Elsevier

Building and Environment

Volume 161, 15 August 2019, 106230
Building and Environment

Sensitivity of Australian roof drainage structures to design rainfall variability and climatic change

https://doi.org/10.1016/j.buildenv.2019.106230Get rights and content

Highlights

  • Hydraulic sensitivities linked to existing and future hydrological uncertainties.

  • Sensitivity to rainfall intensity increases varied from 13% to 406%.

  • Increases in rainfall intensity from spatial variability uncertainty from 2 to 54%.

  • Rainfall intensity projections for the 2090 decade ranged from −15 to +59%.

  • At most sites, exiting uncertainty is greater than future uncertainty for 2 °C warming.

Abstract

The main design determinant for small catchment hydraulic structures is rainfall intensity. Localised relationships between rainfall intensity, frequency and duration (IFD) are used to design each structure to a specified level of performance. However, limitations in rainfall observations introduce uncertainty in IFD values. Further, this uncertainty is compounded by changes in climatic conditions via anthropogenic forcing. Whether this is a cause for concern depends on the structure's sensitivity to deviations away from design rainfall values. Here, we investigate the ability of roof drainage systems to accommodate deviations in design rainfall. We assess the sensitivity of box gutter overflow designs across Australia to spatial variability in IFD values, and projections of IFD values due to climatic change. Different overflow designs were found to have markedly variable responses to rainfall intensity increases, from 13% to 406% before failure. Potential increases in rainfall intensity from spatial variability uncertainty varied from 2 to 54%. While rainfall projections for the 2090 decade ranged from −15 to +59%. Rainfall intensity increases as high as 259% were noted when both sources of uncertainty were combined for a temperature rise scenario of 5 °C. At the majority of locations coupled increases in rainfall intensity were primarily driven by existing rather than future uncertainties for a 2 °C temperature rise scenario. Considering design rainfall uncertainties in terms of design sensitivity shows that adapting to present and future uncertainties can come at no additional cost for some design options while other options need to be altered to reduce the risk of failure.

Introduction

A key function of the built environment is to protect people and property from adverse meteorological conditions. In many design processes, a minimum level of protection from specific weather-related hazards is mandated by design guidelines, building codes and standards. While operational performance can be determined to a relatively high degree of confidence, by empirical tests [1,2] or known physical relationships [3], achieving a similar degree of confidence in design rainfall values for low-probability meteorological extremes is more difficult [[4], [5], [6], [7]]. This is acknowledged in scientific literature [[8], [9], [10]], although the degree to which it would influence design varies between structural elements of the built environment [11,12].

In the design of hydraulic structures, risk is assessed on three factors; the probability of extreme rainfall events occurring, the likelihood of performance failure from such events, and the cost of failure should it occur [13,14]. The probability of extreme rainfall events is characterised using a relationship between Intensity-Frequency-Duration (IFD), Intensity-Duration-Frequency (IDF) or Depth-Duration-Frequency (DDF) [[18], [19], [20]]. Designs for large structures, or networks, are likely to give a greater consideration to parameter uncertainty than small plot-level structures because of the higher associated costs of failure. However, the aggregated costs of multiple plot-level structures simultaneously failing can be substantial, such as when extreme events occur in urban areas [[15], [16], [17]].

Site-specific IFD rainfall values used in the design process are model estimates and are subject to large uncertainty [21]. IFD estimates may not accurately represent actual rainfall values at a site due to gaps in available rainfall data [22,23], the statistical methods used [6,8,10,24] and the climate baseline over which the data has been collected [4,5]. As superior statistical methods are developed, and the quantity and quality of the observations in datasets increase, improved IFD estimates can be produced. For example, the 2016 revision of Australian IFD relationships resulted in changes in intensity values of between +19.7% and −32.6% compared to 1987 IFD values, at point locations in capital cities [25]. IFD estimates can also change with the timespan of available rainfall data because of short to medium term periodic metrological drivers [26,27], and longer term anthropogenic forcing of the climate system [[28], [29], [30], [31]]. These elements combine to create an environment of non-stationary IFD design rainfalls, under which hydraulic designers attempt to maintain a specified performance level over a hydraulic structure's lifetime [[18], [19], [20]].

While methods exist to estimate [5,6,32,33], or account for [[34], [35], [36]], uncertainties in IFD values, the data and technical knowledge required to apply them limits their use in the design of plot-level hydraulic structures. Pappenberger and Beven [11] noted in a review that the technical challenges associated with assessing uncertainty in IFD estimates is a common justification for ignoring the uncertainty in the hydraulic design process. An additional challenge is modelling the influence of climate change on IFD estimates over the structure's operational life. Although the influence of climate change on IFD estimates is difficult to quantify with confidence, the literature indicates that the intensity of rainfall in short-duration events is increasing as more moisture is stored in a warmer atmosphere [37]. The more moisture the atmosphere can hold, the greater is the extreme rainfall event, which forms a basis for updating pre-existing IFD relationships for future temperature increases.

Roof drainage systems are common small catchment hydraulic structures designed using an IFD design rainfall value for a single point location. The objective of this study is to determine the ability of current Australian design standards for roof drainage systems with box gutters to accommodate deviations in design rainfall values. Possible deviations from IFD estimates are assessed based on two sources of variability: spatial variation in proximate IFD values, and changes in climate conditions from anthropogenic forcing. Box gutters located within a buildings perimeter, were chosen as the model structure because exceedance of hydraulic capacity will result in water ingress into the building envelope and performance can be determined satisfactorily using existing hydraulic theory. Although, Australian design guidelines have been used in this study, there are many similarities to design guidelines employed in the USA and Europe.

Section snippets

Box gutter design

The rational (or equilibrium) method [38,39] is a commonly used approach to estimate the peak flow rate in small to medium catchments. Peak flow rate is often the only hydrologic metric considered in the design process of many hydraulic structures, particularly in small urban catchments [[40], [41], [42]]. Its applicability to flood analysis has been questioned for landscape-level catchments with heterogenous surface conditions and greater variation in spatial and temporal patterns of rainfall [

Location of study sites

Nine areas around Australia were included in this study. The study areas were centred on the central business districts (CBD) of Brisbane, Sydney, Canberra, Melbourne, Hobart, Adelaide, Perth, Darwin and Cairns (Fig. 1). The locations include all the state capitals as well as Cairns (which receives some of the most extreme rainfall intensity values in Australia). These locations were chosen as they have a high density of small catchment hydraulic structures and comparatively long-term and

Box gutter sensitivity

The sensitivity of box gutter designs to an increase in flow rate, ΔQMAX, varied substantially, from 1.125 to 4.063, between the overflow types (Table 3). Each overflow design had a unique range of sensitivity with no overlap of ΔQMAX values. The least sensitive design option was the rainhead, which could accommodate increases in flow rates of more than threefold, for a 300 mm wide gutter, to fourfold, for a 600 mm wide gutter. Values of ΔQF for the rainhead are also dependent on width and

Conclusion

Evaluation of potential existing and future deviations in design rainfall intensity in relation to the hydraulic sensitivities of designs can aid the designer in determining if adaption measures are necessary. In this study, deviations in design rainfall values were found to be primarily driven by existing spatial variability rather than future uncertainties at most locations. Therefore, adaptation measures are pertinent not only to possible future climates but are valuable to improve the

Acknowledgements

The authors would like to acknowledge and thank the Australian Government Bureau of Meteorology for the rainfall and temperature datasets used in this research. The data set can be obtained from http://www.bom.gov.au/climate/data/stations/. L. Verstraten is grateful for funding support from the Association of Hydraulic Services Consultants Australia. C. Wasko is grateful for funding support provided by the University of Melbourne Mckenzie Postdoctoral Fellowship.

References (110)

  • F.M. Woldemeskel et al.

    Constraining continuous rainfall simulations for derived design flood estimation

    J. Hydrol.

    (2016)
  • E. Salvadore et al.

    Hydrological modelling of urbanized catchments: a review and future directions

    J. Hydrol.

    (2015)
  • I. Emmanuel et al.

    Influence of rainfall spatial variability on rainfall–runoff modelling: benefit of a simulation approach?

    J. Hydrol.

    (2015)
  • S. Ochoa-Rodriguez et al.

    Impact of spatial and temporal resolution of rainfall inputs on urban hydrodynamic modelling outputs: a multi-catchment investigation

    J. Hydrol.

    (2015)
  • I. Chaubey et al.

    Uncertainty in the model parameters due to spatial variability of rainfall

    J. Hydrol.

    (1999)
  • M.M. Hasan et al.

    Correcting bias in radar Z – R relationships due to uncertainty in point rain gauge networks

    J. Hydrol.

    (2014)
  • L. Pedersen et al.

    Quantification of the spatial variability of rainfall based on a dense network of rain gauges

    Atmos. Res.

    (2010)
  • L. Uusitalo et al.

    An overview of methods to evaluate uncertainty of deterministic models in decision support

    Environ. Model. Softw

    (2015)
  • L. Guan

    Preparation of future weather data to study the impact of climate change on buildings

    Build. Environ.

    (2009)
  • V. Agilan et al.

    What are the best covariates for developing non-stationary rainfall Intensity-Duration-Frequency relationship?

    Adv. Water Resour.

    (2017)
  • S. Westra et al.

    A conditional disaggregation algorithm for generating fine time-scale rainfall data in a warmer climate

    J. Hydrol.

    (2013)
  • C. Wasko et al.

    Continuous rainfall generation for a warmer climate using observed temperature sensitivities

    J. Hydrol.

    (2017)
  • K.H. Beij

    Flow in roof gutters

    Bur. Stand. J. Res.

    (1934)
  • R.W.P. May

    The design of conventional and siphonic roof-drainage systems

    J. Water Environ.

    (1997)
  • D.C. Verdon-Kidd et al.

    Regime shifts in annual maximum rainfall across Australia – implications for intensity–frequency–duration (IFD) relationships

    Hydrol. Earth Syst. Sci.

    (2015)
  • V. Klemeš

    Tall tales about tails of hydrological distributions. II

    J. Hydrol. Eng.

    (2000)
  • V. Klemeš

    Tall tales about tails of hydrological distributions. I

    J. Hydrol. Eng.

    (2000)
  • F. Pappenberger et al.

    Ignorance is bliss: or seven reasons not to use uncertainty analysis

    Water Resour. Res.

    (2006)
  • M.H. Middelmann

    Natural Hazards in Australia: Identifying Risk Analysis Requirements

    (2007)
  • S.-J. Wu et al.

    Risk analysis for flood-control structure under consideration of uncertainties in design flood

    Nat. Hazards

    (2011)
  • M.H. Spekkers et al.

    A statistical analysis of insurance damage claims related to rainfall extremes

    Hydrol. Earth Syst. Sci.

    (2013)
  • M.H. Spekkers et al.

    On the occurrence of rainstorm damage based on home insurance and weather data

    Nat. Hazards Earth Syst. Sci. Discuss.

    (2014)
  • British Standards

    BS EN 12056-3:2000 Gravity Drainage Systems inside Buildings - Part 3: Roof Drainage. Layout and Calculation

    (2000)
  • International Code Council

    International Plumbing Code

    (2015)
  • Standards Australia/Standards New Zealand. AS/NZS 3500.3:2015 Plumbing and Drainage - Stormwater Drainage

    (2018)
  • J. Green et al.

    New design rainfalls for Australia – lessons learned…

  • G.M. Bonnin et al.

    Updates to NOAA precipitation frequency atlases

  • E. Hajani et al.

    Design rainfall estimation: comparison between GEV and LP3 distributions and at-site and regional estimates

    Nat. Hazards

    (2018)
  • Bureau of Meteorology

    Rainfall IFD data system

  • K.E. Kunkel et al.

    Monitoring and understanding trends in extreme storms: state of knowledge

    Bull. Am. Meteorol. Soc.

    (2013)
  • A.S. Taschetto et al.

    An analysis of late twentieth century trends in Australian rainfall

    Int. J. Climatol.

    (2009)
  • C. Wasko et al.

    Reduced spatial extent of extreme storms at higher temperatures

    Geophys. Res. Lett.

    (2016)
  • G. Lenderink et al.

    Increase in hourly precipitation extremes beyond expectations from temperature changes

    Nat. Geosci.

    (2008)
  • K.E. Trenberth

    Changes in precipitation with climate change

    Clim. Res.

    (2011)
  • T.F. Stocker et al.

    IPCC, 2013: Climate Change 2013: the Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change

    (2013)
  • E.A. Lehmann et al.

    Spatial modelling framework for the characterisation of rainfall extremes at different durations and under climate change

    Environmetrics

    (2016)
  • I.B. Gregersen et al.

    Decision strategies for handling the uncertainty of future extreme rainfall under the influence of climate change

    Water Sci. Technol.

    (2012)
  • C. Wasko et al.

    A quasi-global assessment of changes in remotely sensed rainfall extremes with temperature

    Geophys. Res. Lett.

    (2016)
  • T.J. Mulvaney

    On the use of self-registering rain and flood gauges in making observations of the relations of rainfall and flood discharges in a given catchment

    Proc. Inst. Civ. Eng.

    (1851)
  • E. Kuichling

    The relation between the rainfall and the discharge of sewers in populous districts

    Trans. Am. Soc. Civ. Eng.

    (1889)
  • Cited by (8)

    • Automating rainfall recording: Ensuring homogeneity when instruments change

      2022, Journal of Hydrology
      Citation Excerpt :

      The temporal pattern of rainfall effects infiltration, erosivity, vegetation interception (Van Dijk et al., 2005; Vernimmen et al., 2007), as well as risk-based infrastructure design (O’Shea et al., 2021), particularly in urban settings, small catchments, hillslopes, and second order drainage networks (Ball, 1994; Grimaldi et al., 2021; Heneker et al., 2001; Verstraten et al., 2019).

    • A case study on the effects of data temporal resolution on the simulation of water flux extremes using a process-based model at the grassland field scale

      2021, Agricultural Water Management
      Citation Excerpt :

      Such changes would alter the characteristics of the water fluxes generated, as the soil properties of the field will change, meaning the determination of an appropriate resolution to simulate water fluxes may also change from the hourly resolution suggested here. This is analogous to other hydrological studies where, for example, different overflow designs in roof drainage structures have markedly variable responses to rainfall intensity increases (Verstraten et al., 2019). Key model input variables such as precipitation can determine the impacts of simulation time-steps on the performance of hydrological models; for example, the duration and temporal variability of a precipitation event in relation to the rainfall–runoff lag time (Ficchì et al., 2016).

    • Review: Can temperature be used to inform changes to flood extremes with global warming?

      2021, Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences
    • Estimating design hydrologic extremes in a warming climate: Alternatives, uncertainties and the way forward

      2021, Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences
    View all citing articles on Scopus
    View full text