Introduction to HEC-HMS

CE 394K.2 Surface Water Hydrology

Univerity of !e"a at #utin - S$ring% &999
're$ared (y
)avid *. Maid+ent and Set, #,ren
!a(le of Content
Goals of the Exercise
Obtaining the Program and the Data
Procedure
o 1. Importing a Basin Model
o 2. Editing a Basin Model
o . !reating a Precipitation Model
o ". Defining the !ontrol #pecifications
o $. Executing an %E!&%M# Model
o '. (ie)ing %M# *esults
Be !reati+e,, -.aller !ree/ di+ersion tunnel design discharge0
!leaning 1p
-oal of t,e E"ercie
2he intent of this exercise is to introduce 3ou to the structure and some of the functions
of the %E!&%3drologic Modeling #3stem -%E!&%M#04 b3 simulating the runoff
h3drographs resulting from a design storm on .aller !ree/ in 5ustin4 2exas. 2he
ph3sical parameters describing the )atershed )ere de+eloped pre+iousl3 using the
!*.*&PrePro program.
.(taining t,e 'rogra+ and t,e )ata
2he %E!&%M# program can be obtained from the %3drologic Engineering !enter6s
home page at7 http788))).)rc&hec.usace.arm3.mil8
2he program runs on .indo)s 9$4 :2 and 1nix platforms. 5 user6s manual is also
a+ailable at this location. 2he program is loaded in the ;*! in a number of computers in
*m ."<<.
2o run the model for .aller !ree/4 a basin file is needed to specif3 the ph3sical
parameters of the )atershed4 and a map file to gi+e the outline of the drainage areas and
cree/s. 2hese files can be do)nloaded from here7 .aller=!/=1.basin and hms.map
2hese files can also be obtained on the 8class ser+er in the ;*! under
8class8maidment8gish3dro998hmsintro4 and also can be do)nloaded as a >ip file7
)aller.>ip. Ma/e a )or/ing director3 on 3our computer and do)nload these files into it.
'rocedure
?ou ma3 start the %M# b3 clic/ing
Start/'rogra+/,+
5fter a fe) seconds4 a )indo) similar to the follo)ing image should appear7
%enceforth4 this image )ill be referred to as the 'ro0ect )efinition. 5 'ro0ect in %M#
refers to all of the data sets associated )ith a particular model. In this case4 the 'ro0ect is
called ten14 )hich is short for ;a/e 2en/iller. 2his data set is distributed as a standard
example file )ith %E!&%M#. 2he )ecri$tion bar underneath the 'ro0ect name allo)s
3ou to t3pe a detailed name for the actual short 'ro0ect name. If desired4 3ou ma3 clic/
on the ellipsis to the right of the )ecri$tion bar4 and additional space for 3ou to t3pe a
length3 )ecri$tion )ill appear. @or future reference4 an3 time 3ou see an ellipsis in a
)indo) in the %M#4 it means 3ou ma3 access additional space for )riting descripti+e
text.
In the Co+$onent section of the 'ro0ect )efinition4 there are three sub&sections&&the
2ain Model4 the 'reci$itation Model4 and the Control S$ecification. Each
component represents a different element of the model. 2he 2ain Model4 for instance4
contains information rele+ant to the ph3sical attributes of the model4 such as basin areas4
ri+er reach connecti+it34 or reser+oir data. ;i/e)ise4 the 'reci$itation Model holds
rainfall data. @inall34 the Control S$ecification section contains information pertinent to
the timing of the model such as )hen a storm occurred and )hat t3pe of time inter+al 3ou
)ant to use in the model. Each of the sections is explored belo) indi+iduall3.
&. I+$orting a 2ain Model
2he first of the Co+$onent )e )ill consider is the 2ain Model. 2o create a h3drologic
model of .aller !ree/4 )e need to import the basin file that 3ou Aust do)nloaded. In
%M# ProAect .indo)4 use 3ile/4e5 'ro0ect to open a ne) proAect b3 entering the
follo)ing data. :otice that t3ping B)allerB in the ProAect )indo) automaticall3 creates a
subdirector3 called )aller in c7ChmsproA4 )hich is )here ProAect files are stored )hen
%M# is executed.
In the %M# ProAect )indo)4 use Edit/2ain Model/I+$ort to import the basin model
from the director3 c67te+$75or1ing. ?ou6ll see the Basin )indo) open up and some
icons appear. ?ou ma3 be as/ed for the location of the map file. In this case4 in the %M#
#chematic )indo)4 use 3ile/2ain Model #ttri(ute to select the location )here 3ou
ha+e stored the map file. ?ou should no) see a schematic of .aller !ree/ sho)ing the
)atershed and stream map and an o+erla3 of the h3drologic elements.
2he tool bar on the left hand side of the displa3 sho)s the se+en h3drologic elements
contained in %E!&%M#. 2he3 are7
#ubbasin & rainfall&runoff computation on a )atershed
*i+er reach & routing of flo)s from one end of a reach to the other
*eser+oir & routing of flo)s through a le+el&pool reser+oir
Dunction & combination of flo)s from upstream reachs and subbasins
Di+ersion & abstraction of flo) from the stream
#ource & inflo) of )ater from a stream crossing the boundar3 of the modeled
region
#in/ & outflo) of )ater in a stream crossing the boundar3 of the modeled region
-basin outlet0
2he model of .aller !ree/ sho)n abo+e contains onl3 " of these /inds of elements.
2here are 19 h3drologic elements in the model4 made up of E subbasins4 ' ri+er reaches4 $
Aunctions4 and 1 sin/ at the point )here .aller !ree/ flo)s into the !olorado *i+er.
:otice that )hen a stream flo)s through a )atershed4 the additional local runoff from the
drainage area around the stream is not accounted for until the do)nstream end of the
reach )here its flo) is combined at a Aunction )ith the flo) coming from the upstream
reach. 2he Aunctions ha+e been located at points )here roads cross .aller !ree/.
To be turned in: a map of the Waller Creek HMS model
2o screen&capture an image4 do the follo)ing7
Ma/e sure the Sc,e+atic is the top )indo) in 3our terminal.
Depress B!trl&Print #creenB on 3our /e3board. 2his captures the image on 3our
screen and places it on the clipboard.
Open Paint using Start/'rogra+/#cceorie/'aint.
In Paint4 select 'ate from the Edit menu. 5fter a fe) seconds4 the screen image
should appear in the Paint )indo).
Directl3 under the )ord B@ileB in the Paint menu4 there is a dotted rectangle.
Ma/e sure this button is depressed. Enlarge the Paint )indo) so 3ou can see all
of the basin map. Ma/e a rectangle around the basin map b3 holding do)n the left
mouse button and dragging. .hen 3ou ha+e completed the rectangle4 let go of the
mouse buttonF 3our dotted rectangle around the basin map should remain.
1nder Edit in the Paint menu4 select Co$y !o. .e are no) going to sa+e the
basin map to a file.
!hange the Save a ty$e6 bar to 2" 2it+a$. 23pe in a file name and sa+e the file
some)here in 3our 86 account.
Open Microsoft .ord.
@rom the main menu in .ord4 select Inert/'icture/3ro+ 3ile. *etrie+e the basin
image 3ou Aust sa+ed in 'aint. 2he image should appear in 3our .ord )indo).
2. Editing a 2ain Model
5fter first ma/ing sure that the is depressed in the Sc,e+atic4 clic/ on the icon
)hich represents basin 12 )ith the left mouse button. 5fter this icon is highlighted4 hold
do)n the right mouse button and choose Edit from the pop&up menu that appears. In the
upper&right&hand corner of the Su((ain Editor )indo) that )ill appear4 3ou )ill find
the the area of sub&basin 12 -.'<2 sG. /m0. Ma/e a note of this +alue. Go ahead and loo/
up the areas of the other sub&basins as )ell. %M# can )or/ )ith a Basin file in either #I
or English units. In this case4 the file is in #I units.
To be turned in: a table showing the areas of the subbasins. What is the total drainage
area of Waller Creek (sq. km)
*eturning to the %M#4 focus 3our attention on the Sc,e+atic )indo). 2his )indo)
contains a +ariet3 of buttons )hich 3ou ma3 use to assemble and +ie) a basin model.
@irst4 consider the top ro) of buttons. 2he button allo)s 3ou to select -highlight0
items in the model. ?ou ma3 also use the button to drag&and&drop h3drologic
elements from the left&hand&side of the )indo) or to mo+e indi+idual elements )ithin
the model. If 3ou ha+e a model element highlighted4 hold do)n the right mouse button
and select Edit from the pop&up menu to +ie) its properties. 2he button allo)s 3ou
to mo+e -or pan0 the entire model displa3 b3 holding do)n the left mouse button and
mo+ing the mouse. If 3ou )ish to >oom in or out4 3ou ma3 do so b3 depressing the
or button respecti+el3 and selecting a rectangular area in the model to >oom in to or
out from b3 holding do)n the left mouse button and dragging the mouse to dra) a
rectangle. Go ahead and experiment )ith these buttons to understand better ho) each
)or/s.
2he second set of icons under the menu bar allo)s 3ou to choose ho) the h3drologic
elements are represented in the model. @rom left to right4 3ou choices are small
s3mbol 4 large s3mbol 4 standard icon -same icons as those used on the left&
hand&side of the )indo)04 or text name . 2a/e a minute and select each of these
buttonsF notice ho) each button affects the model image differentl3.
:ext4 the third set of icons allo)s 3ou to turn on or off the displa3 of the )atershed
boundaries 4 the ri+er reaches 4 or flo)&direction arro)s on the reaches . 5s
before4 ta/e some time to in+estigate for 3ourself ho) each of these buttons functions.
:otice that if 3ou turn off both and )hile is turned on4 3ou are left )ith a
collection of icons from the elements list. In most cases4 this is ho) 3ou )ould normall3
displa3 models. @or this model4 ho)e+er4 the )atershed boundaries and the ri+er reaches
)ere imported earlier as an image using a techniGue )e )ill encounter in more detail
later. Heep in mind that this image has no influence on the model6s output. In the final
hori>ontal group of buttons4 allo)s 3ou to +ie) a chart summari>ing the results from
a run4 )hile commences computations of a model. @or the time being4 do not use
these buttonsF )e6ll come bac/ to them later.
5t this point4 clic/ on so that the elements in the model are displa3ed as the3 are
along the left side of the )indo). 5s needed4 3ou ma3 drag&and&drop elements into the
model area. Go ahead and tr3 this4 but ma/e sure 3ou /eep the elements 3ou add a)a3
from the elements currentl3 in the model. 2r3 ta/ing a ri+er reach and connecting a
Aunction to each end. 2o do this4 grab one and t)o 6s. %oo/ up one Aunction to
one end of the reach b3 highlighting the Aunction -use 0 and holding do)n the right
mouse button. .hen the pop&up menu appears4 select Connect )o5ntrea+. *epeat
these steps appropriatel3 to connect the stream to the other Aunction. 2o delete the
elements 3ou ha+e added )hen 3ou are finished4 highlight all added elements -hold do)n
the shift /e3 to /eep adding elements0 and then choose )elete Ele+ent from the Edit
menu in the Sc,e+atic )indo). If 3ou )ant4 ta/e some time no) to experiment )ith
some of the other h3drologic elements.

5s the calculation steps abo+e suggest4 the primar3 function of a surface&)ater model
such as the %M# in+ol+es three sets of calculations&&Guantif3ing rainfall losses into the
soil4 con+erting excess rainfall to runoff4 and routing. 5s part of creating a %M# model4
the selection of the processes to be used for each calculation set is made in the 2ain
Model #ttri(ute )indo).
2o access this )indo)4 select 2ain Model #ttri(ute from the 3ile menu in the
Sc,e+atic )indo). 2he follo)ing )indo) should appear7
In this )indo)4 ta/e a loo/ at the )efault Met,od section. 2he method on the left4 9o
*ate4 is )here 3ou choose the process )hich calculates the rainfall losses absorbed b3
the ground. !lic/ on in the 9o *ate section to see 3our choices. #ome options are
SCS Curve 4o. and -reen : #+$t. In this model4 Initial/Contant has been selected.
2his loss relationship means that a Guantit3 of rainfall )ill be absorbed b3 permeable soil
initiall34 and a constant rate )ill be absorbed o+er the time frame of the model.
2he middle section4 !ranfor+4 allo)s 3ou to specif3 ho) to con+ert excess rainfall to
direct runoff. 5gain4 clic/ the to +ie) 3our options4 most of )hich 3ou should
recogni>e. 2his model emplo3s the SCS techniGue. 2he mod!lar/ model referred to
abo+e ta/es gridded rainfall data4 subtracts the losses as specified through the 9o
*ate4 and con+erts the excess rainfall to a runoff h3drograph using a +ariation of )hat is
/no)n as the !lar/ unit h3drograph. ?ou ma3 go else)here to learn more about ho) the
+odClar1 method )or/s. In addition4 #eann *eed4 a graduate student )or/ing )ith Dr.
Maidment4 performed some of the pioneering )or/ on con+erting gridded rainfall to
runoff that the %E! used )hen )riting the %M#. ?ou ma3 more read about this in an on&
line cop3 of his Master6s *eport.

2he final part of the )efault Met,od considers *outing4 )hich is )here 3ou specif3
the process for routing a h3drograph through a ri+er reach. Once again4 clic/ the and
loo/ o+er the choices. 2he Mu1ingu+ is specified here4 )hich is the routing techniGue
used for the reaches in this model.
5t this point4 close the 2ain Model #ttri(ute )indo) b3 clic/ing on Cancel at the
bottom of the )indo). :o) that )e ha+e considered the +arious loss4 transform4 and
routing techniGues to use4 )e )ill loo/ at the actual parameters associated )ith the
chosen techniGues. !hoose )ata-Ed. on the Sc,e+atic )indo) menu bar. ?our choices
should be
#elect 9o *ate and then Initial/Contant first. 2he follo)ing )indo) should appear.
:ote that )hile the Initial/Contant 9o *ate method )as specified +ia another menu4
it is in this table that the method parameters are actuall3 entered. Each basin reGuires an
initial lo Guantit34 a contant lo rate4 and a $ercent i+$ervioune. 2hese +alues
ha+e been selected arbitraril3. If the I imper+ious +alue differs from <4 that I of the
land area is assumed to ha+e no losses and the loss method is applied onl3 to the
remainder of the drainage area. #tudents in the Dept of Geograph3 of the 1ni+ersit3 of
2exas at 5ustin4 ha+e studied ho) the gro)th of the 1ni+ersit3 has changed the land use
patterns o+er time in .aller !ree/. #ee7
http788))).utexas.edu8depts8grg8ustudent8gcraft8)aller8)aller.html
5fter 3ou ha+e loo/ed o+er the data4 clic/ .K. @or future reference4 )hile both .K and
#$$ly transfer the data to the computer6s short&term memor34 the3 differ onl3 in that .K
closes the )indo) )hile #$$ly /eeps the )indo) open.
Once 3ou ha+e closed the Initial/Contant 9o )indo)4 choose )ata-Ed.4
!ranfor+4 and then SCS to +ie) this image7
:ote that the #!# unit h3drograph method reGuires onl3 one parameter for each sub&
basin&&lag time bet)een rainfall and runoff in the subbasin. %M#. 5t this point4 go ahead
and close this )indo).
2he next step in+ol+es entering the parameters for the routing process. @rom )ata-Ed.4
select *eac, *outing and then Mu1ingu+. 2his should cause the follo)ing image to
appear. 2his simulation routes the )ater through the reaches b3 the Mus/ingum method
in )hich H is the tra+el time of a flood )a+e passing through the reach4 J is a measure of
the degree of storage -J K < means a le+el&pool reser+oir or maximum storage4 J K <.$
means a pure transmission reach in )hich there are no storage effects4 and J ranges
bet)een < and <.$0. 2he reach is di+ided into a number of subreaches if necessar3 to
/eep the computations numericall3 stable.

?ou do not ha+e to go through the )ata-Ed. menu to edit element properties. Editing
ma3 also be accomplished b3 highlighting an element4 holding do)n the right mouse
button4 and choosing Edit from the pop&up menu. Before )e mo+e ahead to the
precipitation data4 close the 9ag )indo) and open Su((ain #rea/2ae 3lo5 from
)ata-Ed.4 )hich should loo/ li/e this7
In this part of the %M#4 3ou suppl3 if necessar34 information such as (ain area and
initial (ae flo5 in the s3stem. 2he receion ratio and t,re,old ;/ratio to $ea1
+alues are used b3 the %M# to add baseflo) to direct runoff. In this model )e are not
allo)ing for an3 base flo). .hen 3ou are finishing )ith this )indo)4 close it. In
addition4 3ou should also close the Sc,e+atic )indo) b3 selecting Cloe from the 3ile
menu.
To be turned in: Choose one of the subwatersheds and one of the ri!er rea"hes and
des"ribe in words what the parameter !alues are that are used to "hara"teri#e these
features h$drologi"all$.
3. Creating a 'reci$itation Model
%a+ing established the ph3sical aspects of the model4 )e )ill no) address the rainfall
data. @rom a statistical stud3 of extreme storm rainfall data recorded at gages4 tables and
maps ha+e been prepared for the )hole 1# )hich specif3 the storm precipitation depth to
be expected as a function of the return period of the e+ent and the duration of the
rainfall. 5 table of such +alues is sho)n belo) for 2ra+is !ount34 in )hich the !it3 of
5ustin is located. @or example4 for a 1< 3ear return period e+ent4 )e expect that in $ min
<.L$B of rainfall )ill fall4 in 1$ min 1.''B4 and so on4 up to E2< min -12 hours0 and 1""<
min -2" hours0. 2hese precipitation depths are the +alues )hich )ould be eGualled or
exceeded on a+erage once in 1< 3ears )hen considering a +er3 long period of data. 2he
+alues sho)n for 2ra+is !ount3 )ere extracted from a li+e map ser+er accessible on the
!*.* home page http788))).cr)r.utexas.edu8texas8rainfall8. 2he +alues gi+en in this
ser+er are precipitation intensities in mm8hr4 so the3 )ere con+erted to depths in inches
b3 multipl3ing b3 the rainfall duration and con+erting the resulting depth from mm to
inches. 5s rainfall duration increases4 the cumulati+e depth of rainfall increases but the
a+erage intensit3 o+er the duration decreases because se+ere rainfall cannot be sustained
for +er3 long.
5t present4 %M# can handle rainfall input onl3 in English units -inches0. 2o create a
design precipitation input file4 go to the ProAect .indo) and select Edit8Precipitation
Model8:e). 5 )indo) li/e this appears in )hich 3ou can describe the precipitation file
3ou intend to create4 in this case for a 1< 3ear storm.
5fter clic/ing4 OH4 the follo)ing )indo) appears4 in )hich 3ou should select the
3re<uency-2aed Hy$ot,etical Stor+4 as sho)n. 2he other options refer to different
methods of inputting the precipitation data4 including Grid&based Precipitation )hich
comes from :exrad radar.
.hen 3ou clic/ on O/4 the follo)ing table appears. @ill in the +alues sho)n from the
table abo+e for a 1< 3ear storm -1<I chance of being eGualled or exceeded in an3 3ear0.
!lic/ 5ppl34 and then O/ to complete this step.
@or each proAect4 the %M# then creates an output Data #torage #3stem D## file )hich
stores calculated data from all runs for a gi+en proAect so that results from a pre+ious run
can be directl3 compared to results from a more current run. 2he %M# stores data in the
D## file according to the information gi+en in the 'at,na+e section. 36 is left blan/
because the %M# )ill place the name of the run there.
To be turned in: a s"reen "apture of the design pre"ipitation input file

4. )efining t,e Control S$ecification
2he final piece of the model setup in+ol+es establishing the model6s time limits. Go bac/
to the 'ro0ect )efinition and select Edit/Control S$ecification/4e5. 2he follo)ing
)indo) )ill appear in )hich 3ou can label the specifications.
5fter 3ou clic/ O/4 3ou6ll get the )indo) to specif3 the duration of the simulation in date
and time4 and also the time inter+al of the calculations -1< minutes0. In this case4 the
duration is arbitrar34 long enough to depict the runoff from a 1&da3 storm4 but the 1<
minute time inter+al is part of the Basin file model setup and should remain fixed for this
.aller !ree/ model.

To be turned in: how man$ time inter!als of "omputation will be performed
=. E"ecuting an HMS Model
@inall34 3ou ha+e finished perusing the data in+ol+ed in creating the 2en/iller model. 2o
run the model4 go the HMS Sc,e+atic )indo) and select *un Configuration from the
Si+ulate menu. 2he follo)ing )indo) should appear. !lic/ on each of the three boxes
to highlight and select them4 hit 5dd to record these as the files to be used for *un &4 and
!lose the )indo). 2he three subsections in this )indo) list the ph3sical4 rainfall4 and
timing data sets that ha+e been created for this proAect. 2hough this model onl3 has one
data set in each section4 the %M# is slic/ in that it allo)s the user to ha+e multiple data
sets a+ailable to include con+enientl3 in different runs.
.hen finished4 select #dd4 close the *un Configuration )indo)4 go bac/ to the
'ro0ect )efinition4 and choose Si+ulation Manager from the Si+ulate menu. ?ou
should see an image similar to this7
2he Si+ulation Manager summari>es )hich 2ain I)4 'reci$ I)4 and Control I) ha+e
been chosen for each run. In this )indo)4 *un & is composed of the three sets 3ou Aust
chose in the *un Configuration )indo). @or the moment of truth4 select Co+$uteF if
all goes )ell4 3ou should see
If 3ou do not see this )indo)4 problems ha+e arisen. If so4 %M# gi+es a list of problem
locations and b3 editing the Basin file 3ou ma3 be able to eliminate them.
>. ?ie5ing HMS *eult
2he %M# allo)s 3ou to +ie) results in tabular or graphical form. 2o +ie) a global results
table4 go to the 'ro0ect )efinition and double&clic/ WallerC1.& . @rom the )indo)
)hich appears -the Sc,e+atic )indo)04 go to ?ie5 and choose . ?ou6ll get a
)indo) li/e that sho)n belo) )hich summari>es the pea/ discharge and time4 the total
+olume of storm runoff and the drainage area from )hich it came.
In addition to +ie)ing global results4 3ou ma3 also +ie) results for each element )ithin
the model. 2o do this4 choose in the Sc,e+atic )indo)4 select an element in the
model -!olorado *i+er for example0 )ith the left mouse button4 hold do)n the right
mouse button4 and choose ?ie5 *eult from the pop&up menu )hich appears. Pic/ from
-ra$,4 Su++ary !a(le4 and !i+e-erie !a(le. If 3ou +ie) the -ra$, for !olorado
*i+er4 3ou should see the follo)ing7
2he light red h3drograph is the inflo) data from the small area of .aller !ree/
immediatel3 upsterm of the !olorado *i+er )hich is added to the routed flo) in the
channel to produce the total outflo) cur+e. If 3ou clic/ on a )atershed4 3ou see the
rainfall at the top and the runoff at the bottom7
To be turned in: Make a table showing the peak dis"harge in "ms at ea"h of the si% outlet
points on Waller Creek (& 'un"tions and ( sink).. What is the drainage area abo!e ea"h
of these points What is the peak dis"harge per unit of drainage area ("ms)sq. km) for
these points
2e Creative@@@
2he citi>ens of 5ustin appro+ed a bond issue during 199L )hich authori>es the !it3 to
borro) approximatel3 M2$ million for the construction of a flood di+ersion tunnel for
.aller !ree/. 5t present the 1<< 3ear flood plain for the cree/ co+ers man3 cit3 bloc/s
in do)nto)n 5ustin. 2he goal is to ma/e this area safer from floods and also to
encourage the de+elopment of a ri+er )al/ area4 perhaps similar to that in #an 5ntonio.
5ssume that 3ou are the h3drologist )hose Aob it is to determine the design discharge at
the inlet point of this tunnel )hich is Aust on the do)nstream side of 1$th #treet. 5t this
point an o+erflo) )eir )ill di+ert a significant part of the flood discharge and con+e3 it
to 2o)n ;a/e in a 2< ft diameter tunnel.
#tud3 the data 3ou are pro+ided and if possible4 impro+e the estimates of I imper+ious
co+er on the )atershed4 and the initial and continuing losses. 2he design is to be based
on the 1<< 3ear design precipitation4 not the 1< 3ear storm that 3ou6+e used up to this
point.
To be turned in: a short a""ount of $our design dis"harge stud$. What input data !alues
did $ou use What design dis"harge at (&th St did $ou obtain Show a graph of the
dis"harge h$drograph at (&th St. How "an $ou "he"k the result to see if it is reasonable
Cleaning U$
.elcome to the end of )hat I hope ga+e 3ou more of an insight of ho) topics )e )ill
learn in class. 5t this point4 all 3ou need to do is ma/e sure 3ou ha+e completed the se+en
turn in assignments for this proAect. Go ahead and close the %M# b3 selecting E"it from
the 3ile menu in the 'ro0ect )efinition. .hen the %M# as/s 3ou if 3ou )ant to sa+e the
proAect4 choose 3orget C,ange.
Go to Dr Maidment6s %ome Page