Spatially distributed, physically-based modelling for simulating the impact of climate change on groundwater reserves
9
0
0
Texte intégral
(2) Coupled and integrated modelling ‘Soil model’. Watershed discretised into grid squares climatic data. A B. D. B C. I II. precipitations temperatures solar radiation. pedologic map : majority class for each cultural entity. Homogeneous cultural entities into the grid square slope map: average slope for each cultural entity. Precipitations. Evaporation and transpiration. Snow. Subsurface flow. Percolation. Water balance for each cultural entity. Coupled and integrated modelling. C. Sohier & S. Dautrebande Hydraulique agricole, HA-FUSAGx, Gembloux, Belgium. Ponderation of the entities results by grid square and Water balance for each grid square. Reference: Sohier C., Moeremans B., Deglin D. and Dautrebande S., 2000, Validation of the new catchment hydrological model EPIC-GRID for soil moisture evaluation and discharges simulation. Presented at the Workshop “The future of distributed hydrological modelling”, Leuven, Belgium, April 2000.. ‘Groundwater model’ Hydrogeology Group, GeomaC Dpt , Faculty of Engineering, ULg SUFT3D (Saturated Unsaturated Flow and Transport in 3D) is a finite-element code, - for modelling of the unsaturated and saturated zones in 3D; - allowing local mesh refinements depending on local geological characteristics; - seawater intrusions; - realistic tracer injection conditions; - data and results processing by GMS © and full GIS compatibility References: - Brouyère S., 2003, Modeling tracer injection and well-aquifer interactions: a new mathematical and numerical approach, Water Resources Research 39(3), 1070, doi:10.1029/2002WR001813 - Carabin, G., Dassargues A., 1999, Modeling groundwater with ocean and river interaction, Water Resources Research, 35(8), 2347-2358. - Carabin G., Dassargues A. and Brouyère S., 1998, 3D flow and transport groundwater modelling including river interactions, in Computational Methods in Water Resources XII, Vol. 1: , pp. 569 - 576.. HA-FUSAG. Coupled and integrated modelling. Coupled and integrated modelling Simplified general schema. ‘River Model’ J-F. Deliège & J. Smitz Environment Centre (CEME), ULg. rainfall Hydraulique Agricole FUSAGx. River model sub-model simulates water dynamics in the river network, as the result of input of water fluxes from the soils and of transfers water fluxes from / to the groundwater based on Saint-Venant equations (mass and momentum balances) in a one-dimensional channel. … the surface water sub-model is applied to the entire river network, including the main river and its tributaries. The channel characteristics (slopes, sections, widths, roughness, …) are determined during the pre-processing operations. Reference: - Smitz, J., Everbecq, E., Deliège, J.-F., Descy, J.-P., Wollast, R. and Vanderborght. J.-P., 1997, PEGASE, une méthodologie et un outil de simulation prévisionnelle pour la gestion de la qualité des eaux de surface (PEGASE : a methodology and a forecasting simulation tool for surface water quality management). Tribune de l'Eau. 588, 73-82. … for water quantity evapotranspiration. runoff. Soil Model EPIC Grid. interflow. drainage infiltration Hydrogeology ULG. infiltration. GW model SUFT3D. CEME ULG. River Model. models are implemented in a parallel processing environment data exchange protocol MPI. 2.
(3) 74. 76. 78. Geer basin : hydrology and hydrogeology 60. latitude (deg). 60. Geer basin 58. #. 58. Belgium. P = ETR + QGEER + QOUT + ΔReserves + ΔLosses 810 = 508 + 145 + 88 + 7.5 + 61.5 mm (between 1975 and 1994, after Hallet, 1999) 740 = 525 + 120 + 65 + 15 + 15 mm (between 1951 and 1965, after Monjoie, 1965). Geer basin 450 km2. Unsaturated zone 0. 56. 56 74. 76. 50. 100 Km. 78. longitude (deg). Geer losses. %Kanne. Geer river #. Geer basin 450 km2 A. # Fexhe-Slins. Othée. #. #. Oreye. #. llery. ern ga. north. Waremme. #Bovenistier. Jeneffe. A'. #. ozé. nt mo. -H rion. Ho. so. ry. alle. rn g. e uth. #Awans #LIEGE. water table. lt. fau. 0. 5. 10 Km. #Villers-le-Bouillet. A. A'. geology. : cross-section (fig. 3). Geer basin : hydrogeology and modelling Main characteristics. Loess Sands-conglomerate Grey chalks ‘Hard-ground’ White chalks ‘Smectite of Herve’ Primary bedrock. galleries + pumping. Geer basin : hydrogeology and modelling Main characteristics • piezometric fluctuations up to 15 m: ability of SUFT3D code for considering a highly variable unsaturated zone • seven layers of finite elements, from bottom to top (in general) : - three layers of chalk ; - one layer of hardened chalk named ‘hard ground’’; - one layer of fractured chalk; - one layer of conglomerate / residual sands; - one layer of loess. • the ‘hardground’ is not found in the whole domain; • the conglomerate is replaced northwards by sands; • the thickness of the unsaturated zone can reach 40 m and, when piezometric levels are varying only in the chalk, in this case the 7 layers of finite elements are in the chalk;. 3.
(4) Geer basin : hydrogeology and modelling. Geer basin : hydrogeology and modelling. • the mesh consists of 31423 finite elements (18680 nodes) • a mean element size of about 700 meter • refinements are required in zones where important stresses are applied (faults, galleries, pumping wells, …). Geer basin : hydrogeology and modelling. Geer basin : calibration of the model. High groundwater levels (1983-1984). Low groundwater levels (1991-1992) 200. 1983-1984 1991-1992. 150. 100. 50 50. 100. 150. Measured groundwater head (m). 200. Computed groundwater head (m). 1983-1984. Computed groundwater head (m). 200. 150. 100. 50 50. 100. 150. 200. Measured groundwater head (m). 4.
(5) Geer basin : calibration of the model. Geer basin : calibration of the integrated model in transient conditions. Calculated Darcy’s fluxes. Geer basin : calibration of the integrated model in transient conditions. Calibration: 01/01/1975 – 31/12/1988 Validation: 01/01/1989 – 31/12/1995. 5.
(6) Geer basin : calibration of the integrated model in transient conditions - temporary results. Geer basin : calibration of the integrated model in transient conditions - temporary results. … influence of Δt for GW. Calibration period: 01/01/1975 – 31/12/1988 Validation period: 01/01/1989 – 31/12/1995. Zoom on the period june 80 – dec 81 Influence of the time step. Geer basin : calibration of the integrated model. Geer basin : calibration of the integrated model. 3. M eandailyflowrate(m/s). calibration period (1975 - ...) 10 8 6 4. averaged h in the aquifer for ne = 0.05. 2. 0 01 jan 75. 01 jan 77. 01 jan 79. 01 jan 81. 01 jan 83. 01 jan 85. 3. M eandailyflowrate(m/s). Chronology validation period (1989-1995). calibration period (... - 1988) 10 8 6 4 2. 0 01 jan 85. 01 jan 87. 01 jan 89. 01 jan 91. 01 jan 93. 1977. 01 jan 95. Baseflow (m 3/s). Difference (m 3/s). Pumping + Losses (m 3/s). Reserve. Reserve. (m 3/s). (mm). Hchalk(m). 2.55. 0.89. 1.66. 2. -0.34. -22. -0.44. ‘Dry years’. Chronology total flow rate measured in the Geer at Kanne computed total flow rate in the Geer at Kanne computed base flow. Infiltration (m 3/s). Reference Simulation. 6.
(7) Geer basin : calibration of the integrated model. Geer basin : reference simulation. averaged h in the aquifer for ne = 0.05. ‘Dry years’ Reference Simulation ‘Wet years’. ‘Dry years’ Reference Simulation. Infiltration (m 3/s). Baseflow (m 3/s). Difference (m 3/s). Pumping + Losses (m 3/s). Reserve. Reserve. (m 3/s). (mm). Hchalk(m). 1977. 2.55. 0.89. 1.66. 2. -0.34. -22. -0.44. 1986. 3.49. 1.64. 1.85. 2. -0.15. -10. -0.19. 1983. 4.94. 1.97. 2.97. 2. 0.97. 63. 1.26. Infiltration (m 3/s). Baseflow (m 3/s). Difference (m 3/s). Pumping + Losses (m 3/s). Reserve. Reserve. Hchalk. (m 3/s). (mm). (m). 1977. 2.55. 0.89. 1.66. 2. -0.34. -22. -0.44. 1986. 3.49. 1.64. 1.85. 2. -0.15. -10. -0.19. 1983. 4.94. 1.97. 2.97. 2. 0.97. 63. 1.26. 1988. 5.32. 2.07. 3.25. 2. 1.25. 81. 1.62. Hchalk. ‘Wet years’. IPCC climatic scenarios. Full blue line: CGCM1 scenario (Canada) Dashed blue line: ECHAM4 scenario (Germany) Full red line: HadCM2 scenario, the 4th of the Hadley Centre (UK) experiences. Simulation of the climatic scenarios ECHAM4: 2010 – 2039 period. Infiltration (m 3/s). Baseflow (m 3/s). Difference (m 3/s). Pumping + Losses (m 3/s). Reserve. Reserve. (m 3/s). (mm). (m). 2.55. 0.89. 1.66. 2. -0.34. -22. -0.44. Reference Simulation. ‘Dry years’. 1977 1986. 3.49. 1.64. 1.85. 2. -0.15. -10. -0.19. ECHAM4 30 years 2010 - 2030. ‘Dry years’. 1977. 0.99. 0.38. 0.61. 2. -1.39. -90. -1.80. 1986. 2.36. 0.91. 1.45. 2. -0.55. -35. -0.71. 7.
(8) Simulation of the climatic scenarios ECHAM4: 2010 – 2039 period. Simulation of the climatic scenarios: influence on groundwater levels 147.00 146.00. Piezometric level (m). 145.00 144.00. Reference. 143.00. cgcm 1_1039. 142.00. cgcm 1_4069. 141.00. cgcm 1_7099. 140.00. echam 4_1039. 139.00. echam 4_4069. 138.00. echam 4_7099. 137.00. hadcm 2_1039. 136.00. hadcm 2_4069. 135.00. hadcm 2_7099. 134.00 133.00. h of about 7 m. 132.00 131.00 130.00. Reference Simulation. ‘Wet years’. ECHAM4 30 years 2010 - 2030. ‘Wet years’. 1988 1986. 5.32 3.49. 2.07 1.64. 3.25 1.85. 22. 1.25 -0.15. 81 -10. 1.62 -0.19. 1983. 3.99. 1.19. 2.80. 2. 0.8. 52. 1.04. 1988. 4.09. 1.29. 2.80. 2. 0.80. 52. 1.04. …galleries are out of the saturated zone, … inducing numerical problems Reference. 125.00. cgcm 1_1039. 124.00. cgcm 1_4069. 123.00. cgcm 1_7099. 122.00. echam 4_1039. 121.00. echam 4_4069. 120.00. echam 4_7099. 119.00. hadcm 2_1039. . hadcm 2_4069. 118.00. hadcm 2_7099. 117.00 116.00 115.00 114.00 113.00 112.00 111.00 01/01/96. 01/01/95. 01/01/94. 01/01/93. 01/01/92. 01/01/91. 01/01/90. 01/01/89. 01/01/88. 01/01/87. 01/01/86. 01/01/85. 01/01/84. 01/01/83. 01/01/82. 01/01/81. 01/01/80. 01/01/79. 01/01/78. 01/01/77. 01/01/76. 01/01/75. 01/01/74. 01/01/73. 01/01/72. 01/01/71. 110.00 01/01/70. 01/01/94. 01/01/96. 01/01/93. 01/01/95. 01/01/92. 01/01/91. 01/01/90. 01/01/89. 01/01/88. … the most extreme scenario: ECHAM4 for the 1st period 2010 to 2039 on the basis of incremental changes of T and P/P with regards to the 30 last years. . 126.00. 01/01/69. 01/01/87. Chronology. Conclusions. Simulation of the climatic scenarios: influence on groundwater levels. Piezometric level (m). 01/01/86. 1.26. 01/01/85. 63. 01/01/84. 0.97. 01/01/81. 2. 01/01/83. 2.97. 01/01/80. 1.97. 01/01/82. 4.94. 01/01/79. 1983. 01/01/78. (m). 01/01/77. (mm). 01/01/76. (m 3/s). 01/01/75. HChalk. 01/01/74. Reserve. 01/01/73. Reserve. 01/01/72. Pumping + Losses (m 3/s). 01/01/71. Difference (m 3/s). 01/01/70. Baseflow (m 3/s). 01/01/69. 129.00. Infiltration (m 3/s). . Chronology. … the most extreme scenario: ECHAM4 for the 1st period 2010 to 2039 on the basis of incremental changes of T and P/P with regards to the 30 last years. . for reliable and detailed predictions … ‘physically consistent’ and ‘spatially distributed’ simulations ; the impact of climate changes on groundwater reserves is far from easy to be predicted: here, strong assumptions were chosen: unchanged pumping; unchanged land use; climatic scenarios taken into account by incremental changes with regards to the situation of now; scenarios uncertainty is not taken into account; trends are showing deficits: bad news because it will be probably worst with the introduction of increasing irrigation and introduction of the more frequent climatic extremes, etc. what about impact on the groundwater quality ? still a lot of work and lot of challenges !. 8.
(9) Thank you Credits . . . S. Brouyère & G. Carabin, Hydrogéologie et Géologie de l’Environnement, Dpt GéomaC, ULg C. Sohier & S. Dautrebande, Hydraulique agricole, HA-FUSAGx, Gembloux J-F. Deliège & J. Smitz Centre Environnement (CEME-ULg). More details in Brouyère, S., Carabin, G. and Dassargues, A., 2004, Climate change impacts on groundwater reserves: modelled deficits in a chalky aquifer, Geer basin, Belgium, Hydrogeology Journal, DOI 10.1007/s10040-003-0293-1, 12(2), pp.123134. 9.
(10)
Documents relatifs
Neglecting these other sources when estimating the Redfield ratios is possible only if the choice of the measurement points considered for an estimate, and therefore the definition
modification of the fluctuation level at the X-point upon field reversal could therefore be interpreted as the relevance of the poloidal shear in blob disconnection, as the
Detection number of bipartite graphs and cubic graphs 337 between v(C) and the vertex corresponding to this literal is colored
Nollywood directors and producers started to be invited to film festivals around the world to tell the intriguing story of a video industry that in just a few years had managed
Taken together, these studies establish a critical role for Bcl11b in the maintenance of skin homeostasis and stem cell biology through the regulation of key transcription and
Ainsi dans Paradox on peut placer des droits au niveau des tables mais aussi de chaque champ et le moteur crypte les données dès qu’un mot de passe est défini (SQL Base
In this context, the Tiwouh project proposes a technological solution to support AAC interventions. It is deployed on the field and used by individuals with
The issue of the multiplicity of these dust makers is pivotal to understand the dust formation process, and in this context X-ray observations may allow to reveal an X-ray
