Assessment of the vulnerability and the climate change impact on - PowerPoint PPT Presentation

Assessment of the vulnerability and the climate change impact on the water resources of an insular Mediterranean Watershed S. Nabih, O. Tzoraki, L. Benaabidate and P. Zanis Ourania Tzoraki Ass. Professor University of the Aegean Adapt to Climate Conference Heraklion, June 2019 Université Sidi Mohammed Ben Abdellah Faculté des Sciences et Techniques Aristotle University of Thessaloni

Presentation Layout • Quiz about intermittent rivers • Methodology • Calibration/validation of SWAT model, Sensitivity analysis, CC models prediction, Low and high fmows Indicators (IHA & SMIRES TOOLS) • Results • Recommendations

Do you want to assess your expertise in river fmow intermittence? Ready for the Quiz? Let’s play!

Question 1 Where is the gauging station with the highest proportion of zero? Wrong answer  In Italy Celon River: Freq=19% Wrong answer  In France Albarine River: Freq=17% Good answer  In Cyprus Yialias River: Freq=64% Next question

Question 2 Are intermittent rivers well monitored in France? Which proportion of gauging stations is monitoring intermittent streams? Less than 1% Between 1 and 10% More than 10% Next question

Question 3 What could the maximum value of the mean annual fmow (m 3 /s) within the dataset? Less than 0.1 m 3 /s Between 0.1 and 1 m 3 /s Between 1 and 10 m 3 /s More than 10 m 3 /s Next question

Question 4 Where was this picture taken? In Italy In Spain In Cyprus In the UK Next question

Question 5 Can you click on the picture related to a perennial river? Back to the fjrst page

This catalogue is one of the deliverables of the COST Action CA15113 (SMIRES, Science and Management of Intermittent Rivers and Ephemeral Streams, www.smires.eu), supported by COST (European Cooperation in Science and T echnology). Back to the fjrst page

Case Study Tsiknias watershed

Flood alert system of North Aegean Meteorological station in Automatic telemetric Stypsi station in Kalloni village

Flood alert system of North Aegean Levellogger operation in Agia Paraskevi Measuring the fmow to extract rating- curve

SWAT Model

 Continuous, physically based, and distributed Model The hydrology Water balance  Predict the impact of land management practices on Sediment yield equation Contaminant transport  SWAT subdivides a watershed into subbasin connected by a stream network, and further delineates HRUs (Hydrological response units) consisting of unique combinations of land cover, soils and slopes within each sub-basin. Hydrological Watersh Subbasin response ed s units

Tsiknias Tsiknias Assessment of the Assessment of the • Hydrology ( runoff, Watershed Watershed Climate Change Climate Change interception by the Impacts on the Impacts on the canopy, Objectif of hydrologic Objectif of hydrologic DEM DEM evapotranspiration, the research regime of the research regime of drainage, Tsiknias river Tsiknias river percolation , subsurface runoff , Watershed Watershed DEM setup DEM setup reservoirs , wetlands ) Delineation Delineation • Cl imate ( soil Stream Stream temperature, snow, definition definition climate generator ) • Crop growth, HRUs Outlet definition Outlet definition HRUs • Farm management definition definition (irrigation, rotation, Slope Soil Landuse Slope Soil Landuse Calculation of Calculation of pesticides) sub basins sub basins Weather Weather • Transfer of water parameters parameters Input data Input data into the main reach Tables Tables • Sedimentation , Model Model • Nutrients . Precipitation, Temperature Precipitation, Temperature Calibration Calibration SWAT SWAT Meteorological data from (1955- Model Meteorological data from (1955- Model run run 2016) (Tmin, Tmax, Prec), Flow Evaluation and 2016) (Tmin, Tmax, Prec), Flow Evaluation and (2014-16) validation (2014-16) validation

The SWAT model performance to be evaluated using • R2 (the coeffjcient of determination that evaluates the correlation between two series), • RMSE (the root mean squared error, which evaluates the deviation), • Percent bias (PBIAS) measures the average tendency of the simulated values to be larger or smaller than their observed ones. The optimal value of PBIAS is 0.0, with low-magnitude values indicating accurate model simulation. Positive values indicate overestimation bias, whereas negative values indicate model underestimation bias • the Nash–Sutclifge Effjciency ( NSE) (the goodness of-fjt criterion for the predicated and observed values)

Calibration results 22 1400,0 21 emperature (°C) 1200,0 20 Precipitation (mm) 19 1000,0 18 800,0 17 f(x) = - 0,01x + 17,81 16 600,0 15 400,0 200,0 T 0,0 0 3 6 9 2 5 8 1 4 7 0 agia paraskevi 8 8 8 8 9 9 9 0 0 0 1 - - - - - - - - - - - 9 2 5 8 1 4 7 0 3 6 9 Linear (agia paraskevi) 7 8 8 8 9 9 9 0 0 0 0 9 9 9 9 9 9 9 0 0 0 0 1 1 1 1 1 1 1 2 2 2 2 Linear (agia paraskevi) Linear (agia paraskevi) Agia paraskevi Stipsi

Sensitivity Adjusted values Initial Parameter Defjnition Range values 1 2 3 analysis Soil evaporation ESCO 0-1 0.95 0.95 0.95 0.95 compensation factor Plant uptake EPCO 0-1 1 0.2 0.2 0.2 compensation factor lat_ttime(days) Lateral fmow travel time 0-18 0 1 2 3 GW_DELAY Ground water delay 0-500 31 0 0 0 (days) time 0.04 ALPHA_BF(days) base fmow travel time 0-1 0.048 0.9 0.9 8 RCHRG_DP(fracti deep aquifer 0-1 0.05 0.3 0.3 0.3 on) percolation fraction Threshold depth of water in the shallow GWQMIN (mm) 0-5000 1000 400 400 400 aquifer required for return fmow to occur Ground water revap 0.02- GW_REVAP 0.2 0.2 0.2 0.02 coeffjcient 0.2 Threshold depth of water in the shallow 100 100 REVAPMN (mm) 0-1000 750 1000 aquiferfor revap to 0 0 occur initial SCS runofg curve 72-74-77- 36 CN2 number for moisture 35-98 45 54 83-85-87 condition Depth from soil surface 300-600- Sol_Z(mm) 0-3500 150 300 450 to bottom layer 900 available water 0.38-0.41- Sol_AWC(mm/m 0.2- capacity of the soil 0-1 0.42-0.44- 0.4 5 m) 0.3 layer 0.48 36-136- saturated hydraulic SOL_K (mm/hr) 0-2000 236-336- 2.2 2.6 3.6 conductivity 360 Efgective hydraulic conductivity in -0.01 - CH-K2 (mm/hr) 0 5 15 500 tributary channel 150 alluvium (mm/hr)

Calibration results Daily Statistics of Goodness of Fit Nash-Sutclifge M o n t h l y s im u l a t e d fm o w ( m 3 / s ) Effjciency NSE 0.53 >0.5 Percent Bias PBIAS - 25% 5 Monthly fmow (m3/s) 15.5 4 RMSE Standard 1.24 3 Deviation Error 2 RSR 0.75 <0.7 1 5 4 RMSE/STDEV f(x) = 1,22x - 0,03 0 3 R² = 0,95 4 4 5 5 5 5 6 1 1 1 1 1 1 1 2 - - - - - - - r l t n l t n u p u c c a a 1 J O A J O J J 0 0 0,5 1 1,5 2 2,5 3 3,5 Simulated fmow Monthly observed fmow (m3/s)

RCM institution RCM RCM GCM GCM Abbreviati RCP4 RCP Histori name institutio model model Institut on 5 85 cal n name name ion acronym Acrony m Climate Limited- CLMcom CCLM4-8- CNRM- CNRM- 17 CM5 CERFAC Geocradle tools area Modelling– CLMcom- S Community CNRM 2 5 A Centre national des CNRM ALADIN53 CNRM- CNRM- CM5 CERFAC recherches CNRM- S météorologiques CNRM 4 6 B Koninklijk KNMI RACMO22 EC- ICHEC E EARTH Nederlands Meteorologisch KNMI- Instituut ICHEC 7 8 D Institut Pierre- IPSL-INERIS WRF331F CM5A- IPSL- MR IPSL Simon-Laplace IPSL-IPSL 9 11 E Sveriges SMHI RCA4 CM5A- IPSL- MR IPSL Meteorologiska och HydrologiskaInstitut SMHI-IPSL 10 12 F Climate Limited- CLMcom CCLM4-8- HadGM MOHC 17 2-ES area Modelling– CLMcom- Community MOHC 13 15 H Sveriges SMHI RCA4 HadGM MOHC 2-ES Meteorologiska och SMHI- HydrologiskaInstitut MOHC 20 14 G CLMcom CCLM4-8- MPI- MPI-M Climate Limited- 17 ESM-LR area Modelling– CLMcom- Community MPI 17 19 I Max Planck Institute MPI-CSC REMO200 MPI- MPI-M 9 ESM-LR Magdeburg MPI-MPI 18 3 j

CC models fmow simulation results

Hydrologic alteration values-IHA software The Range of Variability Approach (RVA) compares the variation in the IHA parameters from the pre-impact period to the variation in the post-impact period (or reference vs. alternative scenarios) to determine the extent of the changes. Each IHA parameter is analyzed to determine the frequency with which it falls into one of three RVA categories (Low, Middle, High), as defined by the RVA Category Boundaries. RVA requires at least two years of pre-impact data.

Low and high fmow indicators • Quantiles of fmow The Quantile of fmow is defjned as the fmow value, which exceeds a specifjc time interval (i.e. 90 ή 95). Q90 and Q95 are defjned as discharges that exceeded 90% and 95% of the time, respectively, over the full duration of record at each measurement location. For example the 95 percentile of fmow or Q95, the fmow that is exceeded for 95% of the period of record. They represent warning levels and limiting conditions (Q90) as well as the bases for biological and ecological indices (Q95) and for limits for surface‐water extractions and effmuent discharges. 34

High fmow indicators • Flow Exceedance Probability

Temporal regime plot - SMIRES toolset

RCP45