Water Balance Model (Crawford)

In implementing Small hydropower development the main problem caused by

insufficient investigations are due to inaccurate calculation of the runoff or improper
adoption of water flow for generating power. The installed capacity of power station
become abnormally large or small, so it cause to either the utilization factor of
equipment dropped below a certain level or hydro power potential had not been
utilized effectively.
The hydrologic analysis is key for determining the project output and subsequent
economic benefits. A first step is the determination of the average monthly stream
flows near the proposed site. Where records do not exist, or for short period only, it may
be necessary to synthesize stream flows by correlation with near by stream gauging or
precipitation stations.
Among the hydrologic analysis the flow duration curve is the most basic analysis for
designing a hydropower scheme. The flow duration curve is a most useful tool for
evaluating low flows at a hydropower site. These data can be used to estimate the flow
that can be used for power production 90 to 95 percent of the time. This gives an
immediate indication of the reliability of the power production at a site. Flow-duration
curves are based on continuous stream flow data. On ungaged streams where stream
flow measurements are not available (in most of the SHP sites), precipitation and
potential evapo-transpiration records can be used to calculate continuous flows. The
calculation mimic key hydrologic processes: infiltration of water into the soil profile,
surface runoff, and flow along subsurface flow paths into the stream.
When calculated monthly stream flows are found, they are used like observed flows to
calculate flow duration curve. A sketch of the calculations is shown in Figure 3.11. The
following steps are used to calculate monthly runoff and a flow duration curve from
meteorologic data:

• Assemble five or more years of concurrent rainfall and potential

evapo-transpiration data.
• Estimate the watershed characteristics of the basin.
• Calculate, using a tabular form, the monthly stream flows for five or more years
based on these rainfall and potential evapo-transpiration data.
• Calculate the flow duration curve using the calculated monthly stream flows, as
would be done if observed data were available.

1
Rainfall and Potential Evapotranspiration Data
Streamflow calculations require data on monthly rainfall on the watershed and
monthly potential evapotranspiration. Rainfall data that is observed in or near the
watershed must be found. National meteorologic services or agricultural agencies
maintain records of meteorologic stations. Records at individual stations are often
incomplete. Records from two or three stations are needed to fill in missing records.
Adjustments are usually needed to estimate watershed rainfall from precipitation-gage
records. Rainfalls often increase with elevation, and gages are usually placed in
villages along river valleys. Meteorologic and agricultural services prepare maps of
mean annual rainfall. These maps can be used to adjust gaged rainfall and to estimate
rainfall on the watershed.
Potential evapotranspiration (PET) is the amount of water that would evaporate from
the watershed if water supply is ample. The actual water loss, called actual
evapotranspiration (AET), is less than or equal to the potential evapotranspiration.
The potential evapotranspiration may be estimated and published by a national
meteorologic services office. Potential evapotranspiration is quit uniform from year to
year. A mean monthly distribution of potential evapotranspiration will be sufficient for
purposes of calculating monthly runoff.

Estimating Watershed Characteristics

The tabular calculations of monthly flows use three coefficients that represent
watershed characteristics:
NOMINAL = an index to the soil moisture storage capacity in the watershed, mm(in).
PSUB = the fraction of runoff that moves out of the watershed as baseflow or

2
groundwater flow, dimensionless.
GWF = an index to the rate of discharge from the groundwater storage to the stream,
dimensionless.
NOMINAL, PSUB and GWF are watershed characteristics that will change from one
watershed to another. NOMINAL is the soil moisture storage level that permits half of
any positive monthly water balance to leave the watershed as excess moisture, where
excess moisture is either direct runoff or groundwater flow. The soil moisture storage
level varies and may be less than or greater than NOMINAL. When the soil moisture
storage is less than NOMINAL, the majority of any positive monthly water balance is
retained in the soil moisture. When the soil moisture storage is greater than
NOMINAL, the majority of any positive monthly water balance becomes direct
runoff or and addition to the groundwater storage.
PSUB is the faction of runoff that moves out of the watershed on subsurface flow
paths rather than as direct or surface runoff. The total flow that a watershed provides
consists of surface or direct runoff that creates the peak flows and subsurface flows
that provide the low flows. Low-permeability soils that have low-infiltration
capacities yield large amounts of surface or direct runoff and low sustained discharges.
Soils that have high-infiltration capacities yield higher sustained discharges.
Therefore streams that have high minimum discharges are those with highly
permeable sandy soils and fractured or permeable subsurface geology.
GWF is an index to the time of flow along subsurface flow paths that enter the stream.
It is the fraction of the total volume of water on groundwater flow paths that will enter
the stream in the current month.
These watershed characteristics can be estimated for ungaged streams using the
following guidelines.
NOMINAL = 100 + C × mean annual precipitation
Where C is approximately 0.2 in watersheds where precipitation occurs throughout
the year and 0.25 in watersheds with seasonal rainfall. The value of NOMINAL can
be reduced by up to 25 percent in watersheds with limited vegetation and thin soil
cover.
PSUB = 0.6 median value
PSUB would increase to 0.8 in watersheds known to have highly permeable soils and
would decrease to 0.3 in watersheds with low permeability or thin soils.
GWF = 0.5 median value
GWF would increase to 0.9 in watersheds that have little sustained flow and would
decrease to 0.2 in watersheds known to have reliable sustained flows.
Watershed characteristics can be estimated from limited field data based on historic
streamflows. The characteristic that controls runoff volumes is NOMINAL.
Increasing NOMINAL will decrease runoff and decreasing NOMINAL will increase
runoff. The actual evapotranspiration (AET) increase when NOMINAL is increased,
and the AET/PET ratio in Table 1 will change. When some historic flow data is
available, the calculations in Table 1 can be repeated for different values of
NOMINAL until the calculated flow volume and the historic flow volumes agree. In
some cases, the rainfall amounts that are assumed to occur on the watershed may need

3
to be adjusted. A change in rainfall on a watershed of ±10 percent will change
watershed runoff by at least 10 percent.
The fraction of excess moisture that moves as subsurface flow (PSUB) and the rate of
outflow of subsurface or groundwater flow (GWF) control the low streamflows
between storms. A value of GWF can be found when little rainfall occurs for 2 or
more months. When rainfall does not occur, GWF = 1 – (streamflow volume in
current month/streamflow volume last month). Alternatively, if monthly
measurements of streamflow are made during a period when rainfall does not occur,
GWF = 1 – (streamflow today/streamflow 1 month ago). Calculated values for GWF
will vary, so GWF should be calculated for different months when observed flows are
available. The lowest calculated value of GWF is usually the most reliable estimate.
When NOMINAL and GWF are established, the value of PSUB can be considered.
PSUB increases or decreases the volume of water moving on the subsurface flow
paths. If, for example, the following monthly streamflows are available from field
measurements and calculations, PSUB needs to be increased.
Streamflow June July August September
Observed 1,070 642 385 231
Calculated 720 432 259 155
In this example, the calculated flows are all too low and more subsurface flow volume
is needed. Note that GWF for both the calculated and the observed monthly flows is
0.4, so changing GWF would not improve the calculated streamflows.
The trial calculations to establish watershed characteristics that are outlined above can
be carried out for nearby streams and the watershed characteristics that are found may
be assumed to apply to the ungaged stream that is being studied. Watershed
characteristics usually change only moderately in a region and changes will correlate
with changes in vegetation, soils, and subsurface geology.
Calculations of Monthly Runoff
Calculations of monthly runoff are made using a tabular sheet, Table 3.4A
step-by-step procedure for computing data in the table follows. These calculations can
be done with a hand calculator or adding machine.
Initial Steps
Select the values of the coefficients representing watershed characteristics and enter
them in the column headings: NOMINAL above column (5) , PSUB above column
(13), and GWF above column (16)
Initial or starting conditions are needed for the soil moisture storage [column (4)] and
the groundwater storage [column (14)]. If rainfall is seasonal and the tabular
calculation begins in the dry season, the initial storages will be low. Typical values
would be an initial soil storage of 10 percent of

4
Table 1 Flow-Duration Model
1 2 3 4* 5 6 7 8 9
DATE PRECIP PET MOIST STOR PRECIP/ AET/ AET WATER
mo/yr mm mm STORAGE RATIO PET PET mm BALANCE
mm mm
1/80 356.3 21.7 500.0 1.22 16.42 1.00 21.7 334.6
2/80 196.4 38.4 601.9 1.47 5.12 1.00 38.4 158.1
3/80 145.6 79.1 624.3 1.52 1.84 1.00 79.1 66.5
4/80 59.8 118.3 631.9 1.54 .51 .89 104.9 -45.1
5/80 26.9 155.4 586.7 1.43 .17 .76 118.8 -92.0
6/80 11.6 171.5 494.8 1.21 .07 .63 108.1 -96.5
7/80 19.9 191.6 398.3 .97 .10 .54 103.3 -83.4
8/80 33.7 154.7 314.9 .77 .22 .52 80.2 -46.5
9/80 17.2 137.9 268.5 .65 .12 .41 56.7 -39.5
10/80 299.6 88.9 228.9 .56 3.37 1.00 88.9 210.7
11/80 275.8 41.7 406.8 .99 6.61 1.00 41.7 234.1
12/80 350.0 29.8 525.7 1.28 11.74 1.00 29.8 320.2

10 11 12 13++ 14ξ 15 16 17 18
EXCESS EXCESS DELTA RECHG BEGIN END GW DIRECT TOTAL
MOIST MOIST STORAGE TO GW, STOR STOR FLOW FLOW DISC
RATIO mm mm mm GW, GW, mm mm mm
mm mm
.70 232.7 101.9 141.9 25.0 166.9 106.8 90.7 197.6
.86 .35.7 22.4 82.8 60.1 142.9 91.4 52.9 144.4
.89 58.9 7.6 35.9 51.4 87.4 55.9 23.0 78.9
.00 .0 -45.1 .0 31.5 31.5 20.1 .0 20.1
.00 .0 -92.0 .0 11.3 11.3 7.2 .0 7.2
.00 .0 -96.5 .0 4.1 4.1 2.6 .0 2.6
.00 .0 -83.4 .0 1.5 1.5 .9 .0 .9
.00 .0 -46.5 .0 .5 .5 .3 .0 .3
.00 .0 -39.5 .0 .2 .2 .1 .0 .1
.16 32.8 117.9 20.0 .1 20.1 12.9 12.8 25.7
.49 115.2 118.9 70.3 7.2 77.5 49.6 44.9 94.5
.74 237.7 82.5 145.0 27.9 172.9 110.6 92.7 203.3
* NOMINAL = 410
+ Initial conditions MOIST STORAGE = 500, BEGIN STOR GW = 25.0
++ PSUB = 0.61
ξGWF = 0.64
Note: EXCESS MOIST RATIO in col. (10) was entered two significant figures. EXCESS MOIST in col. (11) was
calculated on a small calculator that retained more than two significant figures for the EXCESS MOIST RATIO

5
Column Comment
(1) Enter the month and year of the data
(2) Enter precipitation (PRECIP) on the watershed for the month (gage
rainfall must be adjusted to represent watershed rainfall)
(3) Enter potential evapotranspiration (PET) on the watershed.
(4) MOIST STORAGE was entered previously as an initial condition or is
MOIST STORAGE plus DELTASTORAGE, column (4) plus column
(12) from the prior month calculation.
(5) Calculate the soil storage ratio (STOR RATIO), the value in column (4)
divided by NOMINAL
(6) Calculate the ratio PRECIP/PET, column (2) divided by column (3).
Enter the result in column (6)
(7) Enter Fig. 2 with PRECIP/PET and the STOR RATIO [column(5)], and
find the value of the ratio of actual to potential evapotranspiration,
AET/PET. Enter the value of AET/PET in column (7)
(8) Calculate AET as PET multiplied by the AET/PET ratio, column (3)
times column (7), and enter AET in column (8).
(9) Calculate the WATER BALAMCE for the month where WATER
BALANCE = PRECIP – AET [column (2) less column (8)]. Enter the
result in column (9).
(10) If the WATER BALANCE is positive, enter Fig. 3 with the STOR
RATIO (column (5)). And find the excess moisture ration (EXCESS
MOIST RATIO). If the WATER BALANCE in column (9) is negative,
the EXCESS MOIST RATIO is zero. Enter the EXCESS MOIST
RATIO in column (10).
(11) Calculate the excess moisture (EXCESS MOIST) where EXCESS
MOIST is the EXCESS MOIST RATIO times the WATER BALANCE,
column (10) times column (9). Enter the result in column (12)
(12) Calculate the change in soil storage (DELTA STORAGE) where DELTA
STORAGE is WATER BALANCE minus EXCESS MOIST column (9)
minus column (11). Enter the result in column (12)
(13) Calculate the recharge to groundwater storage (RECHG TO GW) as
PSUB times EXCESS MOIST, PSUB times column (11). Enter the
result in column (13)
(14) BEGIN STOR GW was entered previously as an initial condition or is
END STOR GW less GW FLOW, column (15) less column (16) from
the prior month calculation.
(15) Calculate the end of month groundwater storage (END STOR GW) by
adding the RECHG to GW in column (13) to the BEGIN STOR GW in
column (14). Enter the result in column (15)
(16) Calculate the groundwater discharge to the stream (GW FLOW) where
GW FLOW is GWF times END STOR GW or GWF times column (15).
Enter the result in column (16).
(17) Calculate the direct runoff to the stream (DIRECT FLOW), where

6
 DIRECT FLOW is EXCESS MOIST less RECHG TO GW, column (11)
minus column (13). Enter the result in column (17)
(18) Calculate the streamflow as the DIRECT FLOW plus GW FLOW,
column (17) plus column (16). Enter the result in column (18). This
result is in millimeters and can be converted to cubic meters for the
month by multiplying by the watershed area in square kilometers times
1000.
NOMINAL, and groundwater storage of 5 percent of NOMINAL. In watersheds with
uniform rainfall throughout the year, an initial soil storage equal to the NOMINAL,
and an initial ground water storage of 20 percent of NOMINAL would be expected.
Calculation that begin in the wet season would have an initial soil storage of 125
percent of NOMINAL and an initial groundwater storage of 40 percent of
NOMINAL.
The effects of initial storages in a water-balance calculation "damp-out" over a period
of 6 to 12 months. There are two ways of avoiding bias caused by unknown initial
storages: The first year of calculated monthly flows could be ignored and not included
in the flow-duration curve. Alternatively, the values of soil and groundwater storage
that are found at the end of the first year of calculations could be accepted as
representative of typical storages at that time of year. These storages could then be
entered as the best estimate of initial conditions for the watershed and calculations
could be redone using these storages.
Enter the initial soil moisture in column (4), and the initial groundwater storage in
column (14)
Reset Initial Conditions for the Next Month
The BEGIN STOR GW for the next month is END STOR GW less GW FLOW,
column (15) minus column (16). Enter this result in column (14) for the next month.
The beginning MOIST STORAGE for the next month is the current MOIST
STORAGE plus DELTA STORAGE, column (4) plus column (12). Enter this result in
column (4) for the next month.
The tabulation steps are repeated until all the months of precipitation and potential
evapotranspiration have been considered. An example of a completed table for
monthly flows is given in Table 1. Note that in Table 1 the EXCESS MOIST RATIO
in column (10) has been entered as two significant figures. The EXCESS MOIST in
column (11)_ was calculated on a small calculator that retained more than two
significant figures for the EXCESS MOIST RATIO.

7
8