Do increases in greenhouse gases cause droughts in Australia? 12


Peter Gallagher reports that even while the coals are still warm, some are already blaming the Victorian fires on increases in greenhouse gases.

The following summarizes indications of decline in droughts in Australia from 1900 to the present, compiled from data provided with the Drought Exceptional Circumstances Report. Some of this information was provided in the submission the the Australian Meteorological Magazine (more about this tomorrow). Drought is defined as the percentage of area with rainfall lower than the 5th percentile. The areas are averaged over seven Australian regions.

1. Moving 30 year Averages

Moving average of climatic trends was illustrated in the DECR (Figure 7), and in Fig 1 from submission to AMM. The figure below shows the overall 30-year running mean of percentage area of exceptionally low rainfall for observations decreasing in almost all areas, and forecasts increasing in all areas with little overlap of the spread of results at 1990.

image003

Figure 2. Overall average (green thick line) of the 30-year running average of percentage area of exceptionally low rainfall for observations is decreasing, in almost all areas (red lines), while models (black lines) are increasing in all areas.

2. Difference of means 1900-1950 vs. 1950-2007

Table 1 is a mean comparison for two different periods. For observations, the mean of droughted area in all 7 regions over the period. Observed droughts decreased significantly from 1900-1950 to 1951-2007. The p values for a non-parametric Mann-Whitney test, used because the observations are not normally distributed, indicate the differences between the periods are highly significant.

Table 1: Mean percentage area of exceptionally low rainfall over time periods. A Mann Whitney rank-sum test shows significant differences between periods.

1900-2007 1900-1967 1951-2007 P 1900-2007 vs. 1951-2007 P 1900-1950 vs. 1951-2007
Observed % Area Drought 5.6±0.5 6.2±0.7 4.9±0.6 0.10 0.004

3. Kruskal-Wallis Rank Sum Test

A non-parametric method for testing equality of population medians among groups, the Kruskal-Wallis test does not assume a normal population, and does assume an identically-shaped and scaled distribution for each group, except for any difference in medians. The result is the same as the Mann-Whitney test, significant decrease in drought in the last 50 years.

P 1900-2007 vs. 1951-2007 P 1900-1950 vs. 1951-2007
Observed % Area Drought 0.10 0.004

4. Differences in Shape

The distribution of observed values appears to be a Pareto (power) distribution, which would be expected for a peaks-over-threshold statistic, emerging from the way drought is defined. A pareto distribution is given by:

$$Pr(X>x)=(\frac{x}{x_m})^{-k}$$

for all $$x ≥ x_{m}$$, where $$x_m$$ called “scale”, and k is a positive parameter called “shape”. When this distribution is used to model the distribution of droughts, the parameter “scale” differs significantly between 1900-1950 to 1951-2007. The R package “Peaks Over Threshold” or POT was used.

Table 2: Shape parameters of a pareto distribution fit to area of exceptionally low rainfall over time periods. P values for probability of false rejection of null hypothesis given by t-tests .

1900-2007 1900-1967 1951-2007 P 1900-2007 vs. 1951-2007 P 1900-1950 vs. 1951-2007
Shape observed 3.2±0.4 3.7±0.6 2.8±0.5 0.22 0.1

6. Quantiles of periods

In yet another analysis we subject Table 3 of the DECR to statistical testing. This table contains the average percentage area having exceptionally low rainfall years for selected 40-year periods and the most recent
decade (1998-2007).


1900-1939 1910-1949 1920-1959 1930-1969 1940-1979 1950-1989 1960-1999 1968-2007 1998-2007
Qld 9.5 6.5 5.5 4.1 3.3 3.1 2.7 2.6 4.7
NSW 5.7 6.9 5.7 6.2 5.8 4.3 4.0 3.8 6.4
Vic&Tas 5.3 6.0 4.2 6.1 5.1 5.0 5.3 5.2 8.5
SW 5.2 7.1 7.2 6.9 7.9 5.9 4.9 4.4 3.4
NW* 6.3 5.3 6.5 7.5 6.5 6.1 4.7 3.5 3.3
MDB 6.1 7.2 5.8 6.4 5.7 4.1 3.5 3.5 6.9
SWWA 2.5 4.7 4.1 6.5 8.3 6.1 6.3 8.5 8.9
Australia 6.4 6.4 6.6 6.4 6.3 5.3 4.6 3.5 3.1

Using the function ‘quantile’ in R, we output the percentage areas for each probability in each 40 year period. Then we lookup the probability for each region using the most recent 40 year period 1968-2007.


Quantiles
5% 10% 50% 90% 95%
3.25 4.05 5.85 7.15 7.60

Regions, area and probability drought has increased.
Qld 2.6 <5%
NSW 3.8 <10%
Vic&Tas 5.2 NS
SW 4.4 NS
NW* 3.5 <10%
MDB 3.5 95%
Australia 3.5 <10%

The results show that over the last 40 years, regions Qld, NSW, NW, MDB, and Australia as a whole have had significantly less area under drought. Only in SWWA has the drought area increased significantly, while Vic&Tas (the region of recent bushfires) and SW have no significant change.

Drought in Australia as a whole has significantly declined.

Conclusions

So here are three different approaches, all showing that the propensity to drought in Australia is decreasing with increases in greenhouses gases (using the data from the DECR, defined as the percentage area with exceptionally low rainfall). Obviously this definition does not take into account the effects of higher temperature. However, it is not self-evident that higher temperatures increase drought either, as pan evaporation seems to be decreasing worldwide.

According to the summary of the pan evaporation workshop at the AAS in November 2004, available on the AAS website:

‘The scientific and wider community should recognise that global atmospheric warming does not necessarily mean a more drying atmosphere or a drier land surface’ (p. 17).

The authors of the summary were Roger Gifford (CSIRO), Neville Nicholls (Monash, formerly BoM), Graham Farquhar (ANU) and Mike Roderick (ANU) – all of them highly regarded researchers.

  • sean

    People want to live in nature, and nature periodically burns. Clearing ground cover around homes is possible but may be an undue burden for the relatively small risk.

    So the real question is why did so many people stay in their homes until it was too late? Did they or the authorities miss judge the risks? Do you have to have regular fire drills like at work?

    We can except the risk of losing homes, provide the homes are known to be empty. It is not a climate problem, it’s humans not managing the evident risks associated with our land usage.

  • sean

    People want to live in nature, and nature periodically burns. Clearing ground cover around homes is possible but may be an undue burden for the relatively small risk.

    So the real question is why did so many people stay in their homes until it was too late? Did they or the authorities miss judge the risks? Do you have to have regular fire drills like at work?

    We can except the risk of losing homes, provide the homes are known to be empty. It is not a climate problem, it’s humans not managing the evident risks associated with our land usage.

  • Geoff Sherrington

    In the DECR “drought” is defined as

    “Box 1: Drought definitions
    Drought can be experienced and hence defined in different ways. There are essentially four types7:
    • Meteorological drought: a period of months to years when atmospheric conditions result in low
    rainfall. This can be exacerbated by high temperatures and high evaporation, low humidity and
    desiccating winds;
    • Agricultural drought: short-term dryness in the surface soil layers (root-zone) at a critical time in the
    growing season.The start and end may lag that of a meteorological drought, depending on the preceding
    soil moisture status;
    • Hydrological drought: prolonged moisture deficits that affect surface or subsurface water supply,
    thereby reducing streamflow, groundwater, dam and lake levels. This may persist long after a
    meteorological drought has ended;
    • Socio-economic drought: the effect of elements of the above droughts on supply and demand of
    economic goods and human well-being.”

    The authors then deal with 3 climate factors, in turn, temperature, rainfall and soil moisture.

    A different combination of these three factors, with different durations and weightings, is plausibly required to create a chosen form of drought from the 4 definitions above. However, there is little or no guidance by the authors as to which climate factors are causal to which definition of drought, particularly the last.

    A model to help farmers, dam builders and decision makers in general, would require the skill to predict when each of the 3 factors is going to combine in time and intensity to create one of the 4 types of drought. Even then, there might be additional factors, such as changes in the types of crops preferred by water users in different years and their different water needs.

    The DECR starting point is to assume that a rise in anthropogenic “greenhouse” gases is going to increase the probability of drought, but the report fails to show how such a rise can be used to predict the concurrence of the 3 nominated factors to produce one or more of the 4 classes of drought.

    David’s reviewers rebuke him for not doing precisely what the authors did not do.

    Moving to the context of the recent Victorian fires, authorities predicted strong, dry, north winds and a high temperature (which is usually a summer consequence). However, the fires might have been less harmful if heavy rain was falling. There might have been far fewer if there had been no lightning. Therefore, a disaster scenario would also need to predict an absence of rain but some lightning. Whether this is easy or hard in this analogy is not the point. The point is that exceptional events appear to require a number of parameters being exceptional at just the same time with just the right intensity. In the Victoria fires case, all needed to come together in the time frame of a day.

    Granted, droughts usually last longer than a bad fire day or even a bad fire season. However, the assumption that a congruence of several events critical to droughts will occur in a nominated time span simply cannot be predicted from a starting assumption that GHGs are heating the earth by a degree or so a century. It is hard enough to hindcast just one of the critical parameters with confidence.

    Model graphs with both historical and projected performance shown in the DECR often show actual observations spending an uncomfortable period outside the confidence limits, however calculated, and even displaying opposite trends. Because this is visually apparent without recourse to mathematical analysis, it is telling.

    The DECR more fully is:

    “An assessment of the impact of climate change
    on the nature and frequency of
    exceptional climatic events. Drought exceptional circumstances.
    K. Hennessya, R. Fawcettb, D. Kironoa, F. Mpelasokaa, D. Jonesb, J. Batholsa,
    P. Whettona, M. Stafford Smitha, M. Howdena, C. Mitchella,b and N. Plummerb
    (aCSIRO, bBureau of Meteorology)
    July 2008

  • Geoff Sherrington

    In the DECR “drought” is defined as

    “Box 1: Drought definitions
    Drought can be experienced and hence defined in different ways. There are essentially four types7:
    • Meteorological drought: a period of months to years when atmospheric conditions result in low
    rainfall. This can be exacerbated by high temperatures and high evaporation, low humidity and
    desiccating winds;
    • Agricultural drought: short-term dryness in the surface soil layers (root-zone) at a critical time in the
    growing season.The start and end may lag that of a meteorological drought, depending on the preceding
    soil moisture status;
    • Hydrological drought: prolonged moisture deficits that affect surface or subsurface water supply,
    thereby reducing streamflow, groundwater, dam and lake levels. This may persist long after a
    meteorological drought has ended;
    • Socio-economic drought: the effect of elements of the above droughts on supply and demand of
    economic goods and human well-being.”

    The authors then deal with 3 climate factors, in turn, temperature, rainfall and soil moisture.

    A different combination of these three factors, with different durations and weightings, is plausibly required to create a chosen form of drought from the 4 definitions above. However, there is little or no guidance by the authors as to which climate factors are causal to which definition of drought, particularly the last.

    A model to help farmers, dam builders and decision makers in general, would require the skill to predict when each of the 3 factors is going to combine in time and intensity to create one of the 4 types of drought. Even then, there might be additional factors, such as changes in the types of crops preferred by water users in different years and their different water needs.

    The DECR starting point is to assume that a rise in anthropogenic “greenhouse” gases is going to increase the probability of drought, but the report fails to show how such a rise can be used to predict the concurrence of the 3 nominated factors to produce one or more of the 4 classes of drought.

    David’s reviewers rebuke him for not doing precisely what the authors did not do.

    Moving to the context of the recent Victorian fires, authorities predicted strong, dry, north winds and a high temperature (which is usually a summer consequence). However, the fires might have been less harmful if heavy rain was falling. There might have been far fewer if there had been no lightning. Therefore, a disaster scenario would also need to predict an absence of rain but some lightning. Whether this is easy or hard in this analogy is not the point. The point is that exceptional events appear to require a number of parameters being exceptional at just the same time with just the right intensity. In the Victoria fires case, all needed to come together in the time frame of a day.

    Granted, droughts usually last longer than a bad fire day or even a bad fire season. However, the assumption that a congruence of several events critical to droughts will occur in a nominated time span simply cannot be predicted from a starting assumption that GHGs are heating the earth by a degree or so a century. It is hard enough to hindcast just one of the critical parameters with confidence.

    Model graphs with both historical and projected performance shown in the DECR often show actual observations spending an uncomfortable period outside the confidence limits, however calculated, and even displaying opposite trends. Because this is visually apparent without recourse to mathematical analysis, it is telling.

    The DECR more fully is:

    “An assessment of the impact of climate change
    on the nature and frequency of
    exceptional climatic events. Drought exceptional circumstances.
    K. Hennessya, R. Fawcettb, D. Kironoa, F. Mpelasokaa, D. Jonesb, J. Batholsa,
    P. Whettona, M. Stafford Smitha, M. Howdena, C. Mitchella,b and N. Plummerb
    (aCSIRO, bBureau of Meteorology)
    July 2008

  • Anonymous

    Thing is, if you can’t do the simple things rigorously, like %area exceptional low rainfall, any more complicated is just trying to look scientific, but waving your hands around. All models are compromises. I would like to see a basic statistical ‘peaks over threshold’ analysis, not assuming 5% as a threshold, cluster identification by autocorrelation, and decent regionalization before even looking at the models.

  • http://landshape.org/enm David Stockwell

    Thing is, if you can’t do the simple things rigorously, like %area exceptional low rainfall, any more complicated is just trying to look scientific, but waving your hands around. All models are compromises. I would like to see a basic statistical ‘peaks over threshold’ analysis, not assuming 5% as a threshold, cluster identification by autocorrelation, and decent regionalization before even looking at the models.

  • Pingback: bateria do laptopa hp compaq()

  • Pingback: massage couple paris 16()

  • Pingback: How to get a girlfriend()

  • Pingback: find here()

  • Pingback: link()

  • Pingback: witryna www()