|
<br />Ilh = difference in water' surface elevation at
<br />the two sections, in feet; and
<br />L = length of the reach (Dalrymple and
<br />Benson, 1967), in feet.
<br />
<br />In Manning's equation, the quantity (L486/n)
<br />AR213 is called conveyance, K, and is computed for each
<br />cross section. The mean conveyance in the reach
<br />between any two sections is computed as the geometric
<br />mean of the conveyance of the two sections, The
<br />discharge equation in terms of conveyance is expressed
<br />as:
<br />
<br />1/2
<br />Q = (KjK2Sf) ,
<br />
<br />In this investigation, n is computed for each
<br />reach of known discharge, the water, surface profile,
<br />and the hydraulic properties of the reach as defined by
<br />the cross sections. The following equation was
<br />primarily used to compute n for this study and is
<br />applicable to a multisection reach of M cross sections,
<br />designated I, 2, 3,.., M - I, M:
<br />
<br />14861 ~ -[ (kl>h,J 1,2 + (Mh,l 2-3 +,.. (kl>h,l (M _ I) ,MJ
<br />n=TI L[" L"3 L(M_I),M ,(7)
<br />1 -+-+
<br />\ ZIZ2 Z,Z3 "'Z(M_j)ZM
<br />
<br />where
<br />
<br />~ = (h+h,l[-(h+h"l,
<br />Z = AR213 (Barnes, 1967),
<br />
<br />ASSUMPTIONS AND LIMITATIONS
<br />
<br />Although efforts were made to strictly follow the
<br />site-selection and data-collection criteria, assumptions
<br />were required for some of the verification
<br />measurements as channel and hydraulic conditions
<br />were not always ideal. The main sources of potential
<br />errors in calculations of roughness coefficients for
<br />dry land channels include changing boundary and
<br />vegetation conditions, discharge measurement
<br />uncertainties, and changing bedforms. The verification
<br />measurements presented in this report are qualified or
<br />rated on the basis of these four factors, Additionally,
<br />
<br />transfer of results presented in this report to similar
<br />sites may be limited because of uncertainties
<br />associated with extrapolating n values obtained for
<br />relatively low flows to flows of greater magnitudes.
<br />The accuracy of verification measurements
<br />presented in this report is rated as either good, fair, or
<br />estimated, These categories correspond to potential
<br />percent errors of less than 10, 15, or 20 percent,
<br />respectively, Verification measurements with an error
<br />of greater than 20 percent were not considered for
<br />publication in this report.
<br />
<br />Changing Boundary Conditions
<br />
<br />(6)
<br />
<br />In the performance of hydraulic computations of
<br />flow in streams that are dominated by sand,sized
<br />material, constant ,bed geometry often is assumed to
<br />persist throughout flow events. Several investigators
<br />indicate sand-dominated streams do not scour
<br />appreciably in a uniform segment of river channel
<br />(Culbertson and Dawdy, 1964; Benson and Dalrymple,
<br />1967), Although most of the channel segments studied
<br />for this report are uniform, scour of substrate material
<br />during the rise and peak of the flow and subsequent
<br />backfill during the flow recession requires
<br />consideration. Accuracy ratings for the n-verification
<br />measurements made in sand-dominated streams where
<br />changing boundary conditions are possible, thus, were
<br />downgraded according to the potential amount of scour
<br />and fill.
<br />Although the channel-geometry changes may
<br />not be as significant as with sand-dominated streams,
<br />flood, stage flows in gravel-bed streams also may
<br />mobilize and transport bed material. Hydraulic
<br />components measured for the verification measure,
<br />ments made in gravel,bed streams, however, probably
<br />were not large enough to cause a considerable change
<br />in boundary conditions resulting from substantial
<br />movement of the bed material.
<br />
<br />Changing Vegetation Conditions
<br />
<br />In arid and semiarid environments, vegetation
<br />commonly grows throughout the main channel of
<br />dryland streams, The vegetation can significantly
<br />impede flow and result in large increases in roughness
<br />coefficients, as suggested by several past studies
<br />(Aldridge and Garrett, 1973; Thomsen and
<br />Hjalmarson, 1991). The force and power of flows,
<br />
<br />Assumptions and Limitations 7
<br />
|