CRINEN : an updated analysis
using data from 2009 onwards
Introduction
This document attempts to review various statements, assertions and
predictions which have formed part of, or been specifically referenced by,
those documents used by OFEV for their “prise de position”, in the light of the
current statistics of 2009 and the first three months of 2010. This is
important in view of the fact that we are now virtually half way in the
2005/2015 period for which forecasting was based, but the situation is not at
all whet was anticipated when these documents were written. These documents are
:-
Specific data on aircraft timetables and actual movements, whether
scheduled airlines, charter flights and private jets (commercial or
non-commercial) has been taken from official airline schedules and from the
data on all aircraft movements as recorded and made available for general
analysis via the Geneva Airport Movement Enquiry (GAME) web-based procedure
available on the ARAG web site[4].
The underlying data for this procedure consists of a record of all detected
aircraft movements at
The particular periods for which these analyses may be performed are as
follows:-
As previously indicated, GAME is used to generate information for the
set of movements for a specified time period and according to specified
criteria. This information is then exported into spreadsheet data, thus
allowing simple, yet powerful, operations designed to select more specialised
information. All of this data is accessible on the ARAG web site via a single
reference document.
In every spreadsheet there may be multiple worksheets, of which the
first is one containing all the information generated by GAME according to the
specified criteria. All or part of this first worksheet may then be copied into
other worksheets, and then sorted in different ways to show different
information.
In some cases, information on the scheduled departure or arrival times
may be taken from the published timetables and copied into the spreadsheets,
thus allowing a comparison between scheduled and actual times of aircraft
movements. It is recognised that in a few cases the actual schedule may have
been modified from the original schedule. However, such modifications are quite
rare and rarely of any large amount, and do not affect much statistical data. There
may also be new scheduled movements on a regular basis: one particular recent
example is the twelve daily flights (six in each direction) between
Executive summary
This report studies the various reports which have been referenced or
created as a consequence of the CRINEN decision, and which have been used to
make a case for the refusal of every curfew proposal. It is the opinion of ARAG
that these reports were deficient in many ways, both in the assumptions made on
traffic growth and in the methodology used to forecast this growth in terms of
nuisances and noise contours.
The report of the
independent expert M. Avi Gil
Although this report was not included in the reference documents listed
by OFEV, it was cited frequently in the AIG Observations document. As well as
being not particularly acceptable to make selective extracts from a document
which has not been made widely, it should be noted that the report predates the
CRINEN process and contains many assertions about traffic patterns which are no
longer accurate. The following paragraphs summarise the main assertions,
specifically invoked in the observations of the AIG but which are no longer
true.
The CRINEN
judgement
In so far as what might be considered as the least demanding curfew
proposal on late evening flights, namely that changing the airport closing time
from
This chapter examines the effects of
extending the existing curfew to 2300, assuming that the half hour “grace
period” where late operations are permitted with airport approval is
maintained.
This difference may have a very big effect, especially during the summer
period when arrivals scheduled for the period from 2300 to 2330 are quite
prevalent, most often from holiday centres in southern regions of Europe
(France, Spain, Italy, Portugal): in the winter period virtually the only such
arrivals are the daily ones from Zurich, arriving in Geneva at around 2315 and
having an exceptionally good record of punctuality (only 17 landings after 2330
in the whole of the 30 week summer period, arriving on average 2 minutes early
and with a standard deviation of only 10 minutes).
Chapter 9 (curfew from 22h00 instead of
It is an obvious thought that this dichotomy could lead to the
suggestion that one could start with a grace period of one hour, decreasing at
a later date to 30 minutes when airlines have had time to rework their
schedules and improve their punctuality record.
The EMPA evaluation
report
As stated in its introduction, the EMPA report, with its noise
calculations and analyses, is very technical and not intended for the wider
public. The guidelines for the various calculations are the various options for
different curfews, as laid down in the CRINEN judgement. The input data is
provided by the AIG airport statistics data for 2005, together with forecasts
for the number of movements and aircraft fleet in 2015, provided by SH&E.
The methodology for the calculations is then defined in a precise manner,
allowing a large number of detailed tables and contour plots to be generated by
a sophisticated computer program.
So can we assume that all the results are correct? Unfortunately not!
Although ordinary people tend to be convinced by such detailed and nice-looking
results, any computer professional will remember the acronym GIGO (Garbage In,
Garbage Out). In other words, if the input and methodology is not correct then
the results will of necessity also be incorrect.
From the start, there has been much criticism of the forecasts provided
to EMPA by SH&I. The most striking criticism was that of assuming growth
patterns which were completely incompatible with the growth statistics at that
time. These growth patterns have in the meantime been considerably affected by
the economic recession, whilst the future growth is still difficult to
forecast, being dependent on many factors (economic growth in
However, there is a more fundamental flaw, which lies in the
methodology. This methodology is one whereby, if flights actually taking place
during a particular late night hour night are to be subject to a curfew, then
they will either be cancelled or rescheduled. If they are rescheduled then they
will all move into the previous hour, WITHOUT any of the movements of the
previous hour themselves moving back. Although highly convenient for the
computer calculations, this is completely at odds with reality, as is obvious
if we remember that even for the companies with the best punctuality
statistics, there is quite a spread of actual movement times around the
scheduled times. Even the most punctual of flights, such as the final Swiss
flight each day from
In connection with this handling of these “delayed” movements, it is
also now the case that the percentage of delayed movements relative to on-time
movements has completely changed from what was assumed in the SH&E
forecasts. In EMPA Table 3-1 the statistics say that in 2005 nearly 40% of all
flights between
For these reasons, the results of the EMPA computer calculations should
be treated with a degree of healthy scepticism. Instead, the approach should be
to see how to frame an curfew rules so as at least to prevent the current
situation from worsening.
The SH&E
evaluation report
The Executive Summary at the start of this report is clearly designed to
create a nightmare scenario, whereby the slightest curfew restriction would
result in catastrophic consequences: millions of Swiss francs lost to the
economy as easyJet moves all its operations elsewhere. This, again, has a
well-known acronym: the FUD (Fear, Uncertainty and Doubt) factor. However, it
has virtually no correlation between this scenario and the actual situation,
nor with what would happen in reality with the introduction of a reasonable
curfew designed to reduce noise nuisance by obliging airline companies to adopt
improved scheduling and punctuality practices. In particular, there is no
reason why easyJet should want to reduce their number of aircraft in
In chapter 4 (recent trends) the SH&E report admits that night
flights, especially in the second night hour, have been growing alarmingly
rapidly, The report then mentions the new measures foreseen by the AIG to
address this growth, in particular a schedule of surcharges ranging from CHF 50
to CHF 9,000 per takeoff for jet aircraft from 2200 to 0600. Since this was
introduced in April 2008 it has been apparent that it has had no real effect,
and possibly even the reverse effect (according to the law of unforeseen
consequences!). This can be seen most easily by studying the history of the
flight which makes by far the largest monetary contribution, namely the flight
from Geneva to Mauritius by the extremely noisy Noise class III Airbus
A340-300. In 2008, after the introduction of the takeoff surcharge, ARAG
calculations show than each takeoff cost an extra 31 CHF, in 2009 an extra 655
CHF and in the first three months of 2010 an extra 1470 CHF. Even this latest
figure, though it may sound a lot, is only about a quarter of the total fees
paid for each flight (which include landing charges and ATC charges). Since
this Airbus carries around 300 passengers, each paying on average over 1000
CHF, it represents only 0.5% of the revenue for each flight. Thus, one may
imagine that the airline feels that it is purchasing the right to have late
flights with a maximum of passengers, whilst the AIG is happy to see the
considerable augmentation of the money available for soundproofing the houses
of people affected by the noise! Two winners (AIG and Air
When this SH&E document specifies numbers and destinations for
scheduled flights it indicates as source the Official Airline Guide, March
2007. It is assumed that this then refers to the schedules for that summer,
i.e. from 25 March to 27 October.
Exhibit 5.7 states that in March 2007 there were arrivals from 17
different places (7 EasyJet, 3 Swiss, two Baboo and 5
others), of which three arrived back at or after 2300. Although the arrival
times quoted do not seem to match those in the timetables document for Summer
2007 which was available on the AIG web site, this is probably reasonably
accurate.
Exhibit 5.8 indicates 12 morning departures prior to 0700: five of EasyJet
and 7 of classical airlines. In relation to this exhibit, it is then stated in
section 5.2.4 that
Prohibiting departures before 0700
at
As is blindingly obvious, EasyJet is not a hub airline, but a point to
point airline. As such, passengers wishing to transit via a hub, with automatic
baggage forwarding, will not use EasyJet flights. This leaves flights of
classical airlines to
The same section then goes on to state that
Restricting arrivals by an hour in
the morning or departures by an hour in the morning would effectively limit
airlines to three daily rotations on most routes.
Again, this
has little truth. The only airline with a base in
In section 5.2.5 (Express Cargo Operations) the SH&E report claims thatexpress all-cargo carriers including DHL, TNT and UPS
would be affected by a morning curfew because
Their customer delivery schedules in
the
Not only is this statement made without any supporting evidence, or
examples, but in the modern world much time-sensitive information now consists
of documents transmitted instantaneously via the Internet.
In section 5.4 (Existing Measures to Reduce Noise Impacts) there are
three particular statements which are no longer true.
The [noise] monitoring system
[MIABA] indicates that aircraft noise levels have been declining despite an
increase in aircraft movements due to the shift to quieter aircraft.
In actual fact, since 2004 the noise registered on the microphone at the
airport has an overall increasing trend[7].
The downward trend in annual noise
surcharge revenues confirms the finding that airport noise levels are
decreasing.
Statistics furnished to ARAG by the AIG show that the revenue from the
general noise surtax increased monotonically from 2005 to 2008. It did fall in
2009, but that was largely counterbalanced by a large increase in the takeoff
surcharge (mainly that Air Mauritius aircraft!).
commercial flight from 2200 to 2400
are restricted to takeoffs that meet noise limits
To be precise, what is restricted is the scheduling of takeoffs of these
noisy aircraft after 2200 (but there are only very rarely any takeoffs
scheduled after 2200 anyway). However, any aircraft for which the takeoff is
planned prior to 2200 (such as the very noisy cargo Boeing 747 of Jade
International) can actually be allowed to take off until 0030!
The predictions of SH&E for the number of aircraft of scheduled
airlines which park overnight at Geneva is stated to be likely to grow from 28
(in March 2005) to between 33 and 38 in 2015. ARAG calculations for March 2009
and March 2010 indicate that the current number is stationary at 27: this
decrease does not appear to have had an enormous effect.
When the SH&E report starts to forecast the impact of any of the
various curfew extensions, it reiterates the incorrect assumptions on the growth
predictions and on the effect of rescheduling. This is best illustrated in
their prediction of the effect of a curfew from 2300, with a grace period of 30
minutes: Chapter 7 in their report.
Exhibit 7.1 shows a prediction for arrivals in 2015 if there is no
intervening change in curfew. This expects the arrivals in the periods
2100-2159 and 2200-2259 to be approximately the same (4,985 and 5,039
respectively), in the 2300-2359 period to be 1,897 and after
The calculations of the effect of a curfew moved back from midnight to 2300 assume that all flights
arriving in the 2300-2400 hour will then almost all move back into the
2200-2300 hour WITHOUT any corresponding moving back of any of the arrivals
already in the 2200-2300 hour back into the 2100-2200 hour. This is absolute
nonsense, flying in the face of reality. As of 2009, ARAG statistics show that almost
half of the arrivals in 2300-2400 are delayed returns of easyJet
In other words, the predictions of changes occurring in such an
unrealistic manner are completely erroneous, and thus render any contours
calculated from them as highly unreliable. In particular, it would be
completely false to accept that there would be a huge surge in the number of
arrivals in the 2200-2259 hour. As is already the case, aircraft arrivals would
gradually tail off towards curfew time.
The same fallacies apply to departures, including the figure of 67
departures after
However, extending the curfew to
2300 would also have a negative impact on noise levels from 2200 to 2300 as
airlines shifted operations to the earlier time period.
This equally invalidates any projection of exhibit 7.6 (Contours for
2005 First Night Hour Traffic including Displaced Second Hour Flights).
The same criticisms of invalidity apply to all of the subsequent
chapters (8, 9 and 10) on extensions of the night curfew. It is especially
flagrant that the economic costs are always dominated by a supposed loss of
earnings because of an invented figure for a reduction in the number of
aircraft stationed at
Chapter 11 of the SH&E report considers the impact of extending the
morning curfew from 0600 to 0700, trying desperately to justify a conclusion
that it would result in
a reduction in six based aircraft
[in 2015] compared to the status quo forecast.
The chapter presents two forecasts of the situation in 2015, in terms of
arrivals and departures in various time periods. Again, the given forecast
numbers are not taken from the EMPA report, which only gives forecasts for the
whole daytime period
The apparent basis for this is a reported feedback from airline
interviews that 10% of flights in the morning 0600 to 0700 period would be
cancelled. Somehow, from this feedback, it is then forecast that there will be
a reduction of 130 flights per day, resulting in the forecast of six less based
aircraft (which then, of course, is multiplied by the assumed “lost revenue” of
each based aircraft to calculate the unbelievable figure of over 17 million CHF
(Exhibit 11.8). However, nowhere is it stated how this figure of 130
cancellations has been decided upon.
A little logical reflection shows the fallacies of the above reasoning.
Firstly, there is no justification for the quoted 10% figure (for which it
would be quite natural for the airline companies to predict on the high side).
Next, the number of based aircraft would only be reduced if departures
scheduled prior to 0700 were to be cancelled (aircraft arriving in this time
frame always leave again the same day). In reality, however, there are few such
regular daily (all weekdays, possibly weekends also) departures. These comprise
one Swiss flight to
Next, we have to work out how, by cancelling 10% of these scheduled
departures, we can simultaneously arrive at a reduction of 260 movements per
year (130 arrivals and 130 departures) and a reduction of 6 based aircraft. 130
arrivals and departures is less than half of one based aircraft, whilst the
loss from 6 based aircraft would be over 2000 movements per year! Completely
incompatible.
Finally, to inject a dose of reality, it is difficult to imagine either that
the classical aircraft departures just before 0700 could not easily be moved a
few minutes later, nor that easyJet would have unachievable difficulties in
rescheduling rotations, perhaps flying more to nearer destinations, to still
have most of their aircraft operating four rotations per day.
Similar criticisms apply to the next chapter, which deals with the
impact of extending the curfew to 0800 on weekends and public holidays. In the
same unexplained manner, it is forecast that such a curfew, applied to the
forecast situation in 2015, would result in 215 flights being cancelled.
Finally, one comments occurring at the end of chapter 15 is worthy of
comment.
An airline official was quoted as saying that
Restricting the operating hours in
Taking the month of March, reported to be the busiest month of the year,
as the basis for an enquiry, a graphical plot of the average number of
movements per hour shows about 35 per hour between 1000 and 1900. Even adding
the two averages for 0600-0800 (8.4 + 22.6)
or for 2200-2400 (13.9 + 3.4) stays well below the capacity of movements
on the runway.
The Observations of
the AIG
As might be expected, this document emphasises those remarks made in the
report of M. Gil, the EMPA report and the SH&E report, which argue against
any extended curfew. Having demonstrated that those remarks hardly ever apply
to the situation as it is now, only any additional comments in this AIG
Observations document, from section 63 onwards, will be specifically addressed.
In section 63 it is suggested
that one should wait to see what has been the effect of the various measures
adopted by the AIG in the Spring of 2008. It is now the Spring of 2010, and the
statistics over the past two years suggest that these measures have had little
or no real impact, and even sometimes arguably a negative impact.
In section 64 the AIG insists that any of the proposed measures would
cause a significant diminution of the numbers of aircraft based in
The second paragraph of section 64 is an even higher level of hyperbole
and misrepresentation of the real facts. In this paragraph a reference is made
to most dramatic consequences, involving a drastic limitation in the number of
European destinations to which it would be impossible for a business person to
go and come back in the same day. As ARAG has demonstrated, neither of the two
simplest curfew possibilities need have anything other than an extremely minor
effect (15 to 30 minutes) on just some destinations served by easyJet.
The final paragraph of section continues the hyperbole, this time with
regard to transit journeys via hub airports. As ARAG has demonstrated, for no
hub other than
Section 65 continues the hyperbole and exaggeration of the facts by
referring to a supposed massive lowering in quality of aerial liaisons
available at
In paragraphs 68 and 69, the document refers to various projects under
study, emphasising that they are merely projects whose feasibility and utility
need to be evaluated. The judgement of ARAG on these possibilities is that they
are ill-defined, with no specified timescale and probably little effect other
than that of generating more income for the AIG.
The first project states that all requests for commercial movements
between 2200 and 0600 would be evaluated by the AIG according to a set of
criteria. There is no attempt to indicate how the relation of the requested
movement to the various criteria would be done.The
suggestion that this evaluation process might be similar to the current process
for derogations after 2400 is not meaningful, since the details of this process
are not known to ARAG.
The second project is that of extending the ban on movements between
2200 and 0600, currently applicable to non-commercial movements, to the
totality of general aviation. However, upon being asked for precisions on this,
the AIG has confirmed that there is no legal definition of general aviation.
The project is thus currently meaningless.
The suggestion is made that, when meteorological conditions permit the
use of either of the two runways 23 and 05, the choice of runway 23 is in some
way more favourable than that of runway 23. In a subsequent meeting between the
AIG and ARAG it has been admitted that there is no foundation for this
statement.
Finally, the AIG asserts that an updating of the noise classification of
aircraft will result in the reinforcement of the dissuasive effect of the
standard noise surtax and the additional surtax applied to late night takeoffs.
ARAG doubts that the validity of this statement, made without any real idea of
which of the aircraft types most frequently using
Conclusions
The great majority of the arguments made against the introduction of any
curfew whatsoever are without foundation in the situation of today.
[1] Evaluation
de limitations supplémentaires des vols nocturnes sur l’Aéroport de Genève,
Calculs et analyses du bruit, EMPA Nr. 445'300, 8 juin 2007
[2] Evaluation of Additional Night
Operating Restrictions at
Prepared
for Aéroport International de Genève
by SH&E, Inc., May 2007
[3] Observations
pour Aéroport International de Genève, Me Olivier Jornot,
5 octobre 2007, OFAC reference 3561/3/31/31-06
[4]
http://www.aragge.ch/game.html
[5]
http://www.aragge.ch/fr/nuisances/bruit/bruit_par_mois.htm
[6]
http://www.aragge.ch/fr/nuisances/bruit/Bruit%20dBA%202004-2009_files/image001.gif
[7] http://www.aragge.ch/fr/nuisances/bruit/Bruit%20dBA%202004-2009_files/image001.gif