OR/17/026 Methodology

From MediaWiki
Jump to navigation Jump to search
Mansour, M M, and Hughes, A G. 2017. Summary of results for national scale recharge modelling under conditions of predicted climate change. British Geological Survey Internal Report, OR/17/026.

Introduction

The following section describes the recharge model (the code itself and its application to the British mainland) and its use with climate change scenarios. For this project climate change scenarios using the 11 member ensembles from the HadCM3 RCM created by FFGWL project (Prudhomme et al., 2012[1]) (rainfall and potential evaporation) have been run through the recharge model. The basis of climate change scenarios used are outlined below. To allow the impact of climate change on recharge to be assessed the modelled daily potential recharge values (mm/d) have been processed in a number of different ways for both groundwater bodies and for river basin management districts. Monthly recharge has been calculated along with seasonal (winter, spring, summer and autumn) totals for different time slices: 2020s, 2050s and 2080s. The methodologies to produce these results is then described in detail.

Model code and its application

Model code

ZOODRM (Mansour and Hughes, 2004[2]) is an Object Oriented model developed by BGS as part of the ZOOM suite of models. It is a distributed recharge model that simulates runoff and recharge processes and provides the output in a gridded form for use with groundwater flow models or on a catchment basis for water balance purposes. It has been applied in both in the UK (e.g. Mansour et al., 2011), to the GB landmass (Mansour et al., 2018[3]) and overseas (e.g. Hughes et al., 2008[4]).

Model instance — application to the GB mainland

The GB-wide recharge model was built using BGS’ code ZOODRM (Mansour and Hughes, 2004[2]; Hughes et al., 2008[4]). Recharge is calculated on a grid with 2 km square cells over the area described by the following National Grid Reference: Bottom Left (40000, -10000) Top right (680000, 1010000). The model has been run from 1st January 1962 to 31st December 2010 and calibrated against the runoff component of river gauged flow. It calculates recharge on a daily basis and aggregates the recharge to a monthly basis (Mansour et al., 2018[3]).

The calculation method used is the modified FAO (Hulme et al., 2001[5]) as proposed by Griffiths et al. (2006)[6]. It uses the distribution of soil parameters and crop parameters obtained from the HOST soil data map ,which includes 33 classes of soil types (Boorman et al., 1995[7]), and the land cover map, Land Cover Map 2000 (Fuller et al., 2002[8]), which includes 9 land use classes. The values of these parameters are obtained from the literature, e.g. Hulme et al. (2001)[5]. The full set of data used for the model are presented in Table 1.

Table 1    Data used for the GB-wide recharge model.
Data Source Reference
Rainfall CEH CERF/GEAR Keller et al., 2005[9]
CEH-GEAR Data set
Potential Evaporation (PE) MORECS PE Hough and Jones, 1997[10]
Landuse LCM2000 Fuller et al., 2002[8]
DEM CEH DTM Morris and Flavin, 1990[11]
River network CEH Moore et al., 1994[12]
Geology BGS Digmap
Soil map HOST Boorman et al., 1995[7]
Crop distribution LCM2000 NERC, 2000[13]

The model calculates potential recharge, which is the amount of water calculated to leave the bottom of the soil zone. It does not, therefore, take into account any modification of recharge resulting from the unsaturated zone and interaction with other, minor aquifers which may lie above the water table.

Climate change datasets — future flows climate

Funded by DEFRA and produced in 2009, UKCP09 provides projections of climate change in the UK (Prudhomme et al., 2012[1]; Murphy et al, 2007[14]; Jenkins et al., 2009[15]; Murphy et al, 2009[16]). The probabilistic climate projections provided by UKCP09 are not fully spatially coherent. To overcome this problem, 11 physically plausible simulations were generated under the medium emissions scenario also known as the A1B SRES emission scenario (IPCC, 2000[17]). Based on the 11 variants of the Hadley Centre Regional Climate Model HadRM3-PPE, which underpins the UKCP09 scenarios, the Centre of Ecology and Hydrology (CEH) applied a bias-correction and downscaling procedure to produce 11 scenarios of Future Flow Climate data. The 11 ensembles consist of an unperturbed example (afgcx) and ten perturbed simulations (Murphy et al., 2009[16]). These data are 1 km gridded climate time variant projections of rainfall (Prudhomme et al., 2012[1]) and potential evaporation (Prudhumme and Williamson, 2013[18]) and allow comparison of results across a range of scales and geographical regions. The data were produced as daily grids from 1st January 1950 to 30th November 2099. The 11 ensembles are named as follows:

  1. afgcx
  2. afixa
  3. afixc
  4. afixh
  5. afixi
  6. afixj
  7. afixk
  8. afixl
  9. afixm
  10. afixo
  11. afixq

The recharge model has been run with rainfall and potential evaporation for all 11 ensembles and the results processed as discussed in the following section.

The output from the 11 ensembles run through the recharge model can be seen as complimentary to Future Flow Hydrology (Prudhomme et al., 2013[19]) which produced an ensemble of daily river flows and monthly groundwater levels for Great Britain. The CERF model produced gridded outputs but didn’t explicitly examine recharge values and how they might vary under conditions of climate change. The monthly groundwater levels were produced from point models (24 overall). The recharge modelling presented here seeks to provide output at a range of scales from gridded 2 km data, groundwater bodies (310) and for the River Basin Management Districts (11).

Processing model output

Groundwater bodies

The following processing was undertaken to produce summary statistics for the groundwater bodies for England and Wales. There are 310 groundwater bodies and they are used for reporting requirements for the EU Water Framework Directive (WFD). Figure 1 shows their distribution.

The results for each ensemble have been analysed for each groundwater body as a whole and presented as colourised spatial plots for each groundwater body.

File:OR17026fig1.jpg
Figure 1    Distribution of groundwater bodies (Contains public sector information licensed under the Open Government Licence v3.0).

Average and percentiles

The mean, standard deviation and the following percentiles: 10, 25, 50, 75, 90 of annual recharge values for each groundwater body have been produced for the following periods: simulated historic (1950–2010), 2020s (2010–2039), 2050s (2040–2069) and 2080s (2070–2099) (see Appendix 1). The percentiles have been calculated from the ranking of the annual recharge value, so the annual recharge for the 10% value is that which has 90% of the values greater than this value. The use of annual recharge in this section is aligned with the calculation of average recharge used by the EA for each groundwater body for WFD reporting purposes.

Generally speaking the values for mean, SD and the percentiles increase in areas of higher recharge, i.e. western England (Cumbria and Cornwall) and Wales and are lower to the east of England. This follows the spatial distribution of rainfall and evaporation over England and Wales, with the highest rainfall/lowest evaporation in the west and lowest rainfall/highest evaporation in the east.

Exceedance of 75% and occurrence under 25%

The 25th percentile and 75th percentile of the total monthly recharge over the simulated historic period (1961–2009) for each month has been calculated (see Appendix 2 - Occurrence under 25% and exceedance of 75% recharge values). Daily recharge values produced by the historic simulation are aggregated to total monthly recharge values. A list of monthly recharge values was derived for each month, ranked from the greatest to the lowest and the 25th and 75th percentiles were calculated. The number of occurrences where future total monthly recharge values, simulated in the period between 2010 and 2099, exceed the 75th percentile and the number of values occurring below the 25th percentile were then calculated for each month. These calculations are presented in Appendix 2 where recharge values are in mm/month.

Monthly changes

The mean monthly recharge values were calculated for each month for the simulated historic period (see Appendix 3 - Mean monthly change). The change between future and historical recharge value in absolute terms was calculated for the 2020s (2010–2039), 2050s (2040–2069) and 2080s (2070–2099).

Change factors summary

Using the standard change factor methodology as used by the Environment Agency for water resource assessment, summary plots were produced of average monthly recharge (Minimum, maximum and median) across all 11 ensembles (see Appendix 4 - Change factors). The change factors (percentage difference between future and historical average monthly recharge for each month) were calculated for each month for all 11 ensembles and for all the groundwater bodies. This was undertaken for the 2050s and 2080s and summarised in plots showing greatest negative change factor, greatest positive change factor, median change factor from all 11 ensembles for each groundwater body for each month. This method, whilst not the only method for undertaking this, produces an appreciation of the range of results from the 11 ensembles.

The detailed methodology is as follows:

  • Produce mean monthly recharge for each ensemble for each groundwater body (310 in all) for 1961–1990 within the historic simulation period and for the 2050s (2040–69) and 2080s (2070–2099)
  • Calculate mean monthly recharge values for each ensemble for each groundwater body (310 in all) for the simulated future periods: the 50s (2040–2069) and the 80s (2070–2099)
  • Calculate future 50s and 80s change factors using the mean monthly recharge for each month for all 11 ensembles for each groundwater body (each month will have 11 ensembles for each groundwater body)
  • Calculate the greatest negative change factor, greatest positive change factor and median value for each groundwater body for each month for the 2050s and 2080s
  • Produce the following sets of plots:
  1. Baseline (1961–90): minimum average monthly recharge, maximum average monthly recharge and median average monthly recharge from all 11 ensembles for each groundwater body for each month (3 x 12 plots)
  2. 2050s (2040–2069): greatest negative change factor, greatest positive change factor, median change factor from all 11 ensembles for each groundwater body for each month (3 x 12 plots)
  3. 2080s (2070–2099): greatest negative change factor, greatest positive change factor, median change factor from all 11 ensembles for each groundwater body for each month (3 x 12 plots)

River basin management districts

To understand the impact of Climate Change on potential recharge on the main catchments used for River Basin Management District (RBMD) planning the model results were summarised over the extents of these catchments (see Appendix 5 - River basin management districts). There are 11 RBMD in England and Wales numbered from 2 to 12 as illustrated by Figure 2.

The results for the RBMDs were averaged for each RBMD as a region and also for each ensemble.

Total volumes of recharge are calculated for each RBMD for 1961–90, 1971–2000 and for the 2020s, 2050s and 2080s.

File:OR17026fig2.jpg
Figure 2    Distribution of river basin management districts (Contains public sector information licensed under the Open Government Licence v3.0).

Monthly changes

The mean monthly recharge value was calculated for each month. The change in recharge value in absolute terms was calculated for the 2020s (2010–2039), 2050s (2040–2069) and 2080s (2070–2099).

These results are presented as annual time series plots for each month (see Appendix 5 - River basin management districts).

Empirical cumulative distribution functions (ECDFS)

ECDFs are produced by ranking the data from the smallest to the largest value. By putting the data in ascending order and then for every assigned value x, the number of data points less than or equal to x is determined. This number is divided by the sample size to calculate a probability of the occurrence of x. Each value is then plotted against the cumulative probability from the smallest to the largest to obtain the ECDF curve. Using this approach allows the median (50%tile) of each distribution to be compared so that change can be assessed. Further the slope of the line can be used to indicate whether the nature of the distribution changes. For example two ECDF plots, one with an increased median value but both with similar slopes means that the distributions are identical, but that the values are generally greater for the ECDF with the higher median value. Increasing slope means that the distribution is more ‘spikey’ with a smaller standard deviation.

ECDFs have been produced by totalling the recharge produced for each RBMD both seasonally (winter, spring, summer and autumn) and monthly for two historic simulation periods (1961–90 and 1971–2000) and for the 2050s and 2080s.

References

  1. 1.0 1.1 1.2 Prudhomme, C, Dadson, S, Morris, D, Williamson, J, Goodsell, G, Crooks, S, Boelee, L, Davies, H, Buys, G, Lafon, T, and Watts, G. 2012. Future Flows Climate: an ensemble of 1 km climate change projections for hydrological application in Great Britain. Earth System Science Data, 4(1), pp.143–148.
  2. 2.0 2.1 Mansour, M M, and Hughes, A G. 2004. User's manual for the distributed recharge model ZOODRM. British Geological Survey, 61pp. (IR/04/150) (Unpublished).
  3. 3.0 3.1 Mansour, M M, Wang, L, Whiteman, M, and Hughes, A G. 2018. Estimation of spatially distributed groundwater potential recharge for the United Kingdom. Quarterly Journal of Engineering Geology and Hydrogeology, 51, 247-263. https://doi.org/10.1144/qjegh2017-051
  4. 4.0 4.1 Hughes, A G, Mansour, M M, and Robins, N S. 2008. Evaluation of distributed recharge in an upland semi-arid karst system: the West Bank Mountain Aquifer. Hydrogeology Journal. 10.1007/s10040-008-0273-6.
  5. 5.0 5.1 Hulme, P, Rushton, K, and Fletcher, S T. 2001. Estimating recharge in UK catchments. IAHS PUBLICATION. Jul:33-42. Cite error: Invalid <ref> tag; name "Hulme 2001" defined multiple times with different content
  6. Griffiths, J, Keller, V, Morris, D, and Young, A R. 2006. Continuous Estimation of River Flows (CERF) — Technical Report: Task 1.3: Model scheme for representing rainfall interception and soil moisture. Environment Agency R & D Project W6–101. Centre for Ecology and Hydrology, Wallingford, UK.
  7. 7.0 7.1 Boorman, D B, Hollist, J M, and Lilly, A. 1995. Hydrology of soil types: a hydrologically-based classification of the soils of the United Kingdom. Institute of Hydrology Report No. 126. Wallingford, UK Cite error: Invalid <ref> tag; name "Boorman 1995" defined multiple times with different content
  8. 8.0 8.1 Fuller, R M, Smith, G M, Sanderson, J M, Hill, R A, and Thomson, A G. 2002. The UK Land Cover Map 2000: construction of a parcel-based vector map from satellite images. The Cartographic Journal, 39(1), 15–25. Cite error: Invalid <ref> tag; name "Fuller 2002" defined multiple times with different content
  9. Keller, V, Young, A R, Morris, D, and Davies, H. 2005. Continuous Estimation of River Flows (CERF). Technical Report: Estimation of Precipitation Inputs, Environment Agency R {&} D Project Report WD-101, Centre for Ecology and Hydrology, Wallingford
  10. Hough, M N, and Jones, R J A. 1997. The United Kingdom Meteorological Office rainfall and evaporation calculation system: MORECS version 2.0 — an overview. Hydrology and Earth System Sciences 1 (2): 227–239.
  11. Morris, D G, and Flavin, R W. 1990. A digital terrain model for hydrology, Proceeding of Fourth International Symposium on Spatial Data Handling, July 23–27, Zurich, Vol. 1 pp.250–262.
  12. Moore, R V, Morris, D G, and Flavin, R W. 1994. Sub-set of UK digital 1: 50 000 scale river centre-line network. NERC, Institute of Hydrology, Wallingford.
  13. Natural Environment Research Council (NERC) 2000. Countryside Survey 2000 Module 7. Land Cover Map 2000 Final Report. Centre for Ecology and Hydrology, Wallingford, UK .
  14. Murphy, J M, Booth, B B B, Collins, M, Harris, G R, Sexton, D M H, and Webb, M J. 2007. A methodology for probabilistic predictions of regional climate change from perturbed physics ensembles. Phil. Trans. R. Soc. A 365, 1993–2028.
  15. Jenkins, G J, Murphy, J M, Sexton, D S, Lowe, J A, Jones, P, and Kilsby, C G. 2009, UK Climate Projections: Briefing report, Met Office Hadley Centre, Exeter, UK.
  16. 16.0 16.1 Murphy, J M, Sexton, D M H, Jenkins, G J, Boorman, P M, Booth, B B B, Brown, C C, Clark, R T, Collins, M, Harris, G R, Kendon, E J, Betts, R A, Brown, S J, Howard, T P, Humphrey, K A, McCarthy, M P, McDonald, R E, Stephens, A, Wallace, C, Warren, R, Wilby, R, and Wood, R A. 2009, ‘UK Climate Projections’ Science Report: Climate change projections. Met Office Hadley Centre, Exeter.
  17. IPCC: 2000, Special report on emissions scenarios (SRES): A special report of Working Group III of the Intergovernmental Panel on Climate Change, Cambridge University Press, Cambridge, p.599.
  18. Prudhomme, C, and Williamson, J. 2013. Derivation of RCM-driven potential evapotranspiration for hydrological climate change impact analysis in Great Britain: a comparison of methods and associated uncertainty in future projections. Hydrology and Earth System Sciences, 17(4), p.1365.
  19. Prudhomme, C, Haxton, T, Crooks, S, Jackson, C, Barkwith, A, Williamson, J, Kelvin, J, Mackay, J, Wang, L, Young, A, and Watts, G. 2013. Future Flows Hydrology: an ensemble of daily river flow and monthly groundwater levels for use for climate change impact assessment across Great Britain. Earth System Science Data, 5(1), pp.101–107.