city, town, and village roads within Dutchess County. State and private roads were

not surveyed. By using New York State’s official inventory, all 5,275 entries did not have to be

created manually. Over the course of the last two summers, several mistakes in the database

were identified and corrected. These mistakes included missing roads, incorrect sections,

incorrect lengths, incorrect surfaces, and roads in the database which do not presently exist.

These mistakes were recorded in a separate “Missing Roads and Paper Roads” report. To

simplify the process, this database was split into individual databases for each town, village,

and city. An example of the inventory data can be found in Appendix D.

4

Data Collection

The task of surveying the roads was approached one municipality at a time. When

possible, each municipality was divided into sections by the county roads that crossed it. These

county roads were rated first, and then the smaller local roads within the sections were rated.

This method made the task easier, and allowed for fewer missed roads after each municipality

was completed. The roads were rated while driving, which minimized the need to stop.

Condition Data

Each road was graded using the same criteria in order to maintain consistent ratings.

This criteria was determined using the Surface Condition Survey Sheet (as shown in Appendix

C), provided by CLRP. The Surface Condition Survey Sheet allows for an organized recording of

the condition of the entire street system. Its purpose is to identify existing distresses in a road

surface which affects the performance of the road. The resulting PCI highlights structural or

material defects that lead to deterioration of road performance and eventual failure.

Surface Condition Surveys were available for Asphalt Treatment, Surface Treatment,

Concrete, and Unpaved roadways. Only asphalt roads were surveyed for this project. The

document separates the distresses in asphalt into the following eight categories:

● Alligator Cracking

● Longitudinal/Transverse Cracking

● Edge Cracking

● Potholes/Patching

● Roughness

● Rutting

● Bleeding/Raveling

● Drainage

The following explanation was provided by the Cornell Asset Management Program -

Roads and Streets 2014 User Guide.

5

Longitudinal/Transverse Cracking

Longitudinal cracks are cracks that run parallel to the roadway centerline. Longitudinal

cracks are usually found at construction joints and between lanes. Transverse cracks run

perpendicular to the roadway centerline. Transverse cracks are generally spaced at regular

intervals and caused by expansion and contraction of the road surface material.

Both types of cracks can also be reflective, appearing above joints and cracks in

underlying pavements.

Alligator Cracking

Alligator cracking refers to interconnected crack patterns that resemble alligator skin or

chicken wire. Pavement pieces range in size from one to six inches on a side.

6

Patching/Potholes

Patching refers to areas where the original pavement has been removed and

subsequently replaced but is showing deterioration. Potholes are areas where portions of the

road pavement have broken and loss of pavement has resulted in a bowl shaped depression.

Edge Cracking

Edge cracking refers to cracks adjacent and parallel to the edge of the pavement. While

generally confined to the outer one to two feet of pavement, edge cracking can progress into

the travel lane.

7

Drainage

Drainage severities are judged by the ability for runoff to flow from the paved area to a

location that does not influence roadway conditions. Visual indicators of drainage problems

include accumulation of debris and sand and high water marks. (Evaluation during or after a

rainfall event can be extremely beneficial.)

Roughness

Pavement roughness is defined as irregularities in roadway surface which adversely

affects the comfort of the ride.

Rutting

Rutting refers to channels in the wheel paths. Rutting causes water to drain along the

road surface rather than drain to the edge of the road.

8

Bleeding

Bleeding refers to the excess asphalt material on the surface of the roadway. Bleeding

can be a safety problem due to decreased skid resistance.

Raveling

Raveling is the wearing away of the pavement surface caused by the dislodging of the

surface aggregate particles and the loss of asphalt binder. Raveling includes the loss of fine and

coarse aggregate. The surface becomes characterized as very rough and pitted with the obvious

loss of aggregate.

Most of the above categories are rated by both severity of the distress, and extent of

distress. Potholes/Patching are only rated by extent, and Bleeding/Raveling and Drainage are

only rated by severity. The definitions for Severity and Extent as defined by the CAMP-RS User

Manual are shown in the table below.

9

Severity: Extent:

Low

Present but not causing immediate problems

Low

Isolated to a few locations

Moderate

Needs attention before it becomes a problem

Moderate

Between 10 to 30 percent of the road is affected by the distress

High

Maintenance is needed immediately as the feature is a problem

High

More than 30 percent of the road segment is affected.

CAMP-RS Software:

The Cornell Asset Management Program - Roads & Streets (CAMP-RS) Software that

CLRP provides is designed to be an objective and efficient means of surveying and analyzing the

condition of the roads. CAMP-RS is used to generate an accurate inventory of the roads within

the municipality's jurisdiction and objectively survey the condition of these roads.

Each entry in the inventory included road name, Road Identification Number (RIN),

section information, dimensions, surface type, and traffic. Road location (by municipality) and

jurisdiction (by either county, town, or village/city) was added to the entries as well.

An example of the inventory data can be found in Appendix D. The DOT ID provided by

the State for each road was used as the RIN. In order to generate unique RINs, sections

numbers were added to the DOT ID and separated with a dash.

After the inventory was compiled, the road sections were surveyed and a Pavement

Condition Index (PCI) was determined. The PCI’s ranged from 0-100, and were determined

based on the extent and severity of the various failures found over the length of the road.

These factors and their numerical weight can be found in Appendix C, the Pavement Condition

Survey.

The software is capable of generating a priority value using the PCI value, AADT traffic

data, and functionality class of the road. This priority is usually an inverse of the PCI value. The

10

software can also suggest a repair type and cost based on the road condition. This can be

combined with the total road maintenance budget to determine how much money should be

allocated to each road. However, the only data analyzed for this report was the PCI value. The

repairs and budget allocation were left to the discretion of the Highway Superintendents from

each municipality.

Potential Repairs:

There are four basic steps leading to a repair plan: taking inventory of the road network,

gathering road condition data, determining the appropriate repairs, and prioritizing repairs in

order to make a repair schedule.

The CAMP-RS program generates suggested repairs based on the data collected. These

repairs are grouped by repair strategy and repair category. The repair category is determined

by the extent and severity of road distress, and the repair strategy is the solution to the

distresses. A decision tree is used by the software to determine a repair strategy from a repair

category. An example of a decision tree for Alligator cracking is shown below.

Decision Tree;

Alligator Cracking - Asphalt Pavement - Repair Categories

Severity Extent

Low Moderate High

Low 41 44 44

Moderate 43 43 45

High 43 46 48

The decision tree generates a two digit number that defines the surface type and the

repair category. The first digit (4) refers to asphalt surface, and the second digit (1-8) refers to

the repair category. These decision trees can be altered by the municipality to suit their needs.

The six road surface repair categories are preventative maintenance, deferred action,

routine maintenance, corrective maintenance, rehabilitation, and reconstruction. The following

list defines each category, and its correlating number code.

11

1. Preventative maintenance is action designed to stop deterioration before it becomes a

larger problem that needs a different fix. Sealing cracks is an example of this.

2. Routine maintenance is for roads in good condition. This is usually a cost effective

option which includes local patching, or removal of bumps. Routine maintenance may

prevent larger problems from occurring.

3. Corrective maintenance is maintenance which corrects an existing defect. It is more

limited than, and not as cost effective, as other maintenance. Examples include boxing

out and patching, or installing subsurface drainage. Corrective maintenance differs from

rehabilitation in that corrective maintenance is less extensive.

4. Deferred action is for sections of road beyond the point of effective preventative

maintenance, but not deteriorated to where they need rehabilitation.

5. Rehabilitation differs from reconstruction in whether base replacement or recycling is

done, and usually includes overlays or extensive recycling.

6. Reconstruction is the complete removal and replacement of failed pavement, and may

involve other changes to the road such as drainage work or widening of the road.

Reconstruction is generally the most expensive repair by a large margin.

Prioritizing repairs is necessary to create an effective road repair schedule. Routine

maintenance is more cost effective than letting roads deteriorate to a point where

reconstruction or rehabilitation is necessary. However, routine maintenance requires an

accurate evaluation of the roads, and a long-term plan for the road maintenance. As shown in

Figure 1 and 2 below, routine and preventative maintenance done in the early life of a road are

much less expensive in the long run. However, if the road is allowed to deteriorate to a very

poor pavement condition, the cost is amplified.

12

Results:

Analysis

County

The total Pavement Condition Index data generated from CAMP-RS was collected and

analyzed in hopes of drawing effective conclusions. The roads were divided by jurisdiction for

the analysis to provide more accurate and useful data to Dutchess County as well as the

municipalities. The first chart in each section shows the unweighted PCI Frequency. The second

shows how many miles of road fall into each PCI Value. The results of the unweighted

frequency and weighted frequency are very similar.

Pavement Condition Index Data: County Roads

13

These results indicate the majority of County roads are kept in good to very good

condition. By comparison, there are relatively few roads in poor or very poor condition. While

the road network could always be improved, this is the desired spectrum for road condition.

To provide a geographic perspective, the county roads were divided by the municipality

they were located in. The resulting weighted and unweighted frequencies were compared to

the County averages in Appendix E. The weighted averages were highlighted with a range of

colors from green to red to show their respective values. Municipalities highlighted in red are

below average locations, and municipalities highlighted in green are above average locations.

The percent difference from the average is also displayed. A color-coded map showing the PCI

scores of County roads by municipality can be found in Appendix F.

Local Pavement Condition Index Data: Local Roads

The Local roads are mostly in kept in good and fair condition. There are several roads

that are in poor or very poor condition, and in need of reconstruction. The Local roads tend to

have a wider range in values and a lower average score than County roads, which is to be

expected.

14

Similar charts were provided for each municipality. The unweighted and weighted

averages for each municipality were calculated as an effective means of comparison. These

results are shown in Appendix G. The weighted averages were highlighted with a range of

colors from green to red to show their respective values. Municipalities highlighted in red are

below average, and municipalities highlighted in green are above average. The percent

difference from the average is also displayed. A map color coded to show each municipalities

weighted average PCI scores by municipality can be found in Appendix H.

Different municipalities have different populations and varying budgets, so this data

should not be used to judge a municipality’s performance, only the relative condition of their

roads.

Due to the construction of new roads and errors found in the database, several

municipalities saw a change in total miles of road. This change was shown in the Appendix I.

There was no change in total miles of road of County roads.

Functional Class Analysis

An analysis of the deterioration of roads over time by functional class was completed for

all CHIPS (Federal Aid) Eligible Roads. The functional classes considered are 12 (Principal

Arterial), 16 (Minor Arterial), and 17 (Major Collector). This analysis includes 6 years’ worth of

PCI’s for most of these roads (2011-2016). Due to time constraints, these roads have not yet

been surveyed for 2017. The graph below shows the Yearly Weighted Average PCI for each

class. While each road deteriorates with time, the positive trend of this graph indicates that the

condition of the road network is increasing with time. This is especially important since these

roads carry the bulk of local traffic and are a key part of Dutchess County’s infrastructure.

Error Analysis

55.00

60.00

65.00

70.00

75.00

80.00

85.00

90.00

95.00

2011 2012 2013 2014 2015 2016

PC

I

Year

CHIPS Roads Weighted PCI by FC

FC 12

FC 16

FC 17

15

A simple error analysis was conducted to determine the error between average PCI

values obtained each summer. The percent error was less than 6%. While this is not negligible,

it is low enough to not discredit the data. This error could be due to differences in scoring and

human error. However, it could simply be due to difference in pavement condition in the

different municipalities and geographic locations surveyed.

Observations

While evaluating in excess of 2000 miles of roads, certain patterns and characteristics of

road decay became apparent. Over time, longitudinal cracks often splintered and became

alligator cracks, and eventually pot holes. Many roads also had longitudinal cracking along the

centerline. The centerline cracking appeared to most likely be caused by the paving seam not

being sealed during construction or repairs. Corners at intersections often had edge cracking

that wasn’t present at other locations along the road as well, even if the intersection had a

large shoulder. Asphalt curbs, which were found in some of the more populous municipalities,

appeared to do an excellent job at delaying the development of edge cracking. However, it is

noted that this may be more correlation than causation, since some town codes may have

changed to require them in recent years.

High traffic tended to increase the amount of cracking present but did little to affect the

roughness or drivability of the road. The issue which caused major hazards seemed to have

other issues such as improper drainage or lack of any identifiable maintenance.

Conclusion:

The Cornell Asset Management Program - Roads and Streets, was an effective tool for

evaluating the roads in Dutchess County. The Cornell Local Roads Program provided the training

necessary to complete the road evaluation and to use the CAMP-RS software. Dutchess County

now has a CAMP-RS road inventory with an objective, quantitative, and quality score for all of

the roads in the county. The County’s database of the local roads has also been updated to

more accurately reflect its inventory.

Average PCI Weighted Average PCI

Summer 2016 82.8 81.6

Summer 2017 79.6 77.2

Difference 3.1 4.3

Percent Difference 3.8% 5.3%

Comparison of Average Scores Between Summers

16

In general, the county roads’ condition exceeded the condition of local town, village, or

city roads. Roads in areas with steep slopes were often in poorer condition than roads in areas

with flat land, typically correlating to the population of the immediate area.

It is important that the database be updated regularly since roads deteriorate over time.

After obtaining several cycles worth of data through diligent updating, as is the case with CHIPS

Eligible Roads, the combined data could be used to determine road condition trends. These

trends can be used to generate a comprehensive view of road deterioration over time, and

predict the cost associated with it. This data will allow for better planning, lower costs, and the

development of a better and safer road network.

17

Appendices:

Appendix A: Map of Dutchess County

18

Appendix B: Map of Areas Surveyed Each Summer

19

Appendix C: Sample Pavement Condition Survey

20

Appendix D: Sample Inventory Sheet

21

Appendix E: PCI Comparison of County Roads by Municipality

Municipality Miles of Road

Inspected Average

PCI Weighted Average PCI Percent Difference

All Dutchess County 393.0 84.9 82.5 -

Town of Fishkill 5.2 92.3 93.0 10.7%

Village of Millerton 0.6 92.0 92.0 9.9%

Town of Red Hook 17.6 89.6 89.7 7.1%

Town of La Grange 22.9 88.5 88.1 5.5%

Town of Stanford 28.1 88.5 88.0 5.4%

Town of Wappinger 21.1 89.3 87.9 5.8%

Town of Pleasant Valley 12.9 88.1 87.9 5.1%

Town of Clinton 25.3 87.0 86.6 3.7%

Town of Rhinebeck 13.7 86.0 85.0 2.2%

Village of Fishkill 0.1 85.0 85.0 1.5%

Town of Poughkeepsie 9.7 84.8 84.6 1.2%

Village of Wappingers Falls 0.04 84.0 84.0 0.3%

Town of Hyde Park 20.8 85.4 83.9 1.1%

Town of East Fishkill 17.2 82.3 83.4 -1.0%

Town of Milan 23.2 84.1 82.5 -0.5%

Town of Pawling 20.8 83.1 82.4 -1.2%

Village of Tivoli 1.4 81.0 81.6 -2.9%

Town of Beekman 14.5 83.9 81.0 -1.5%

Town of Union Vale 28.9 82.0 80.2 -3.1%

Town of Washington 18.6 83.5 79.9 -2.5%

Village of Pawling 1.4 83.3 79.5 -2.8%

Village of Red Hook 0.5 79.0 79.0 -5.6%

Town of North East 32.3 81.6 76.2 -5.8%

Town of Dover 23.3 81.3 76.0 -6.1%

Town of Pine Plains 10.1 82.9 75.9 -5.2%

Town of Amenia 22.7 75.6 70.4 -12.9%

City of Beacon N/A N/A N/A N/A

City of Poughkeepsie N/A N/A N/A N/A

Village of Millbrook N/A N/A N/A N/A

Village of Rhinebeck N/A N/A N/A N/A

*All Municipalities labelled as N/A do not contain any county roads

22

Appendix F: PCI County Roads Map

23

Appendix G: PCI Comparison of Local Roads by Municipality

Municipality Miles of Road

Inspected Average

PCI Weighted

Average PCI Percent

Difference

All Local Roads 1571.5 81.6 79.5 -

Town of Pleasant Valley 70.5 85.7 85.4 6.2%

Town of Amenia 31.7 85.5 84.5 5.6%

City of Beacon 50.6 84.4 83.9 4.5%

City of Poughkeepsie 67.4 84.3 83.9 4.4%

Village of Red Hook 8.4 83.8 83.4 3.8%

Town of Pine Plains 38.9 84.7 83.3 4.3%

Town of North East 35.4 84.4 83.2 4.1%

Town of Red Hook 58.0 82.8 83.2 3.1%

Town of Wappinger 98.3 83.7 82.3 3.0%

Village of Fishkill 5.6 83.2 82.2 2.7%

Village of Millerton 4.9 81.4 81.8 1.4%

Town of Beekman 66.8 83.1 81.7 2.3%

Village of Tivoli 5.1 79.1 81.2 -0.4%

Town of La Grange 118.2 82.7 80.7 1.5%

Village of Pawling 12.1 81.6 80.2 0.5%

Town of Poughkeepsie 139.9 80.1 80.0 -0.6%

Town of Rhinebeck 57.4 82.1 79.7 0.4%

Town of Dover 64.2 83.0 79.1 0.7%

Town of Milan 51.8 78.9 78.7 -2.2%

Village of Rhinebeck 10.7 80.3 78.5 -1.4%

Town of Fishkill 57.4 80.4 78.2 -1.6%

Town of Washington 64.6 82.2 77.3 -1.0%

Village of Millbrook 7.9 69.9 75.9 -9.4%

Town of East Fishkill 196.3 78.9 75.1 -4.4%

Town of Hyde Park 93.3 77.4 73.9 -6.0%

Village of Wappingers Falls 11.8 74.5 73.2 -8.3%

Town of Stanford 62.6 79.2 73.0 -5.6%

Town of Union Vale 45.3 75.2 72.1 -8.6%

Town of Pawling 53.4 68.5 66.3 -16.3%

Town of Clinton 69.1 70.7 66.0 -15.1%

24

Appendix H: PCI Local Roads Map

25

Appendix I: Difference in Total Miles of Local Road between NYS and Survey

Results

Municipality Original Miles

of Road* Miles of Road Inspected**

Updated Mileage***

Percent Difference

Total Local Roads 2050.3 1571.47 1957.89 -4.51%

Dutchess County 393.0 393.0 393.0 0.0%

City of Beacon 50.60 50.52 50.59 0.0%

City of Poughkeepsie 67.40 67.67 67.43 0.0%

Town of Amenia 31.70 25.55 31.69 0.0%

Town of Beekman 66.80 68.97 70.02 4.8%

Town of Clinton 69.10 58.71 69.13 0.0%

Town of Dover 64.20 46.04 64.20 0.0%

Town of East Fishkill 196.30 209.76 214.12 9.1%

Town of Fishkill 57.40 56.03 57.43 0.1%

Town of Hyde Park 93.30 93.29 96.22 3.1%

Town of La Grange 118.20 120.02 120.02 1.5%

Town of Milan 51.80 45.53 51.75 -0.1%

Town of North East 35.40 33.11 35.36 -0.1%

Town of Pawling 53.40 43.28 53.43 0.1%

Town of Pine Plains 38.90 38.87 38.87 -0.1%

Town of Pleasant Valley 70.50 67.97 70.46 -0.1%

Town of Poughkeepsie 139.90 138.96 139.85 0.0%

Town of Red Hook 58.00 57.25 57.90 -0.2%

Town of Rhinebeck 57.40 56.32 57.35 -0.1%

Town of Stanford 62.60 51.60 62.62 0.0%

Town of Union Vale 45.30 42.85 46.28 2.2%

Town of Wappinger 98.30 97.86 98.25 -0.1%

Town of Washington 64.60 36.21 64.58 0.0%

Village of Fishkill 5.60 5.54 5.61 0.2%

Village of Millbrook 7.90 7.67 7.87 -0.4%

Village of Millerton 4.90 4.86 4.86 -0.8%

Village of Pawling 12.10 12.12 12.12 0.2%

Village of Red Hook 8.40 7.97 8.38 -0.2%

Village of Rhinebeck 10.70 10.69 10.69 -0.1%

Village of Tivoli 5.10 4.81 5.09 -0.2% *Original miles of road according to 2015 NYS Local Highway Inventory **Miles of inspected asphalt roads during Summer 2016 and Summer 2017 ***Updated Mileage combines miles of inspected asphalt roads with miles of uninspected dirt/concrete roads