Risk Characterization

As shown in Figure 1.3, risk characterization is an important step in the overall process of risk management, ttis section describes the methodology and pro-

Likelihood that soybean rust reaches severity > 20% by late June if found in late May

Fig. 1.5. Likelihood that soybean rust will reach a severity of over 20 percent by late June if found in late May (from Del Ponte and Yang 2006).

J events In 10 yeart

01-1

Likelihood that soybean rust reaches severity > 20% by late June if found in late May

J events In 10 yeart

01-1

Fig. 1.5. Likelihood that soybean rust will reach a severity of over 20 percent by late June if found in late May (from Del Ponte and Yang 2006).

Fig. 1.6. Percent decrease in total cereal production for Niger as a function of the National Rainfall Index (Gommes 1998).

Risk Levels

Fig. 1.7. Drought risk for Gujarat, India, determined by integrating risk maps for both agricultural and meteorological drought (Chopra 2006).

Percent change in kharif rainfall from normal

19 20 21 22 23 24 Seasonal mean temperature (°C)

Fig. 1.8. Relationship between (a) monsoon season food production and seasonal rainfall and (b) regional wheat yields with seasonal temperature (Government of India 2004).

Percent change in kharif rainfall from normal

19 20 21 22 23 24 Seasonal mean temperature (°C)

Fig. 1.8. Relationship between (a) monsoon season food production and seasonal rainfall and (b) regional wheat yields with seasonal temperature (Government of India 2004).

Risk Levels

ZlNonsfc

V.r. Hiuli

Fig. 1.7. Drought risk for Gujarat, India, determined by integrating risk maps for both agricultural and meteorological drought (Chopra 2006).

vides some illustrative results for characterizing levels of risk associated with both weather extremes and climate anomalies.

WeatherExtremes tte return period (also know as the recurrence interval of an event) is a statistical measure of how often an extreme event of a given magnitude is likely to be equalled or exceeded, within a given time frame. For example, a "fifty-year rainfall event" is one which will, on the average, be equalled or exceeded once in any fifty-year period. Note: it does not mean that the event occurs every fifty years.

tte likelihood or probability that an event of specific magnitude will be equalled or exceeded in any given year is the inverse of the return period, that is, 1 / Return Period

A one in fifty-year event has one chance in fifty of occurring in any specified year, that is, its probability is 1/50. ttus the probability equals 0.02, or 2%.

In some cases it is useful to know the probability that an event of at least a given magnitude will occur within a specified number of years, say five years, ttis probability can be calculated using the following equation:

probability of occurrence in n years =1-

(1 - probability of occurrence in anyyear)"

For example, the probability that an event with a probability of 0.2 will occur in the next five years is:

Note again that this probability applies only on average, and cannot be considered a forecast.

Table 1.1 provides return periods and probabilities for given extremes in daily rainfall, based on observed data for Delhi and Pune, India. It is clear that extreme rainfall events of a given magnitude at Delhi are substantially more frequent than those observed in Pune.

Similarly, return periods and probabilities for specified values of maximum air temperature and extreme wind speeds are given in Tables 1.2 and 1.3, respectively, tte results show that both extreme high temperatures and extreme wind gusts are much more common at Delhi, relative to Pune.

Table 1.1. Return Periods for Daily Rainfalls of Given Amounts, for Delhi and Pune, India. Based on Data for 1969 to 2004, inclusive. [Data courtesy of India Meteorological Department]

Daily Rainfall of at Least (mm)

Delhi

Pune

Return Period (y)

Probability

Return Period (y)

Probability

50

1.1

0.94

1.2

0.80

75

1.3

0.75

2.3

0.40

100

2.0

0.49

5.8

0.20

125

3.6

0.28

16

0.06

150

6.9

0.14

48

0.02

175

14

0.07

140

0.01

200

28

0.04

>400

0.00

Table 1.2. Return Periods for Maximum Temperatures of Given Amounts, for Delhi and Pune, India. Based on Data for 1969 to 2004, inclusive. [Data courtesy of India Meteorological Department]

Maximum Temperature ofatLeast (°C)

Delhi

Pune

Return Period (y)

Probability

Return Period (y)

Probability

41

1

1

1.6

0.61

42

1

0.99

6.2

0.16

43

1.2

0.86

31

0.03

44

1.9

0.53

160

0.01

45

4.0

0.25

>800

0.0

46

9.7

0.10

47

25

0.04

Table 1.3. Return Periods for Maximum Annual Wind Gusts of Given Amounts, for Delhi and Pune, India. Based on Data for 1969 to 2004, inclusive. [Data courtesy of India Meteorological Department]

Daily Annual Wind Gust of at Least (km h~')

Delhi

Pune

Return Period (y)

Probability

Return Period (y)

Probability

50

1

1

1.1

0.90

75

1.1

0.92

2.2

0.46

100

2.4

0.42

6.6

0.15

125

9.1

0.11

23

0.04

150

41

0.02

83

ClimateAnomalies

Drought has a major impact on agricultural production, making it an important risk condition. Figure 1.9 shows the frequency of drought for Delhi, where in this instance drought is defined as months when the rainfall is at or below the ten-per-centile for that month. It is clear that, based on this indicator, there is a high risk of at least a brief drought occurring in any given year, tte risk of a prolonged drought is also very real.

Fig. 1.9. ^e frequency of drought for Delhi, India, for 1969 - 2004. Drought is defined here as a month when the rainfall is at or below the ten-percentile for that month. Data courtesy of India Meteorological Department.

1969 1974 1979 1984 1989 1994 1999 2004

Year

Fig. 1.9. ^e frequency of drought for Delhi, India, for 1969 - 2004. Drought is defined here as a month when the rainfall is at or below the ten-percentile for that month. Data courtesy of India Meteorological Department.

Changing Risk

Agriculture is one of the main sectors likely to be impacted by climate change, ^is section presents the results of analyses designed to show how risk levels for rainfall and temperature extremes, and drought, are projected to change over the remainder of the current century. One such change of importance to India is illustrated in Figure 1.10. A substantial increase in drought risk is expected during the current century. One consequence is a major decline in irrigated wheat yields in northern India during the coming decades (Figure 1.11). Clearly, and as would be expected, as the time horizon increases the confidence in the projections decreases.

Future changes in risk are estimated using the outputs of selected global climate models (GCMs)1 run for a range of greenhouse gas emission scenarios (Figure 1.12). Table 1.4 lists the combination of models and emission scenarios on which the risk projections are based.

Differences in climate projections give rise to uncertainties in the estimated values of future climate risks, ^ere are numerous sources of uncertainty in projections of the likelihood components of climate-related risks, ^ese include uncertainties in greenhouse gas emissions as well as in modelling the complex interactions and responses of the atmospheric and ocean systems. Policy and decision makers need to be cognizant of uncertainties in projections of the likelihood components of extreme events.

Observed Dryness Index

Hump

Fig. 1.10. Areas in India prone to drought (a) today and (b) in the mid 21st century, determined using a dryness index, ^e light shading indicates areas where rain exceeds evaporation, ^e darker shading identifies regions where evaporation is greater than precipitation - the darker the shading the drier the region, except that urban areas have the darkest shading (Schreiner 2004).

Fig.1.11. Simulated impact of global climate change on irrigatedwheat yields in North India (Aggarwal 2002).

Fig. 1.10. Areas in India prone to drought (a) today and (b) in the mid 21st century, determined using a dryness index, ^e light shading indicates areas where rain exceeds evaporation, ^e darker shading identifies regions where evaporation is greater than precipitation - the darker the shading the drier the region, except that urban areas have the darkest shading (Schreiner 2004).

Best estimates of future risk levels are based on an average of the estimates using a multi model and emission scenario ensemble, tte range in uncertainty is determined using a model and emission scenario combination that produces the maximum and minimum rate of change in future risk levels.

Projected changes in the return periods of extreme dailyrainfall events (Figure 1.13) are based on estimates using a multi model and emission scenario ensemble (see Table 1.4). It is anticipated that global warming will reduce the return periods

Exposure Dryness Index

Exposure Dryness Index

Tablel.4. Available Combinations of Global Climate Models and Emission Scenarios1

CGCM

CSIRO

Hadley

NIES

GFDL

See Text

A1B

T, P1

T, P

T, P

T, P

S

W1

A1F

T, P

T, P

T, P

T, P

S

w

AIT

T, P,S

T, P, S

T, P, S

T, P

s

w

A2

T, P, S

T, P, S

T, P, S

T, P

s

w

B1

T, P, S

T, P, S

T, P, S

T, P

s

w

B2

T, P, S

T, P, S

T, P, S

T, P

s

w

Fig. 1.12. Scenarios of C02 gas emissions and consequential atmospheric concentrations of C02 (from IPCC 2001).

of extreme daily rainfall events for Delhi - that is, the likelihood of such extreme events will increase in the future.

Projected changes in the return periods of extreme maximum temperature (Figure 1.14) are based on estimates using a multi model and emission scenario ensemble (see Table 1.4). It is anticipated that global warming will also reduce the return periods of extreme maximum temperatures for Delhi.

Estimates of changes in maximum wind gusts are based on the assumption that such wind gusts will increase by 2.5, 5 and 10 per cent per degree of global warming. ttus the emission scenarios listed in Table 1.4 are explicitly included in the estimates. tte best estimate of the increase in maximum wind gusts is determined

100 125 150 175 200 225 25C Daily Rainfall (mm)

Fig. 1.13. Relationship between daily rainfall and return period for Delhi, India, for present day (black line) and 2050 (blue lines), ^e uncertainty envelope shows the maximum and minimum estimates of return periods for 2050, based on all possible combinations of the available global climate models and emission scenarios.

100 125 150 175 200 225 25C Daily Rainfall (mm)

Fig. 1.13. Relationship between daily rainfall and return period for Delhi, India, for present day (black line) and 2050 (blue lines), ^e uncertainty envelope shows the maximum and minimum estimates of return periods for 2050, based on all possible combinations of the available global climate models and emission scenarios.

Fig. 1.14. Relationship between maximum temperature and return period for Delhi, India, for present day (black line) and 2050 (blue lines), ^e uncertainty envelope shows the maximum and minimum estimates of return periods for 2050, based on all possible combinations of the available global climate models and emission scenarios.

44 45 46 47 48 49 50

Maximum Temperature (C)

Fig. 1.14. Relationship between maximum temperature and return period for Delhi, India, for present day (black line) and 2050 (blue lines), ^e uncertainty envelope shows the maximum and minimum estimates of return periods for 2050, based on all possible combinations of the available global climate models and emission scenarios.

by averaging the ensemble of estimates for all combinations of percentage increase and emission scenarios. As indicated in Figure 1.15, global warming will likely reduce the return periods of maximum wind gusts for Delhi.

Maximum

W rid

Gust

Fig. 1.15. Relationship between peak wind gust and return period for Delhi, India, for present day (black line) and 2050 (blue lines), ^e uncertainty envelope shows the maximum and minimum estimates of return periods for 2050, based on all possible combinations of the percentage increases and emission scenarios.

Maximum

W rid

Gust

Fig. 1.15. Relationship between peak wind gust and return period for Delhi, India, for present day (black line) and 2050 (blue lines), ^e uncertainty envelope shows the maximum and minimum estimates of return periods for 2050, based on all possible combinations of the percentage increases and emission scenarios.

0 0

Post a comment