Managing Risk in Class G airspace

A note by the LAA

V 5 - 26 Feb 2012

Introduction

An earlier paper by the GAA revised on 25 Feb 2012 analysed all recorded mid-air collisions in class G airspace and extracted factual data about the nature of the risk. It concluded that the majority of collisions occurred when taking off or landing or when flying close to the airfield or flying site. It showed that predominantly, aeroplanes collide with aeroplanes and gliders collide with gliders or glider tugs. This paper uses the data from the study to consider how that risk is distributed in order to inform discussions with the CAA Department of Airspace Policy concerning the Future Airspace Strategy (FAS) as it relates to Class G airspace.

Aim

The aim of this paper is touse actual collision data todetermine how collision risk is distributed and to consider how and where that impacts on FAS policies for Class G airspace.

Validity of Data

The data on collisions used in the previous paper represented all events over 37 years which are a statistically significant data set. The data and its analysis have been considered against Category A airprox events by UKAB but perhaps because of the sheer volume of airprox events over the period, the comparison was inconclusive. Nonetheless the Director UKAB supports the earlier paper.

Whilst the particular location of an individual event does not point to a specific risk at that spot, the frequency of similar events in similar situations suggests the nature of a more general risk which may be used as the basis for consideration.

Distribution of Risk

The highest risk of collision risk for all GA occurs in and around busy GA airfields and flying sites (71%) with the remainder (29%) spread over the FIR broadly in line with likely traffic density.

The overwhelming risk to gliders is with other gliders and glider tugs in and around the launch site and in cross country thermals (97%) with only 3% of collisions occurring in cruise flight in the open FIR.

The risk to aeroplanes and helicopters is more evenly divided 55% in and around airfields and 45% in cruise flight.

Collisions Associated with an Airfield or Flying Site

Of the 178 aircraft involved in relevant collisions, 124 (70%) were taking off or landing or manoeuvring close to the airfield or flying site:

  • 39of those were powered aircraft or helicopters (31%)
  • 85of those were gliders or glider tugs (69%)

The majority of collisions involving aeroplanes, helicopters, gliders and all glider tugs are associated with the taking off and landing and manoeuvring flight in the vicinity of the airfield. These aircraft are often following a common pattern, using the same runway, joining or leaving a circuit or soaring in the same thermal or on the same ridge. In these circumstances local traffic density is very high and pilots generally know that other aircraft are nearby using common procedures but occasionally they fail to see and avoid a confliction. Collisions in and around GA airfields and flying sites constitute the main collision risk to GA (71%).

Collisions in the Open FIR

Of the 178 aircraft involved in relevant collisions, 40 (22%) were cruising away from the airfield or flying siteand 14 (8%) were in cross country thermals; a total of 54 (30%). Of these:

  • 33 were powered aircraft (61%)
  • 3 were gliderscruising (6%)
  • 14 were gliders in cross country thermals (26%)
  • 4 were military aircraft (7%)

The majority of mutual glider collisions in the open FIR occur in cross country thermals (14); these area special case and are considered separately in the conclusion. The focus for consideration of cruise events arethe 33 powered aircraft, 3 gliders and 4 military aircraft. This constitutes 22% of all relevant collisions. The distribution of these collisions generally mirrors probable traffic density across the FIR – none in Scotland, few in the North and more in the South (See Annex A and B). Whilst it might be expected that the risk of collision would be much higher at choke points, particularly in funnel areas andclose to major CTRs where traffic is heavier, this is not borne out by the actual distribution of events. Indeed, given the increase in traffic density in these areas, the risk of collision appears to be lower than elsewhere in the open FIR.

Collisions in the open FIR including those in choke points and densely utilised areas constitute only a minor part of collision risk to GA.

The Risk in DenselyUtilised Areas

There have been no collisions in high density areas around choke points and very few in the surrounding areas. It seems likely that aircraft flying close to the boundary of CAS or in a corridor bounded by CAS benefit from protection in that conflicts are very unlikely to come across the CAS boundary so the maximum angle between courses which could lead to a collision issignificantly reduced. Take for example the Luton/Heathrow gap. Because of the CAS, little traffic flies North or South and an aircraft flying any distance must track within about ±12˚ of East or West. The possible collision angle is thus limited to ±6˚off the nose.

Pilots flying such a track in this airspace will be aware that traffic density is high and will be alert in much the same way they would be following a TCAS warning. Moreover they will know the direction of threat and will be able to focus their lookout ahead. Taking the maximum angle between courses which is likely to lead to a collision in the open FIR to be about ±120˚ (collision angle ±60˚) suggests that the critical visual search area is reduced by some 90%. Similarly where an aircraft flies close to a CAS boundary as they pass a CTR, the pilot can focus attention to one side and ahead, which is a much easier task than looking out all around, reducing risk by at least 50%.

The 1991 paper by Andrews on unalerted visual acquisition supports this by proposing that the probability of seeing a threat reduces significantly as the search area is widened.

Hence it appears that collision risk is not increased when flying close to controlled or restricted airspace but could be reduced, even without the benefit of radar services, arguing against the proposition set out in the 21st Century Class G slide.

Conclusion

The demonstrated collision risk takes 2 forms; concentrated high risk over and around busy GA airfields and launch sites and distributed low risk through the FIR. Whilst the obviously increased traffic density around CAS and choke points appears to increase risk, as suggested in the work analysing 21st Century Class G airspace requirements, other factors must be already mitigating this as the actual number of collisions has been zero. It is likely that the boundaries of CAS act as a barrier narrowing the area that pilots must search to acquire conflicting traffic. Combined with heightened pilot awareness this reduces the probability of collision markedly.

The provision of an ATS service at an airfield does not seem to prevent collisions occurring and at many airfields and particularly launch sites, any form of control would be impracticable. Strong pilot skills in understanding the situation and the likely intentions of other aircraft combined with well practised lookout skills could help reduce this risk.

In the open FIR, the risk is much lower and relies on low traffic density and continual good lookout. GA traffic alert systems can warn of nearby transponding traffic but do not normally provide an accurate direction. The volume of airspace is large and the risk well distributed suggesting that to be effective,an ATS traffic service would need to be provided everywhere.

Technical solutions are not well adapted to deal with any of the GA issues discussed. The analysis 21st Century Class G airspace requirement identified some systems but they are not necessarily useful in any or all the circumstances discussed. Transponder carriage does not have an advantage without a universal ATS traffic service or universal carriage of an adequate traffic alert system which does not yet exist. Moreover the operations of a large part of GA are not amenable to interaction with an ATS service the ATS resources are unlikely to be affordable or practicable. ADSB(out) is presently beyond the means of GA because of the requirement for the GPS and the installation to be certified. ADSB(in) is available but without ADSB(out) or an ADSB TIS it cannot function. FLARM works for gliders and some GA aircraft but is not certified. Nevertheless it is currently making by far the greatest contribution to safety in this area and is the only system that can mitigate in dynamic situations including flight in cross country thermals.

Prepared by LAA

Revised 26 February 2012

Appendix 1 & 2: Distribution of collisions away from an ATZ or Flying Site (see separate document)