Modelling of the Mount Hope River watershed with BASINS ...journalsonline.org/american-journal-of-engineering...Dellman et al. (2002) considered that with the growing need to monitor

American Journal of Engineering, Science and Technology (AJEST) Volume 5, 2020

1

Modelling of the Mount Hope River watershed with BASINS-

HSPF to simulate discharges, sediment and snowpack

Juan M. Stella1*

1 Tecnológico de Monterrey, Monterrey, México

Abstract

Hydrological simulations of Mount Hope River watersheds located in northeast Connecticut,

were performed using the Unites States Environment Protection Agency (EPA) model

BASINS-HSPF (Better Assessment Science Integrating Point and Nonpoint Sources -

Hydrological Simulation Program Fortran) from 1997 to 1998. The simulation was performed

at monthly time step during one year and at sub-watershed scale. The purpose of the project

was to determine the efficacy of BASINS-HSPF in a forestry watershed for the simulation of

monthly discharges, sediments transport and snowpack. The model was calibrated using

measured discharges of the Mount Hope River and the results applied for the monthly

simulation of sediment transport and snowpack, these parameters simulated were compared

with sediment transported and snowpack calculated by another means. Some problems such

as time consuming for parameters calibration, were identified to make the model work,

besides that the results show that the model can be used for prediction of monthly discharges,

sediment transported and snow pack with a high degree of accuracy.

Keywords: BASINS-HSPF, watershed model, simulation, discharge, sediment, snowpack

Introduction

A watershed is the geographic area of land where all of the water that is under it or drains off

it goes into the same outlet, this is a drainage catchment that divides the landscape into

hydrologically defined areas, Gibson, Carlson, Simpson, Smeltzer, Gerritson, Chapra,

Heiskary, Jones and Kennedy (2000). The U.S. is divided and subdivided into successively

smaller hydrologic units which are classified in 4 levels, Regions, Sub regions, Accounting

Units, and Cataloging Units, United States Geological Survey (USGS, 2006). Watershed

models typically simulate flow and associated water quality characteristics as a series of

hydrologic and hydraulic processes, these processes include surface runoff, Parajuli and

Ouyang (2013). Urbanization can alter the hydrology of a watershed, particularly the

magnitude and frequency of storm related flooding, quantifying changes in stream flow that

result from urbanization are critical for planning and hydraulic structures systems, detention

basins, and other storm water-management facilities Zarriello (1999). Because data on storm-

runoff volume and flood flow in specific areas are commonly unavailable, and future changes

in these flow characteristics that result from urbanization cannot be measured directly,

planners and engineers have come to rely on hydrologic models for this information Zarriello

(1999).

Hydrologic models are essential and effective tools for investigating the complex nature of

processes that affect surface and subsurface hydrology, soil erosion, and the transport of

chemical pollutants in catchments and for assessing the impacts of land use changes,

agricultural activities, and best management practices on these hydrologic processes, United

states Environmental Protection Agency (USEPA, 2001). For example, major watershed

restoration efforts are under way in Illinois to reduce sediment loads and nutrient

concentrations and to improve the ecosystem along the Illinois River and its tributaries Singh,


2

Knapp, Arnold and Demissie (2005). As part of restoration efforts, hydrologic models for the

Illinois River Basin are being applied and evaluated in the Illinois State Water Survey (Singh

et al. 2005). Part of the overall process in applying models for the Illinois River watershed is

a determination of which the models will perform best under varying watershed scales in

simulating hydrology, sediments, and nutrients Singh et al. (2005).

Many computer hydrologic models have been developed to simulate watershed hydrology

and water quality processes, the Hydrological Simulation Program FORTRAN (HSPF) is a

product of U.S. Environmental Protection Agency (USEPA, 2001), which is a comprehensive

model used for modeling processes related to water quantity and quality in watersheds of

various sizes and complexities Bicknell, Imhoff, Kittle, Donigian, and Johanson (2001), this

model is included within Version 3.0 of the modeling framework, referred as Better

Assessment Science Integrating Point and Nonpoint Sources- Hydrological Simulation

Program FORTRAN (BASINS-HSPF), developed by the United States Environmental

Protection Agency (USEPA, 2001).

BASINS-HSPF model uses information such as the time series of rainfall, temperature and

solar radiation; land vegetative cover characteristics such as land-use patterns; and land

management practices to simulate the processes that occur in a watershed (Bicknell et al.,

2001). The result of those simulations is a time series of the quantity and quality of runoff

from an urban or agricultural watershed such as flow rate, sediment load, and nutrient and

pesticide concentrations (Bicknell et al. 2001).

The importance of BASINS-HSPF come of the fact that is a comprehensive, continuous

model designed to simulate catchment hydrology and the associated water quality Duru

(1999). BASINS-HSPF system is designed so that the various simulation and utility modules

can be used conveniently, either individually or in group with a top-down approach that

emphasize structured design Dellman, Ruiz, Manwaring and Nelson (2002).

BASINS-HSPF considers a wide range of parameters that impact on hydrology and water

quality, therefore for Duru (1999) the application of the model could be wide. Evaluation of

BASINS-HSPF that is essentially equivalent to calibration of its parameters is, therefore, an

important effort toward understanding the model and bringing it into practical application

Duru (1999).

Dellman et al. (2002) considered that with the growing need to monitor the environment, it

became apparent that a method for rapid analysis of those effects was needed, BASINS-

HSPF is an analytical tool that apply mathematical models designed to simulate watershed

hydrology and water quality for both conventional and toxic organic pollutants in natural and

urban systems and predict possible environmental problems in the watershed (Dellman et al.

2002).

BASINS-HSPF model incorporates the watershed-scale Agricultural Runoff Model (ARM)

and Non-Point Source (NPS) models into a basin-scale analysis framework that includes

pollutant transport and transformation in stream channels (Bicknell et al. 2001). This set of

modules arranged in a hierarchical structure that must include a methodic data management

component, which permit the continuous simulation of a comprehensive range of hydrologic

and water-quality processes (Bicknell et al. 2001).

The objective of this study is to model the Mount Hope River catchment applying BASINS-

HSPF hydrologic model, calibrate the parameters of the model in such way that the simulated


3

monthly discharges match the observed discharges, and then apply BASINS-HSPF to

simulate sediment transported and snowpack, and compare the results with sediment

transported and snowpack calculated by other means.

Methods

Description of the study site

The Mount Hope river is tributary of the Thames River (Figure 1) has a total length of 23 km,

and the drainage area at the United States Geological Service (USGS 2005) gage # 01121000

is 74,0 km2 and is located close to the town of Warrenville in the northeast of the state of

Connecticut, New England Region, northeast of USA. The discharges gage datum is 102,3

meters above sea level National Geodetic Vertical Datum of 1929 (NG5D29) (USGS 2005).

a

0 25 5012.5 Kilometers

b

0 5 102.5 Kilometers

c

0 1 20.5 Kilometers

Figure 1: a) The west branch of the Thames River watershed (Mount Hope river is tributary

of Thames River) in black color in the state of Connecticut. b) The east branch of

the Thames river watershed with the Mount Hope river watershed in black color

and the Agronomy Farm site (black star). c) Mount Hope river watershed, and the

gage discharge site (black star).

Table 1, shows the vegetative cover of the Mount Hope River basin, where the percentages of

land use in the watershed are based on 1990 Land Use and Land Cover data calculated by

Bighinatti (2005).


4

Table 1: Attributes of Mount Hope River Basin (Bighinatti, 2005)

Monthly data for testing BASINS-HSPF were obtained for the years 1997 and 1998,

discharge data for the Mount Hope River were obtained from the USGS gage # 01121000, at

Warrenville (USGS, 2005). The discharge attributes are based on data from United States

Geological Survey website from 1941 to 2003 (USGS, 2005), the lowest recorded lowest

seven day average flow that occurs (on average) once every ten years (7Q10) in the Mount

Hope River at Warrenville was 0.011 m3/s on 1957, minimum discharge 0.071 m

3/s in 1958,

maximum discharge 6.853 m3/s on 1974, median discharge 0.425 m

3/s and mean discharge

0.850 m3/s (USGS, 2005).

Precipitation, potential ET, potential surface evaporation, air temperature, dew point

temperature, wind speed, and solar radiation data were obtained from the University of

Connecticut, Agronomy Farm located at 41°47'42" N and 72°13'42" W approximately 11,26

km from the Mount Hope gage (Figure 1). Table 2 summarizes the monthly mean values of

air temperature, humidity, solar radiation and precipitation in the Mount Hope River

watershed from 1958 to 2003.

Table 2: Monthly mean values of air temperature, humidity, solar radiation and precipitation

in the Mount Hope River watershed.

Month Air temperature Humidity Radiation Precipitation

[oC] [%] [W/m] [mm]

January -3 92 24 96

February -1 89 57 88

March 2 86 72 113

April 8 83 105 115

May 14 78 179 101

June 19 75 244 113

July 22 77 218 97

August 21 73 215 104

September 17 74 170 117

October 11 81 97 116

November 6 67 89 116

December -1 70 63 106

Land Use in the Watershed Value Unit

Barren Land 1.4 [%]

Forest 84.4 [%]

Non-forested Vegetation 8.3 [%]

Open Water 2.1 [%]

Urban 2.8 [%]

Wetland 1.0 [%]

Stratified Drift 4.2 [%]


5

A USGS digital elevation model (DEM) layer and a predigitized stream network data layer

from the National Hydrography Dataset (NHD) were used to setup BASINS-HSPF in the

Mount Hope River watershed (USGS, 2005). A digitized soil information layer from the

Natural Resources Conservation Service-State Soil Geographic Database (NRCS-STATSGO)

soil database and land use-land cover data layer from the Unites States Geological Service-

Geographic Information Retrieval and Analysis System (USGS-GIRAS) database were used

for further sub classification of areas in the watershed (USGS, 2005).

BASINS-HSPF model application in the Mount Hope River watershed

Based on its topography and existing stream network, the Mont Hope River watershed was

divided into 12 smaller, hydrologically connected sub watersheds and associated stream

reaches using the automatic delineation tool GIS interface.

Each sub watershed of the Mount Hope River watershed was partitioned into pervious and

impervious areas based on land uses such as urban, agriculture, forest, barren, and

wetland/water areas. Since BASINS-HSPF did not automatically create segments based on

soil types, the dominant soil (Hydrologic Soil Group B) was assumed to be representative of

the Mount Hope River watershed soil conditions, such an approach has been used in some

previous BASINS-HSPF studies (Donigian et al. 1984) and Jones and Winterstein (2000).

Because major hydrologic differences occur between pervious and impervious land use types

and since forest is the major pervious land use in the Mount Hope River watershed, all

pervious land segments in the model were assigned the same hydrologic parameters. The

storage routing scheme was used for channel routing. The function tables (FTABLES) were

not modified, and the default volume-discharge relationship was used as calculated by the

model based on DEM and NHD data.

The Muskingum routing scheme was used for channel routing and the Soil Conservation

Service (SCS) curve number method was used for runoff simulation; thus, only the daily

precipitation and maximum and minimum air temperature data were input from the climate

station.

The entire Mount Hope River watershed was assumed to be totally forested in character so

that the IMPLND module was not applicable. Thus, only the parameters required to operate

PERLND and RCHRES modules for runoff simulation were calibrated.

During the calibration process in this study Bicknell et al. (1996) specifies the numerical

range of BASINS-HSPF model parameters and Donigian and Davis (1978) provides general

evaluation guidelines for setting numerical values of the parameters. Furthermore, Donigian,

Imhoff, Bicknell and Kittl. (1984) describes the entire application process of BASINS-HSPF,

to demonstrate the decisions, procedures, and results involved in a typical application.

Furthermore Donigian, Beyerlein, Davis and Crawford (1977) present a thorough discussion

of the parameters in the application of the BASINS-HSPF to agricultural watersheds in

Georgia and Michigan. Initial parameter values in this study were nominated by integrating

information from the above sources with information from earlier BASINS-HSPF related

experiences such as provided by Moore, Matheny, Tyre, Sabatini and Klaine (1988) and

Chew, Moore, and Smith (1991).


6

The parameters and the ranges within which their values were varied were selected based on

other BASINS-HSPF evaluation studies, specifically Chew et al. (1991); Jones and

Winterstein (2000) and the BASINS Technical Note 6 (USEPA, 2001).

The descriptive equations used to represent these processes in BASINS-HSPF operation are

given by Bicknell et al. (1996) and the parameters required to execute BASINS-HSPF model

for sediment simulation can be grouped under the four categories: watershed, meteorological,

hydrological, and sediment are given by Wang, Duru, Hjelmfelt, Qiu and Thompson (1999).

The SNOW module simulates snowmelt contributions from the land surface derived from the

fall, accumulation, and melting of snow. This module is necessary for a complete hydrologic

package since much of the runoff, particularly in the northern half of the United States, is

directly related to snow conditions. Two options are available for modeling the processes

involved in snow accumulation and melt on a land segment: an energy balance and a degree-

day approach. The energy balance approach is based on work by the US Corps of Engineers

(USACE, 1956), Anderson and Crawford (1964), and Anderson (1968). Empirical

relationships are employed when physical relationships are not well understood. The snow

algorithms use meteorological data to: determine whether precipitation occurs as rain or

snow; simulate an energy balance for the snow pack; determine the effect of the heat fluxes

on the snow pack, and; calculate the resulting melt that reaches the land surface.

The second snowmelt method uses a temperature index, or "degree-day" approach. Most

processes from the snow algorithms are employed; however, snowmelt due to atmospheric

heat exchange is calculated using the air temperature and an empirical degree-day factor.

This approach minimizes the requirements for meteorological data to precipitation and air

temperature. The reader is referred to Rango and Martinec (1995) for a summary of the

degree-day method. SNOW may require six meteorological time series for each land segment

simulated, depending on the option chosen. They are:

A value from each of these time series is input to SNOW at the start of each simulation

interval. However, some of the meteorological time series are only used intermittently for

calculating rates. One such example is the calculation of the potential rate of evaporation

from the snow pack.

Air temperature is used to determine whether precipitation is falling as rain or as snow. The

critical temperature TSNOW may be automatically adjusted upward by up to one degree in

unsaturated conditions, based on the dew point. This adjustment is optional when using the

temperature index method, and is made only if the input dew point time series is supplied

(Bicknell et al., 2001).

Sediment transport calculation

Mtalo, Killingtveit and Ndomba (2008) in their research presented a sediment rating curve

given by Equation 1 is a regression line with coefficients chosen for a rising flow condition

for the Mount Hope application with square-r equal to 0.63.


7

bQa*ansportSedimentTr [1]

Where:

Q Observed discharges

a 0.2286

b 2.2271

Snow Water Equivalent (Snowpack) calculation

The Snow water equivalent (SWE) was calculated applying Maidment (1993) methodology,

using the relationship between surface air temperature and water density curve and the

relationship between SWE water density and depth of water (Equation 2).

SS ρ*d*0.01SWE [2]

Where:

SWE Snow Water Equivalent

ds Water depth

ρS Water density

Statistical analysis

The simulated stream discharges, sediment transported and snowpack for the years 1997 and

1998 along the Mount Hope River were compared with observed stream water temperatures

and calculated sediment transported and snowpack using the Nash - Sutcliffe model of

efficiency (Nash and Sutcliffe 1970) given by Equation 3 and linear regression.

n

1i

2

i

n

1i

2

ii

)OO(

)SO(

1NS [3]

Where:

Oi Observed discharges

O Mean of observed discharges

Si Simulated discharges

n Number of steps modeled

Nash–Sutcliffe efficiencies can range from −∞ to 1. An efficiency of 1 corresponds to a

perfect match between simulated and observed data, therefore that the model is very accurate.

An efficiency of 0 indicates that the model predictions are as accurate as the mean of the


8

observed data, an efficiency less than 0 occurs when the observed mean is a better predictor

than the simulated, Nash and Sutcliffe (1970).

Results and Discussion

Hydrologic calibration

The hydrologic components of BASINS-HSPF were calibrated to fit the observed daily

discharge data for one year period from October 1997 to September 1998. Initial values of the

model parameters were varied iteratively within a reasonable range during the calibration

runs until the maximum coefficient of determination square-r between observed and

simulated discharges data was obtained.

Figure 2, shows the observed and simulated monthly discharges at the Mount Hope River

from October 1997 to September 1998. The model showed a high sensitivity to the variations

in the intensity of the rain but failed to achieve a close value of the observed peak discharge

in event in March 1997 with one peak discharge of 5.5 m3/s for the observed and 3.4 m

3/s for

simulated and other event in June 1998 with almost the same peak discharges, 3.7 m3/s for

the observed and 3.6 m3/s for the simulated peak discharges. The shape of the concentration

and recession curves of the simulated events follows the trends of the observed discharges.

Figure 2: Monthly observed and simulated discharges at the Mount Hope River from October

1997 to September 1998.

Figure 3, shows the monthly the calculated and simulated sediment transported at the Mount

Hope River from October 1997 to September 1998. The model shows a high sensitivity to the


9

variations in the intensity of the discharges and achieve similar values for the sediment

transported event in March 1998 with sediment transported of 113 mg/l for the calculated and

129 mg/l for simulated. The shape of the concentration and recession curves of the simulated

events follows the trends of the calculated sediment transported.

Figure 3: Monthly calculated and simulated sediment transported at the Mount Hope River

from October 1997 to September 1998.

Figure 4, shows the monthly the calculated and simulated sediment transported at the Mount

Hope River from October 1997 to September 1998. The model shows a high sensitivity to the

variations in the temperature and achieve similar values for the snowpack accumulated event

in December1997 with a snowpack of 77.3 mm for the calculated and 86.2 mml for

simulated. The shape of the concentration and recession curves of the simulated events

follows the trends of the calculated snowpack.


10

Figure 4: Monthly calculated and simulated snowpack at the Mount Hope River from

October 1997 to September 1998.

Parameters and statistical analysis

Table 3 shows the linear regression slope and square-r, and Nash – Sutcliffe coefficient (NS)

between the simulated and observed discharges, sediment transported and snowpack.

Table 3: Linear regression slope and square-r, and NS for discharges, sediment and

snowpack.

Discharges Sediment transported Snowpack

Square-r 0.78 0.78 0.76

NS 0.72 0.77 0.62

slope 0.59 0.83 0.76

Table 4 shows the peaks obtained between simulated and observed-calculated discharges,

sediment and snowpack.

Table 4: Peaks between simulated and observed-calculated discharges, sediment and

snowpack.

Discharges Sediment transported Snowpack

Simulated 3.4 129 86

Observed-Calculated 5.5 113 77


11

The simulation performed applying BASINS-HSPF at the Mount Hope River from October

1997 to September 1998, shows a high degree of accuracy between simulated and observed

discharges and simulated and calculated sediment transported and snowpack.

The results after calibration between the simulated and observed discharges performed a

square-r over 0.78 and Nash – Sutcliffe coefficient over 0.72. The trend line for the simulated

discharges shows that the model underestimated the discharges, with a slope of 0.59.

The results after calibration between the simulated and observed sediment transported

performed a square-r over 0.78 and Nash – Sutcliffe coefficient over 0.77. The trend line

between the simulated and calculated sediment transported shows that the model predicts

accurately the sediment transported, with a slope over 0.83.

The results after calibration between the simulated and observed snowpack performed a

square-r over 0.76 and Nash – Sutcliffe coefficient over 0.62. The trend line between the

simulated and calculated snowpack shows that the model predicts accurately the sediment

transported, with a slope over 0.76.

Conclusions

The hydrologic simulation model BASINS-HSPF is one of the most commonly used

hydrologic models, due to its complexity BASINS-HSPF requires extensive data input for an

accurate simulation. The agreement between simulated and observed monthly discharges,

sediment transported and snowpack was good with a square-r coefficient of 0.76 and Nash –

Sutcliffe coefficient over 0.62.

Modeling field activities with BASINS-HSPF is very difficult because the variables and the

data that determine the magnitude of discharges and other parameters are difficult to find and

have control when the BASINS-HSPF files are prepared. The discharges generated by

BASINS-HSPF model follow the trend of discharge observed in monthly simulation. The

snow and sediment simulated by BASINS-HSPF couldn’t be compared with observed data,

because there were not, but the trend of this simulation and the calculated values have a high

degree of relationship.

References

1) Anderson, E.A., and N.H. Crawford. 1964. The Synthesis of Continuous Snowmelt

Runoff Hydrographs on a Digital Computer. Technical Report 36. Department of

Civil Engineering, Stanford University, Stanford, California, 103 p.

2) Anderson, E.A. 1968. Development and Testing of Snow Pack Energy Balance

Equations, Water Resources Research, 4(1): 19-37.

3) Bicknell, B.R., Imhoff, J.C., Kittle, J.L., Jr., Donigian, A.S., Jr. and Johanson, R.C.

2001. Hydrological Simulation Program -- FORTRAN, User's manual for Release 12.

U.S. Environmental Protection Agency, EPA/600/R-97/080, Environmental Research

Laboratory, Athens, GA.

4) Bighinatti, S.J. 2005. Investigations of flow-duration curves and application to

estimating discharge on ungauged streams. Master Thesis. Storrs, University of

Connecticut.


12

5) Chew, Y. C., L.W. Moore, and R. H. Smith. 1991. Hydrological Simulation of

Tennessee=s North Reelfoot Creek Watershed. Res. Jour. Water Pollut. Control Fed.

63(10).

6) Dellman, P. N., Ruiz, C. E., Manwaring, C. T., and Nelson, E. J. 2002. Watershed

Modeling System Hydrological Simulation Program; Watershed Model User

Documentation and Tutorial (No. ERDC/EL-SR-02-1). Engineer Research and

Development Center Vicksburg MS Environmental Lab.

7) Donigian, A. S., D. C. Beyerlein, H. R. Davis, and N. H. Crawford. 1977. Agricultural

Runoff Management (ARM) Model Version 11: Refinement and Testing, U. S.

Environmental Protection Agency, Environmental Research Laboratory, Athens,

Georgia.

8) Donigian, A. S. and H. H. Davis. 1978. User’s Manual for Agricultural Runoff

Management (ARM) Model EPA-600/3-78-080. U. S. Environmental Protection

Agency, Environmental Research Laboratory, Athens, Georgia.

9) Donigian, A. S., J. C. Imhoff, B. R. Bicknell, and J. K. Kittl. 1984. Application Guide

for Hydrological Simulation Program - FORTRAN (HSPF), EPA-600/3-84-065, U. S.

Environmental Protection Agency, Environmental Research Laboratory, Athens,

Georgia.

10) Duru, J. O. 1999. Evaluating HSPF for simulating sediment yield from a Claypan

agricultural watershed in Central Missouri. The Society for Engineering in

Agricultural, Forest, and Biological Systems. In International Meeting (pp. 18-21).

11) Gibson, G.R., R. Carlson, J. Simpson, E. Smeltzer, J. Gerritson, S. Chapra, S.

Heiskary, J. Jones, R. Kennedy. 2000. Nutrient criteria technical guidance manual:

lakes and reservoirs (EPA-822-B00-001). U.S. Environmental Protection Agency,

Washington, D.C.

12) Jones, P.M. and T.A. Winterstein. 2000. Characterization of Rainfall-Runoff

Response and Estimation of the Effect of Wetland Restoration on Runoff, Herone

Lake Basin, Southwestern Minnesota, 1991-97. USGS, WRIR 00-4095. Mounds

View, Minnesota.

13) Maidment, D.R. 1993. Handbook of Hydrology. Mc Graw-Hill, Inc., 1424 p., New

York.

14) Moore, W., H. Matheny, T. Tyre, D. Sabatini, and S. J. Klaine. 1988. Agricultural

runoff modeling in a small west Tennessee watershed. Research Journal Water

Pollution Control Federation 60(2): 242-249.

15) Mtalo, F. W., Killingtveit, Å., & Ndomba, P. M. 2008. Developing an Excellent

Sediment Rating Curve from One Hydrological Year Sampling Programme Data:

Approach. Journal of Urban and Environmental Engineering, 2(1), 21-27.

16) Parajuli, P. B., and Ouyang, Y. 2013. Watershed-scale hydrological modeling

methods and applications. INTECH Open Access Publisher.

17) Rango, A., and Martinec, J. 1995. Revisiting the degree‐day method for snowmelt

computations. Journal of the American Water Resources Association, 31(4): 657-669.


13

18) Singh, J., Knapp, H. V., Arnold, J. G. and Demissie, M. 2005. Hydrological modeling

of the iroquois river watershed using HSPF and SWAT. Journal of the American

Water Resources Association, 41(2): 343-360.

19) U.S. Army Corps of Engineers (USACE). 1956. Snow Hydrology, Summary Report

of the Snow Investigations, North Pacific Division. Portland, Oregon, 437p.

20) U.S. Environmental Protection Agency (USEPA). 2001. BASINS-HSPF Version 3.1,

User's Manual. EPA-823-B-01-001, Washington, D.C.

21) United States Geological Survey (USGS). 2005. Daily Streamflow for Connecticut.

http://waterdata.usgs.gov/nwis/nwisman/?site_no=01121000&agency_cd=USGS

22) United States Geological Survey (USGS). 2006. http://water.usgs.gov/ GIS/huc.html

23) Wang, M., Duru, J. O., Hjelmfelt, A. T., Qiu, Z., and Thompson, A. 1999.

Hydrological Simulation of a Claypan Watershed Using HSPF.

http://www.fse.missouri.edu/home/hjelmfelta/Publications_files/PDF/seattleasce.pdf

24) Zarriello, P. J. 1999. Watershed modeling approach to assessing the hydrologic effects

of future development in the Ninemile Creek Basin, Onondaga County, New York.

US Department of the Interior, US Geological Survey.

Modelling of the Mount Hope River watershed with BASINS ...journalsonline.org/american-journal-of-engineering...Dellman et al. (2002) considered that with the growing need to monitor

Documents