doi 10 1016j landusepol 2009 08 019 L H a b a A R A K W F W H m T m V o a o c r r m l U d a G 0 d Land Use Policy 26S (2009) S251–S264 Contents lists available at ScienceDirect Land Use Policy journa.
Trang 1Contents lists available atScienceDirect Land Use Policy
j o u r n a l h o m e p a g e :w w w e l s e v i e r c o m / l o c a t e / l a n d u s e p o l
a Department of Civil and Environmental Engineering, Imperial College, London SW7 2BU, United Kingdom
b Engineering Science, Oxford University, United Kingdom
a r t i c l e i n f o
Article history:
Received 24 August 2009
Accepted 24 August 2009
Keywords:
Water resources
Flood risk
Water quality
a b s t r a c t
Human activities have profoundly changed the land on which we live In particular, land use and land management change affect the hydrology that determines flood hazard, water resources (for human and environmental needs) and the transport and dilution of pollutants It is increasingly recognised that the management of land and water are inextricably linked (e.g Defra, 2004) “Historical context, state
of the science and current management issues” section of this paper addresses the science underlying those linkages, for both rural and urban areas In “Historical context, state of the science and current management issues” section we discuss future drivers for change and their management implications Detailed analyses are available for flood risk, from the Foresight Future Flooding project (Evans et al., 2004a,b) and other recent studies, and so we use flooding as an exemplar, with a more limited treatment
of water resource and water quality aspects Finally in “Science needs and developments” section we discuss science needs and likely progress This paper does not address the important topic of water demand except for some reference to the Environment Agency’s Water Resources Strategy for England and Wales (Environment Agency, 2009)
© 2009 Queen’s Printer and Controller of HMSO Published by Elsevier Ltd All rights reserved
Historical context, state of the science and current
management issues
The urban environment
Urban development provides a useful illustration of some of the
most obvious effects of land use change on water management
Vegetated soils are replaced with impermeable surfaces, increasing
overland flow and reducing infiltration, bypassing the natural
stor-age and attenuation of the subsurface In addition, the conveyance
of runoff to streams is modified Overland runoff is conventionally
collected by piped storm-water drainage systems and conveyed
rapidly to the nearest stream The result is a greater volume of
runoff, discharging in a shorter time, potentially leading to
dra-matically increased flood peaks, but also reduced low flows and
less groundwater recharge
Urbanisation effects on fluvial floods
The size of the effect of urban development on streamflow will
depend on the natural response of the catchment The effects will be
夽 While the Government Office for Science commissioned this review, the views
are those of the author(s), are independent of Government, and do not constitute
Government policy.
∗ Corresponding author.
E-mail address: h.wheater@imperial.ac.uk (H Wheater).
greatest where natural runoff is low, in catchments with permeable soils and geology, and can include changes in flood seasonality Natural catchments in the UK mainly flood after prolonged rainfall
in winter, when soils are already wet and storm runoff is readily generated Urban catchments are not so seriously affected by these antecedent conditions and respond more rapidly to rainfall This means that intense summer rainfall may become a major cause of flooding (Institute of Hydrology, 1999)
It is expected that the relative effects of urbanisation will reduce
in larger, rarer floods, but current design guidance to quantify this
is highly speculative
For larger catchments, the effects are more complex, as the loca-tion of development within the catchment will affect its response For example, urban development located near to the outlet of a catchment may generate runoff before the main response of the natural catchment arrives The overall effect of urbanisation on the catchment flood peak will depend on the relative magnitude and timing of the constituent responses
These effects have been well known for some 40 years (see e.g Hall, 1984), and to mitigate them, engineered solutions have rou-tinely been adopted to reduce flood peaks through the provision
of storage One common solution is the construction of a reser-voir to provide “detention storage.”Crooks et al (2000)report on the effects of 30 years of urbanisation on two sub-catchments of the Thames, showing an apparent increase in flood frequency with urbanisation, followed by a reduction as storage solutions were implemented
0264-8377/$ – see front matter © 2009 Queen’s Printer and Controller of HMSO Published by Elsevier Ltd All rights reserved.
Trang 2There is much interest in Sustainable Urban Drainage Systems
(SUDS) to manage urban runoff and associated problems of water
quality Various design solutions can be implemented, for example
restoring the infiltration of rainfall into the soil by directing storm
runoff to engineered soakaways, or seeking to retard flows within
the storm sewer system (Verworn, 2002) However, a lack of clear
responsibilities for design and maintenance have limited uptake of
SUDS in England and Wales The official review of the UK’s 2007
summer floods (Pitt, 2008) highlights the current problems of
gov-ernance of water in the urban environment Pitt also comments
on the increasing density of urbanisation He proposes solutions
such as planning controls on paved areas within areas of domestic
housing
While well-developed design guidelines are available for
con-ventional storage, based on a substantial body of research (Hall
et al., 1993), the research base to support SUDS applications is
much more limited There is no clear understanding of the effects
of extreme rainfall on the performance of SUDS, and there is
sub-stantial anecdotal evidence that control of local-scale installations
is ineffective, leading to errors in construction and defective
oper-ation (Packman, pers comm.)
Urban stormwater flooding
“Urbanisation effects on fluvial floods” section above addressed
the effects of urban development on river flooding There are also
major issues of flooding due to surface runoff within the urban
envi-ronment This type of flooding is a major cause of insurance claims
for flood damage Storm runoff is normally channelled via gully
pots, into storm sewers, which are usually designed to
accommo-date relatively frequent events Under more extreme conditions,
these sewers will start to surcharge (flow full under pressure), and
as pressures build up, manhole covers can lift and the sewers
dis-charge to the surface Such flows combine with surface runoff to
generate flooding of roads and properties Urban flooding is often
complex Sewer flooding can arise when pipes exceed their
capac-ity, become blocked, have their capacity limited by river flooding,
or a combination of these factors Divided management
responsi-bilities are a problem in this area One of the recommendations of
the Pitt report (2008) is for clear overall responsibility for urban
flooding in England and Wales
There are technical problems in urban flood design The
fre-quency of surface flooding for storm sewers is not a design criterion,
is often not known, and will vary greatly for different systems
There has been a lack of technical capability to address this
prob-lem But in the past few years, models have been developed to
represent the surface routing of overland flows, and associated
storm sewer interactions, supported by high resolution
topo-graphic data, for example from LIDAR airborne remote sensing
systems (Djordjevi ´c et al., 2004) This offers exciting potential for a
paradigm shift in the design of the urban environment to manage
flood risk
Floodplain development
Finally in this discussion of urban flooding, we turn to issues of
development on floodplains Many major towns and cities are
adja-cent to rivers, and there are continuing economic pressures to build
in river floodplains However, floodplains have precisely the
func-tion that their name suggests; rivers can be expected naturally to
flow beyond their banks every few years The natural functioning of
a floodplain is to store and subsequently release flood waters,
atten-uating a flood as it travels downstream Over the past century or
more, floodplains have been increasingly used for urban and
agri-cultural development, and the need to protect that development
has led to engineered disconnection of the river from its floodplain
The result is a loss of flood attenuation, and increases in flood risk
downstream This remains an issue of concern for the major Euro-pean rivers such as the Rhine Levels of flood protection for some German cities have significantly decreased and active efforts have been made in recent years to recreate floodplain storage The same issues arise in the UK, although little work is available to quantify the effects of historic changes There is now interest in the UK in the potential for the return of floodplain land to an active water storage role, for example by reducing the level of flood protection
of agricultural floodplain land (see below)
Recent moves have been made by the UK Government to strengthen the role of the Environment Agency in the planning pro-cess in England and Wales (CLG, 2006), and also to raise awareness
of planners of the risks of flooding A particular problem, high-lighted by the 2007 floods, is the location of strategically important utility infrastructure in floodplains It is also not uncommon for emergency services, hospitals and residential homes for the elderly
to be located in floodplains
Water resource and water quality tissues Towns and cities need water supplies, which are often imported from other catchment areas After use, this water is conventionally routed through the sewer system, treated, and discharged to the local river Urbanisation reduces natural water infiltration into soil,
so that in urban rivers, effluent discharge may be a dominant com-ponent of river flows, particularly under the low flow conditions of summer
The release of treated effluents to streams has long been a major source of pollution, and nutrients have been a particular concern
EU legislation, in the form of the Urban Wastewater Treatment Directive, has required major treatment works to introduce ter-tiary treatment to reduce nutrient loads, but this requirement does not extend to the large numbers of small treatment facilities.Jarvie
et al (2007)report observations of phosphorus in the river Lam-bourn in Berkshire These measurements show the effect of sewage effluent on phosphorus loads in the river, the reduction in phospho-rus when treatment was improved, and the subsequent release of phosphorus from river sediments as the system re-equilibriated
In addition to the discharge of treated effluents, there is poten-tial for pollution from urban storm runoff, which can include oils and heavy metals Urban storm drainage systems normally include simple devices, such as gully pots, to collect sediments and asso-ciated pollutants, while one of the roles of SUDS, discussed above,
is to reduce pollutant discharge Particular problems arise where storm and foul sewers are combined Under extreme flows, treat-ment facilities are unable to accept the storm discharges, and overflows of sewage effluent to watercourses can occur This is a concern for pollution of the Thames in London, and is one of the motivations for major investment in a new interceptor sewer There is also scope in urban areas for a wide range of pollutants
to be released to the water environment from accidents, spillages, broken pipes and illegal activities In recent years, industrial pollu-tion of surface water systems in the UK has been greatly reduced in response to tighter regulatory controls But in the subsurface, there
is a legacy of pollution of soils and groundwater, with long-term consequences Groundwater in urban environments is commonly polluted and is not suitable as a potable resource
The management of water in the urban environment can signif-icantly modify hydrological impacts The harvesting of rainwater from roofs can reduce both storm runoff and the demand for other water resources, while the re-use of so-called ‘grey water’ at a domestic scale is technically feasible, although not currently eco-nomic (Liu et al., 2007) Vegetation can be used to attenuate and reduce runoff and associated pollution, either at the scale of ‘Green Roofs’ or in larger scale implementation of SUDS In water-limited areas, the management of urban water has been intensified, and
Trang 3associated investment has been made to reduce pollutant
load-ing In Singapore, for example, runoff is collected as a resource
from approximately 50 per cent of urban areas, and wastewater
is treated to create ‘NewWater’, mainly for industry, but in part for
domestic supply
Changes in water management can have important effects In
London, a large lowering of groundwater levels occurred due to
industrial water use after the industrial revolution In recent years,
that use has reduced, and rising groundwater levels have posed
major problems for subsurface infrastructure, including basements
and the underground rail system Pumping is a solution, but due to
historical pollution, the water that it produces is not suitable for
water supply
The mismanagement of water can also have significant impacts
Leaking water distribution systems provide a source of water to the
subsurface, and leaking sewer systems are a source of pollution A
striking example is in Riyadh, Saudi Arabia, where the leakage of
imported water has created major problems of rising groundwater,
and perennial flows in the once ephemeral Wadi Hanifah
The rural environment
Afforestation
While urbanisation is a dramatic change to the natural
environ-ment, the effects of other land use changes are more subtle The first
long-term experimental hydrological research programme in the
UK, based initially on the Plynlimon catchments in Wales (Hudson
and Gilman, 1993), was initiated some 40 years ago in response
to concerns for the effects of afforestation on water resources
This research showed that high rates of evaporation of
intercep-tion storage (water wetting the surface of leaves) had important
effects on the water balance This finding was consistent with
a large body of international literature which shows that in the
long term, afforestation reduces flows due to increased
evapora-tion (Bosch and Hewlett, 1982) However, there has been some
ambiguity about the effects of lowland broad leaf forests, with
con-trasting views put forward (Calder, 2007; Roberts and Rosier, 2005)
Although long-term effects are relatively well defined, in the short
and medium term, effects on flows may be very different Studies
byRobinson (1986)show that the drainage practices widely used
at that time to establish forests in the UK uplands gave rise to an
increase in storm runoff, an effect that may last for many years
Field drainage
In the 1970s, attention in the UK turned to the effects of
agricul-tural drainage on flooding (Robinson and Rycroft, 1999; Robinson,
1990) Under-draining, the use of underground pipe systems to
drain soils to improve production, is a common agricultural
prac-tice and the UK is one of the most extensively under-drained
countries in Europe Much of this drainage occurred between the
1940s and the 1980s, encouraged by government grants (Robinson
and Armstrong, 1988) In the UK, very low-permeability soils often
have a secondary treatment such as ‘subsoiling’ or ‘moling’, to
improve the flow of water to the drains Although the cessation of
grants in 1984 has meant that there has been little new land being
drained, existing drainage is still maintained to varying degrees
(Armstrong and Harris, 1996)
The installation of field drains will generally cause a reduction
in surface and near-surface runoff due to a lowering of the water
table and an increase in the available storage capacity of the soil
However, runoff from drained land may be faster or slower than
from undrained land depending on the nature of the soil and its
management (Armstrong and Harris, 1996), as well as the
tim-ing and intensity of rainfall.Reid and Parkinson (1984)illustrated
how runoff response from drained fields varied seasonally,
depend-ing on antecedent moisture conditions A reduction in the time to
peak flow and an increase in the magnitude of peak flows has been reported in relation to installation of field drains in a 16 km2clay catchment in North East England (Robinson et al., 1985)
The drainage of soils rich in organic matter may have both short and long-term effects Lowering the water table in peatlands will increase the amount of available storage capacity in the short term but will also increase organic matter decomposition rates, resulting in a subsequent decrease in available storage as the organic matter content decreases (Holden et al., 2004) and hence potentially an increase in flood peaks in the long term, as well as long-term soil damage
Agricultural intensification The floods that have affected England and Wales since 2000 have reinforced more general concerns that changing agricultural prac-tices in the UK may have increased the risk of flooding (Wheater, 2006) This is not an issue solely confined to the UK and similar con-cerns have been raised elsewhere across northern Europe (Evrard
et al., 2007; Pinter et al., 2006; Bronstert et al., 2002; Pfister et al., 2004; Savenije HHG, 1995) Prior to World War II, the UK agri-cultural landscape was characterised by small fields with dense hedgerows and natural meandering rivers The subsequent drive for increased productivity in farming brought about major changes (O’Connell et al., 2007) These include the loss of hedgerows and
an increase in field size, the installation of land drains connecting hilltop to river channel, and channelised rivers with no riparian zone This landscape change has been accompanied by changing patterns of agricultural land use and the intensification of pro-duction, although recent changes in agricultural policy have led
to some de-intensification in the past few years
Changes in arable production have been associated with changes in cropping and land cultivation practice and the increas-ing use of heavy machinery There have been pressures to work land when soil moisture conditions are unsuitable, and to work land unsuitable for purpose
In the uplands, the source areas for the UK’s major rivers, land use is dominated by grassland production, mainly for sheep In Wales, 72 per cent of agricultural land was estimated to be under grassland production in 2005, almost exclusively to support sheep farming Sheep numbers in Great Britain doubled between 1950 and 1990 as a result of farm support payments based on stock num-bers (Fuller and Gough, 1999) Associated with this change has been
an increase in the amount of improved pasture in upland areas, which has been created by draining, ploughing, and reseeding, also financially supported by government and EU incentives (James and Alexander, 1998) These increased numbers also led to the use of less suitable land for grazing, supporting higher stock intensities
on marginal land
The degradation of soil structure, due to either arable or grazing intensification, can lead to reduction in soil infiltration rates and available storage capacities, increasing rapid runoff in the form of overland flow (e.g.Heathwaite et al., 1990; Bronstert et al., 2002; Carroll et al., 2004; O’Connell et al., 2004) There are concerns in the
UK and elsewhere in Northern Europe that this may increase the risk of flooding (Holman et al., 2003; Stevens et al., 2002; Boardman
et al., 1994; Burt, 2001) However, the role of land use management
in enhancing or ameliorating UK flood risk has been identified as
an unanswered question in a major review commissioned by Defra (O’Connell et al., 2004) The research cited above mainly focuses on the role of land management intensification at the scale of individ-ual fields The catchment-scale effects remain largely unresolved Beven et al (2008)attempted to identify the catchment-scale effects of land use and land management change by interrogation
of catchment-scale data, but failed to identify a clear relationship between land use and land management and river flows This does
Trang 4not mean that such a relationship does not exist, but rather that
errors in catchment-scale measurements and the multi-faceted
nature of catchment change, combined with climate variability,
do not allow such effects to be detected, although they may be
substantial
A recent multi-scale experimental and modelling study has been
established at Pontbren, in mid-Wales, to provide data and
mod-els to address this issue (Marshall et al., 2009; Jackson et al., 2008;
Wheater et al., 2008a,b) The soils at Pontbren mainly comprise
heavy clay, with a history of land drainage, and are predominantly
grazed by sheep Pontbren’s land management background is also
noteworthy From the 1970s to the 1990s, sheep numbers increased
by a factor of six, and animal weights doubled Since that time,
farmers have reduced stocking densities, moved to smaller and
hardier breeds, and started reinstating hedgerows and shelter belts
Experimental studies have shown rapid improvement in soil
struc-ture and permeability associated with the establishment of tree
shelter belts, and modelling studies have been used to investigate
both field and small catchment-scale effects While runoff volumes
are not significantly changed by the planting of shelter belts,
impor-tant changes to runoff peaks are indicated Simulations suggest
that for frequent events, the median effect of reverting to 1990s
patterns of land use would be to increase flood peaks by 13 per
cent Conversely, introducing optimally placed tree shelter belts
to the current land use would reduce peak flow by 29 per cent,
and introducing full woodland cover would reduce flows by 50
per cent Considering an extreme event, the corresponding median
effects are a 5 per cent increase and a 5 per cent and 36 per cent
reduction, respectively While some of these effects are not large,
neither are they insignificant It is worth noting that such
inter-vention measures also have benefits for diffuse pollution and for
wildlife habitats, but there is currently no framework for integrated
assessment of these possible benefits
The above discussion has related to the generation of runoff
from fields and hillslopes, i.e the amount and timing of water
enter-ing rivers The routenter-ing of flows down rivers is affected by floodplain
management As we have seen in “Floodplain development”
sec-tion, there has been a disconnection of rivers from their floodplains,
itself associated with the provision of increased flood protection
for agricultural land But it is possible to reduce flood risk
down-stream by reducing the level of protection for agricultural land in
floodplains, and allowing the natural storage and attenuation of
water associated with floodplain inundation to be re-established
However, issues are not straightforward; agricultural areas with
flood defences can act as washlands in high flows, storing water
above its natural level and reducing the peaks of flood hydrographs
Hence removing the surrounding flood protection banks entirely
may in some cases perversely increase flood risk to downstream
areas In addition, there are social and economic issues associated
with their removal, for example with compensation payments paid
to landowners rather than tenants
Water quality, sediments, geomorphology and habitats
The flow regime of a catchment, in combination with its sources
of sediment supply, determines the geomorphological behaviour of
a river This means that land management practices have
implica-tions for suspended sediments, river geometry and bedform This
has implications for habitats and infrastructure
Defra (2008)states that up to 75 per cent of sediment loading to
rivers can be attributed to agriculture.Collins et al (1997)analyse
sediments on the Upper Severn and note the signature of
acceler-ated erosion of soils due to afforestation and deforestation and the
erosion of pasture soils.Hatfield et al (2008)analysed
sedimen-tation rates in NW England Pulses of erosion and sediment flux
were associated with mining and deforestation in the 19th century,
agricultural intensification in the mid-20th century, and a recent possible signal of climate change Henshaw and Thorne (2008) report significant increases in sediment bed-load when compar-ing improved and unimproved pasture at Pontbren Major stream incision has taken place in areas of improved pasture Field ditches, gullies and over-steepened banks provide sources of coarse sedi-ment These authors also identify a recent reduction in bed-load yields, possibly associated with measures taken by the farmers to protect banks from cattle and reinstate hedgerows and woodland shelter belts
Land management is also strongly associated with chemical water quality The major pollutants of concern for the UK are nutri-ents, particularly nitrate and phosphates.Defra (2008)estimates that 60 per cent of nitrate pollution and 25 per cent of phosphates
in English waters originates from agriculture In response to the
EU Nitrates Directive (1991), controls are placed on agriculture through the designation of Nitrate Vulnerable Zones (NVZs), which have recently been revised for England, increasing the area designated from 50 per cent to 70 per cent This means that in most of England and Wales, limits are imposed on the permissible loading of organic and inorganic nitrogen applications, and on their timing, and no application is allowed on areas defined as being at high risk of runoff There are important tensions between agriculture and water quality, and these raise associated policy issues.Lovett et al (2006)demonstrate that nutrient standards in the River Slea could only be met by taking substantial areas of land out of agricultural production
An overarching issue is where the responsibility for nitrate pol-lution should lie.Lovett et al (2006) also provide an economic analysis that indicates that the cost of nitrate reduction in agricul-tural production is four times that of nitrate removal by treatment
of the potable water supply Similar issues apply to phosphorus
In a modelling study of the potential impacts of phosphorus man-agement on the rivers of eastern England,Wheater and Daldorph (2003)conclude that while constraints on agricultural manage-ment can reduce river concentrations significantly, it would be hugely expensive to achieve the low levels of phosphorus concen-tration associated with good ecological status
Relatively little attention has been paid to biological water qual-ity, although faecal indicator fluxes are of concern.Hunter et al (1999)show the effects of sheep grazing on faecal bacteria lev-els in an upland environment in England, andOliver et al (2005) show significant E coli loads in runoff from land grazed by cattle and treated with livestock wastes.Crowther et al (2002) demon-strate the effects of lowland pastoral agriculture on coliform, E coli and Streptococci concentrations on river waters and hence on marine bathing water quality ButKay et al (2005), in an analysis
of the 1600 km2Ribble catchment in NW England, concluded that urban areas were the dominant source of faecal indicators for that catchment
Many other pollutants are relevant to the rural environment, but space precludes their detailed treatment here Some are of local concern (e.g pollution from sheep dips), others reflect specific inci-dents (e.g disposal of carcasses in the Foot and Mouth epidemic) and some are widespread and may reflect long-range pollution (e.g acid deposition, which is dependent on land cover)
A principal driver for water resources management is the EU Water Framework Directive This focuses on ecological quality and is a driver for the protection of water quality and of aquatic habitats It involves a timetable for the achievement of environ-mental targets However, particular issues arise with respect to groundwater and groundwater-dominated rivers, such as the chalk streams of South East England Chalk groundwater is recharged
by water travelling through an unsaturated zone that can be up
to 100 m deep Recent research has addressed uncertainties in
Trang 5Fig 1 The flooding system Upper image: the fluvial/coastal flooding system Lower image: the intra-urban flooding system (fromEvans et al., 2004a,b ).
the transport of nitrate in this unsaturated zone, confirming that
rates of movement are of the order of 1 m per year This means
that a legacy of decades of nitrate history is moving slowly to
groundwater—the so-called ‘nutrient time-bomb.’ (Jackson et al.,
2007)
Land use futures: a water management perspective
The relationship between land use and flooding was explored in the 2004 Foresight Future Flooding project (Evans et al., 2004a,b), which was recently revised as part of the Pitt Review of 2007 floods
Trang 6Fig 2 Combined climate change and socio-economic futures used in the Foresight
2004 analysis.
(Pitt, 2008; Evans et al., 2008) We first summarise key issues from
these reviews, and then discuss the water resource and water
qual-ity perspectives
Future magnitude, distribution and drivers of flood risk in
England and Wales
The aim of the 2004 Foresight Future Flooding project was to
use the best available science to provide a challenging vision for
flood and coastal defence in the UK between 2030 and 2100 and
so inform long-term policy It employed two forms of analysis—a
quantitative, probabilistic, computer analysis using very large
Geo-graphical Information System (GIS) databases based on the Risk
Assessment for System Planning (RASP) system developed by the
Environment Agency, and a qualitative analysis The latter used a
structured method to draw out evidence-based expert knowledge
to estimate approximately how big an impact the various drivers
and responses might have on flood risk under different future
sce-narios, and then ranked them in order of impact on flood risk
The project saw the flooding system as being composed of two
sub-systems, the catchment and coastal flooding system, and the
‘intra-urban’ system, where flooding arises from events within
urban areas This is in contrast to river and coastal flooding, where
water enters urban areas from outside (Fig 1)
The analysis used four combinations of climate and
socio-economic scenarios to create alternative pictures of possible
futures (Fig 2), drawing on UKCIP02 and the Foresight Futures
socio-economic scenarios (SPRU et al (1999); OST 2002)
Magnitude and distribution of potential future flood risk
The quantitative analysis was first run under a ‘business as usual’
assumption for flood risk management, with government
contin-uing to spend the same amount of money and following the same policies as in 2004 The results of this were striking, with flood risk increasing in the 2080s under all four scenarios, as illustrated in Tables 1 and 2
The distribution of flood risk was shown by maps such as those
inFig 3, which compared the present and future distributions of economic damage
The concentration of flood risk around the coast and in the major urban areas is obvious, as is the lesser severity of future flood risk under the Global Sustainability scenario compared with the high-growth, high climate change, low-regulation World Markets scenario
It should of course be borne in mind that UK government expen-diture on flood risk management has increased considerably since the publication of the Future Flooding report in 2004, thereby reducing the growth of future risk under the ‘business as usual’ sce-nario Nevertheless the flood risk multipliers and the distributions
of future flood risk shown inFig 3remain of interest in showing the potential for flood risk to increase in the future
Drivers of future flood risk What then were the drivers of these large increases in future flood risk? Here we draw on the Pitt Review update of the orig-inal qualitative analysis, which grouped the drivers according to
a Source/Pathway/Receptor (SPR) classification as shown below in Table 3:
The top 12 drivers, graded by national flood risk multiplier in the 2080s, are shown inTable 4:
It can be seen that while climate change drivers feature highly, many drivers connected to land use are also prominent including infrastructure, buildings and contents, urbanisation and intra-urban runoff In addition, the flood risk created by climate change is dependent not only on the increased frequency of flooding, but also
on the distribution and number of receptor assets in the floodplain,
a function of the degree of regulation under the different scenarios
inTable 5
The Future Flooding analyses went on to show that with port-folios of structural and non-structural responses, implemented in
a sustainable way, the future risks could be pulled back to a level around that of the present day The top 12 responses are shown below
It can be seen that responses related to land use rank alongside engineering responses as the most powerful in controlling future flood risk
Land use and the drivers of future flood risk
We now examine more closely some of the flood risk drivers and responses in the context of land use
Table 1
Flood risks expressed as Expected Annual Damage (EAD) and the baseline costs of flood defence for the business as usual option—catchment and coastal, 2080s.
Present day World markets National enterprise Local stewardship Global sustainability
Table 2
Flood risks expressed as Expected Annual Damage (EAD) and the baseline costs of flood defence for the business as usual option—intra-urban, 2080s.
Present day World markets National enterprise Local stewardship Global sustainability
Trang 7Fig 3 Foresight futures: comparative risk – Expected Annual Damage – residential and commercial properties, 2080s.
Trang 8Table 3
Combined list of fluvial/coastal and intra-urban drivers.
Climate change
Fluvial systems and processes
River morphology and sediment supply P
Urban systems and processes
Sewer conveyance blockage and sedimentation P Impact of external flooding on intra-urban drainage systems P
Socio-economics (now includes rural and intra-urban
receptors and all types of flooding: river, coastal, pluvial
and coincident)
Climate change drivers
The high ranking of coastal drivers draws attention to their
importance and to the choices which must be made between
pro-viding high levels of funding to resist rising coastal threats,
realign-ing defences, or abandonrealign-ing large tracts of land to the sea The
connection between land use in coastal areas and future flood risk is
obvious
Precipitation is a pervasive driver for non-coastal flood risk Although future precipitation is highly uncertain, increased fre-quency of extreme events is expected This raises concerns for consequential impacts including intra-urban, fluvial and ground-water flooding Increased flood risk for urban and rural areas will impact on lives, infrastructure, agricultural production and ecosystems, while increased floodplain flows have implications
Table 4
National ranking of drivers, graded by national flood risk multiplier—2080s.
Trang 9Table 5
Response rankings for the 2080s.
for land use management The possibility of having to find the
space through our riverside towns and cities to accommodate
flood flows up to 40 per cent greater than today’s values presents
great challenges not only in engineering terms but particularly to
urban planning It contrasts awkwardly with Government policy
of reusing brownfield sites Many of these originated as waterside
developments during the industrial revolution, using water as a
source of power and transport Increased flooding will also impact
on the role of agricultural land management in flood mitigation, as
well as affecting agricultural land and productivity
Managed realignment is often seen as a solution to coastal flood
and erosion defence problems However, the cost effectiveness of
this measure may be less than was believed (Rupp-Armstrong,
2008) and it may not be as widely adopted as originally
envis-aged Two factors have a bearing here The first is the increasing
cost of managed realignment schemes The second is the rising
value of agricultural land in the UK and the greater awareness
of food security as an issue due to climate change and changing
world markets (Brown and Funk, 2008; IAASTD, 2008) Although
less than 1 per cent of flood damage affects the agricultural sector
(Evans et al., 2004a,b), a large proportion of the most
agricultur-ally productive land in England and Wales is dependent on flood
protection and land drainage All the scenarios in our 2004 report
reveal high exposure to flood risk in the Fens of East Anglia The
increased importance now being placed on future food security
may require response options to be re-evaluated to reduce flood
risk and to maintain standards of land drainage in areas of national
agricultural importance
Urbanisation
Urbanisation acts as a driver of flood risk by increasing runoff
which affects communities downstream, and by increasing the
assets at risk of flooding
The effects of urbanisation on runoff are well known With-out mitigation, urbanisation increases flood risk The key issue
is the extent to which mitigation measures are implemented, either at catchment or local scale (see below) So the effect
of this driver is heavily dependent on socio-economic scenar-ios
Urban areas are also impacted by floods, a process exacerbated
by population growth, household distribution and human attitudes and desires Development on floodplains is of great concern It puts property and infrastructure at risk of flooding, and affects the trans-mission of floodplain flows The Pitt Review (Pitt, 2008) shows vividly that a number of recently constructed housing estates were flooded in 2007 Decisions on where to build houses, factories and other infrastructure are now recognised as a key tool in managing future flood risks The importance of protecting vital infrastructure from flooding is also clear
However, this issue is not a simple one The 2004 Foresight flooding reports (Evans et al., 2004a,b) drew attention to the need to balance flood management against other economic, social and envi-ronmental needs, especially the demand for new housing It would
be controversial to ban redevelopment of brownfield sites that lie in the floodplain, but are behind well-managed flood defences afford-ing a high standard of protection This applies to much of London The need is perhaps for more sharply targeted policy instruments Future urban flood risk will be affected by changes in the way
in which urban areas are managed, their characteristics, and how planning and management change in the context of social and cli-mate change Important effects here may include the renewal of existing urban spaces, new urban forms, new densities of develop-ment, more green space, and encroachment into green belts While changes in existing urban form are certain to occur, the fabric of urban areas changes relatively slowly in the UK For example, the current rate of replacement of the housing stock is 0.1 per cent per
Trang 10annum and the rate of addition to that stock is 1 per cent In
addi-tion, 22 per cent of all land in England is already in some urban
usage and there is limited scope for further urban expansion This
limitation is compounded under some scenarios used in the
Fore-sight Future Flooding project which suggest that the UK could be
short of agricultural land for food production
A significant percentage of insurance claims for flood damage
originate from outside of the floodplain They arise from the
pres-ence of groundwater, local flooding in the form of ‘muddy floods’
(runoff from nearby hills and fields), and from intra-urban flooding
Urban drainage systems and processes
As we noted in “Historical context, state of the science and
cur-rent management issues” section, a number of drivers connected
with the urban drainage system also are of relevance to urban land
use via their interaction with the form and function of the urban
area, and are likely to become a more important factor in limiting
flood risk in the future
Building development, operation and form include
opportuni-ties to manage local flood risk though actions taken at the building
level Examples include the use of permeable surfacing in car
parks and rainwater harvesting Responses from the various
stake-holders are also included (i.e individual behaviour) together with
responses that relate to actions when flooding does occur
(miti-gation) However, even where there is control over urbanisation,
‘creep’ adds hard surfaces in an uncontrolled and unpredictable
manner
Source controls comprise a range of possibilities Classical
solutions to increased flood risk include construction of storage
reservoirs to attenuate flows More recent methods focus on SUDS,
although lack of clear responsibilities has limited their uptake in
England and Wales The 2008 Pitt Review also highlights increasing
density of urbanisation—proposing controls on paved areas within
domestic housing, for example
Rural land management
As was the case for urban water management, rural land has
a role both as a driver of flood risk, and a receptor The role of
rural land management in flood runoff generation has been
dis-cussed above Although there has been some de-intensification of
agriculture in recent years, this followed dramatic intensification
of UK agriculture over the previous 30 years, in response to
agri-cultural policy and economic and social pressures It is thought
that this intensification has increased runoff generation at the local
and small catchment scales, and a significant proportion of UK
soils are classed as degraded Clearly the opportunity exists for
land management to mitigate flood risk, both in the context of
runoff generation, and by the potential use of agricultural
flood-plain land for flood storage and attenuation (Morris et al., 2005)
Such intervention measures also have benefits for diffuse
pollu-tion and for wildlife habitats, as we noted above, but there is
currently no framework for the integrated assessment of these
benefits
There is also a growing perception that changes in peri-urban
land use and land management may have a significant impact
on flood risk Agricultural intensification, together with
addi-tional urbanisation of the peri-urban area, has produced significant
changes in the volumes of runoff that enter the urban area from the
peri-urban area, including the effects of reduced infiltration and
increased overland flow
Farm land is more tolerant of flooding than urban land, and the
unit costs of damage are much lower there While flooding and soil
waterlogging in some intensively farmed areas can result in
signif-icant losses of agricultural output, in others this is not the case
Changing polices towards agriculture and environment suggest there could be benefit in ‘setting back’ some previous agricultural flood defences to restore ‘natural’ floodplains in ways which pro-vide benefits in terms of flood storage and enhanced biodiversity Promoted by financial rewards to land managers, these measures could support rural livelihoods
With respect to the alternative futures explored in the Fore-sight Future Flooding project, the constituents of the driver set (Table 4) are mainly ranked as having a medium impact on future flood risk, although the impact of urban and rural land use is perceived to be particularly high for the utilitarian world market and national enterprise scenarios Agriculture is shown to exert a medium influence as a receptor under most scenarios The reason for this assessment varies between scenarios due to differences in land use, damage costs, and the degree of exposure to flooding For rural land use as a pathway and a receptor, flood risk is mainly a function of societal preferences evident in agricultural and environ-mental policy drivers Similarly, the contribution of urbanisation to flood risk is influenced by socio-economic factors which shape the nature and rate of urban development
Land use, water resources and water quality Future flooding in the UK has received considerable attention, with extensive scenario definition and analysis undertaken in two Foresight studies But other aspects of water management futures are less well developed
Water resources for the future For England and Wales, anEnvironment Agency, 2001report on Water Resources for the Future (Environment Agency, 2001) gave projections of water demand under a number of scenarios similar
to those used by Foresight Future Flooding This has recently been updated, with a 2009 report on the Water Resources Strategy for England and Wales (Environment Agency, 2009)
The problem of the provision of water is compounded by the fact that precipitation is biased towards the North and West of Britain whereas population and hence consumption are biased towards the South and East The current water resources status shows relatively large areas of the South East as either over-abstracted or over-licensed, and the scenario adopted for 2050s climate shows reductions in mean monthly Summer and Autumn river flows of up to 80 per cent Scenarios are used to support projections of future demand on the basis of variables includ-ing population growth They suggest changes by 2050 ranginclud-ing from a 35 per cent increase for ‘uncontrolled demand’ to a 15 per cent reduction for ‘sustainable behaviour’ One possible growth area is water use for irrigation At the moment this accounts for only 1–2 per cent of water use, although this demand is naturally at its peak at times and in areas of water shortage Use of water for irrigation could rise by 25 per cent by 2020, and given the large changes in growing conditions projected for the 2050s (seeFig 4), could be much higher by then However, water resource futures are highly uncertain The Environment Agency report (2009)points out that more than 60 per cent of water consumed in food and goods and services used is imported,
so the UK economy is particularly vulnerable to global water scarcity
Although the provision of extra water is a challenge, propos-als such as storing more water in mid-Wales and transferring it eastwards to the Thames basin and beyond are under active study (Thames Water, 2009) The problems of realising such transfers are several They include the current structure of the water industry
in England and Wales, funding, the environmental impacts of dif-ferent water quality levels, and the possibility of invasive species