sequently
1 summed
particular
- all other
and the
esent the
:idence of
at in the
eptable Z
se object
m. Other
; included
ccessfully
ng to be
dentifiers
ve object
justment.
nts were
d by the
resent the
s of the
although
3ht of the
tic errors
1ce to the
tes of the
readsheet
etric task
mates of
J
Ss and
precision are an important initial indicator of data quality and
can be derived from the least squares process. The precision
estimates of the targeted floating marker points were in the
range of 32mm - +5mm in elevation which is sufficient for the
purposes of development of the 3D flow model A more
important measure of quality is accuracy and this is often far
more difficult to quantify. The best estimates of accuracy are
determined by comparing derived estimates of coordinates to
the values of known check points. Such a check was not
possible to instigate because independent check points could
not be placed upon the dynamic water surface. However
examination of the plan positions of the marker points is
revealing (Figure 2). What can be identified is the alignment of
the lines which clearly join the floating points in a regular and
systematic pattern which is consistent with the directions of
flow. As a further check it is intended to compare these 2D
flow vectors with those computed by the 3D computerised flow
model. This will help to confirm both the accuracy of the water
surface morphology and possibly the 3D flow model itself.
2.6 DEM creation
The final coordinates were loaded into the Intergraph
Siteworks terrain modelling package for visualisation and
further processing. The 3D points were triangulated to form a
surface which could be contoured and used to create an
isometric grid representation (Figure 3).
3D water
Figure 3, Tsometric view. of
surface model
3. Integration of DEM into flow model
The next stage of this research will involve the use of water
surface data, in combination with digital elevation models of
the river-bed, to increase our understanding of flow processes
in confluent channels. Such understanding is critical because of
the existence of confluences as key nodes in fluvial systems, as
well as the parallel between confluences and points of
discharge of polluting substances into rivers. Existing research
into confluence dynamics (e.g. Biron et al, 1993) has
illustrated the importance of three-dimensional flow structure
as a control on the mixing process. This flow structure is
thought to vary with the precise morphology of each
confluence, and the discharge ratio and the turbulent intensity
of the confluent flows. Field and laboratory investigations
allow the understanding of specific combinations of flow
structure controls, but they take time and cost money to
instigate. One alternative is the use of numerical simulation,
and although such methods have been used effectively for two-
dimensional problems (e.g. Lane et al, 1994b; Lane et al.,
1995), the nature of confluence flows requires a three-
dimensional treatment (Lane, in press). If Computational Fluid
Dynamics code can be used to simulate effectively three-
dimensional flow structure in field and laboratory measured
confluences, then this can be extended to the simulation of
confluence flow processes with different controlling conditions.
The water surface data derived from this series of field work
will be used for three purposes:
e in combination with bed morphological information to
determine both water depth and bedslope, and hence to
determine hydraulic slope, so providing a first estimate of
X
SSL b 7 PS Dd
ERHEBEN.
PD SE SS ADS
eap Pe KIBET ISR
(rupe HE
dnd 9. PAPQqPQU DET DI ICH Pod Scoto RSS
GAS PPS te Sede pede EPA SEES ISO.
A V 4h c] 9. {CP qp <> 4 b t qp <> <> <> TR ABS CP d « ^p CF D » wt IS CaS
AIR SE I KE SE RRR
ORBI o SSR ERE EDS
EN IE NASSEN
DR reed o SO Saee Ho SOS LESE
| GNT O SEC N DIA
E220 RSS A A eS oS SISK
ARRETE SELS SS ee oo SSI
SSL Se Oe Se So
SSH HU
LISE 2 O2 SOS ESS SL ISA HEIR A RSIS TY
KNIE ZEILE WX ood oC o Pe X, DC D D d oq X, RI XQ |
A Nene reo eo aee See eed P IEEE ESEIRALSSEE ISIE )
FTN Deco aeo SL SSL
ENS AU Coe ooo Deo o eo er Se o Oe e OO
(8990295992025 4,5 HSID e
LI Seo o eoe eee Ld o o oe oe e o Oe S S
loeo eroe oe Oe eee e eoa A IRIEL eO
(11385222040 ZEITEN
sig Re So Soo
HEERES QOO OO OO
S
©
Q
[
E
Frs
+
EX
HO
RQ
XO
RS
[XO
OQ
Um E ©
Em
0
C=
103
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B7. Vienna 1996
