PUB in practice: case studies - Runoff Prediction in Ungauged Basins: Synthesis across Processes, Places and Scales - page 332

distribution of discharge from one or several rain storms.
Thus, in data-scarce regions, reservoir areas in combin-
ation with regional reservoir area

with significant and expanding urbanisation, accurate par-
ameters and runoff characteristics for this type of built
environment are needed.

volume relations and
publicly available rain records can provide the catchment
discharges to validate hydrological models such as the
Thornthwaite

-

Mather model.
Especially in semi-arid Africa, irrigated agriculture from
reservoirs is seen as one of the major ways of increasing
food production. This will have a direct impact on stream
discharge ( Sachs and McArthur, 2005 ; UN Millennium
Project). An indirect method, as presented here, can not
only help to assess water resources and to monitor water
stress, but also to understand the runoff generation in the
study catchments, and allow impact assessment of small
reservoirs on the available water resources.

- Description of the study area

The study area is a 32-hectare mesquite grassland and a
13-hectare residential development (referred to as the
grassland and urban catchments, respectively) in the city
of Sierra Vista in south-eastern Arizona at 1300 m eleva-
tion (see Figure 11.65 ). Topographic relief is moderate,
with a 31 m elevation difference between the highest
point in the natural catchment and the outlet of the urban-
ised catchment. Construction within the urbanised catch-
ment was conducted from 2001 to 2005 and is typical of
most tract-style housing in the south-western USA. The
site was completely graded and building pads for houses
were compacted prior to construction. Houses 185 m 2 or
larger are built on relatively uniform lots 1670 m 2 or
larger and have similar building materials and landscaping
( Figure 11.65 ). Streets are asphalt, 7.3 m wide, with
rounded curbs. About 90% of roofs are sloped (25% to
35%) with corrugated cementitious tiles; the rest are low-
slope (2% to 8%) with elastomeric coating. The tile roofs
discharge runoff distributed along eaves, mostly without
gutters, while flat roofs discharge through focused down-
spouts. A 1-metre wide pervious right of way exists
between sidewalks and the street. Storm drainage is via
surface streets except for a 1.3-hectare area that drains to
the catchment outlet via a 61 cm corrugated metal pipe.
Vegetation is immature, with only small areas of canopy
cover. All pervious surfaces are covered with 2 to 4 cm
diameter gravel mulch, about 10 cm deep, except for a
few small irrigated turf areas. About 10% of yards have
pervious weed barrier fabric underlying the gravel mulch.
Stormwater runoff from the natural catchment is routed
through the urbanised catchment. Runoff from both
catchments is of short duration and baseflow is absent.
Vegetation on the natural catchment consists of 3 to 6 m
tall mesquite trees (Prosopis velutina), with relatively
abundant inter-canopy grass up to 1 m tall. Vegetation
transitions from mostly grass in the upper reaches to
mostly mesquite in the lower reaches, and is seasonally
dormant.
Stream stage was measured at 1-minute intervals by an
automated bubble gauge upstream of a 90-degree v-notch
weir at the channel connecting the two catchments and at
the outlet of the combined catchments ( Figure 11.65 ) from
May 2005 until September 2008. Rainfall data were col-
lected at 1-minute intervals at four weighing recording rain
gauges in 2005 and 2006 (gauges 401, 402, 403 and 404),
and at two additional rain gauges in 2007 and 2008
(gauges 420 and 424). From August 2006 onward, each

11.16 MODEL ENHANCEMENTS FOR
URBAN RUNOFF PREDICTIONS
IN THE SOUTH-WEST USA
j. r. kennedy, d. c. goodrich and
c. l. unkrich

The issue from societal and hydrological perspectives
Population growth and urbanisation have occurred rapidly
in the American south-west over the past several decades
and they are projected to exceed those of other regions of
the USA in the future. Urbanisation often leads to an
increase in storm runoff. Urban rainfall runoff models

and regionalisation approaches typically consider storm
runoff volume due to the increase in impervious surfaces,
but less commonly consider the effect of changes in the
infiltration rates of pervious surfaces in the built environ-
ment, or the effect of routing runoff from impermeable
rooftops across permeable soils. Woltemade (2010) found
a decrease in infiltrability in newer developments (post-
2000), where the use of heavy machinery for site devel-
opment was more common (as it is in the south-west
USA). Also, in recent years, the increase in stormwater
runoff associated with urbanisation has begun to be con-
sidered as a potentially renewable water resource. This
runoff can be re-used directly, through rainwater harvest-
ing efforts or recharge of aquifers through focused infil-
tration in detention basins or dry wells, or indirectly, by
routing runoff to natural stream drainages where it can
recharge. In arid environments, where upland surface
recharge is minimal, increased runoff from urbanisation
can lead to increased recharge, as rainfall that previously
would have infiltrated and then evaporated or transpired is
instead routed to an area where deep infiltration and
recharge can occur. If PUB is to be successful in basins

-