Impacts Scenarios

1
Impacts Scenarios

• IPCC (1994). A scenario is not a prediction. A
  coherent, internally consistent and plausible
  description of possible future state of the
  world Four criteria should be met.
• Consistent with global predictions
• Physically plausible
• Applicable in impact assessment
• Representative
• Straight forward to obtain

2
UKCIP http//www.cru.uea.ac.uk/link/
3
(No Transcript)
4
(No Transcript)
5
Seasonal Variation
6
More detail on UK scenarios
7
Scenarios of hydrological application

• Scenarios should be as simple as possible
• Extra complexities add cost and hide
  uncertainties in process
• Example Requirements, Impact assessment

8
Impact Assessment
Landuse Change
9
Comparison between modelled and observed rainfall
current climate
Mean 1/?
10
Comparison of high flows

• Compare high flows using cumulative probability
  distribution and/or flood frequency statistics

PQgtq11/n, FQ(q1)1-1/n
PQgtq22/n, FQ(q1)1-2/n
11
1.00
0.90
0.80
0.70
0.60
Relative cumulative Frequency
0.50
Observed floods
0.40
Simulated floods
0.30
0.20
0.10
0.00
0
100
200
300
Annual maxima
Fit and extreme value distribution to each and
extrapolate to low probability events
12
Alternative for comparison of less extreme flows
Flow duration curve Very similar to
procedure for extreme flows Operate on all flows
not just annual extremes Flow duration curve
displays the time flow are exceeded
Rank all flows Calculate exceeded
13
Low flows compared using Q95 (flow rate esceeded
95 of time)
14
Implementation of modelsApply only one scenario

• Calibrate both rainfall and rainfall-runoff model
  for current climates
• Should have generated a PE record for a future
  climate in which the mean temperature has
  increased by 20C and the forest has matured in
  the Forsinard catchment. 20C increase corresponds
  to the high UKICP scenario (2050).
• Need to generate a synthetic rainfall series that
  is consistent with medium-high UKICP scenario.
• Change parameter in rainfall model to reflect
  average increase of 8 rainfall. (only partly
  consistent why?)
• Rerun TOPMODEL with new PE and rainfall record
• Compare lowflows and high flows of current and
  future scenarios using Q95 and flow duration
  curves respecively.

15
References