Airspace & Risk

At the IFATCA annual conference held in Istanbul Turkey in April 2007 IFATCA amended it’s policy regarding ATC at aerodromes; the policy now reads:

Air Traffic Control service shall be provided at aerodromes that:
have published IFR approach, departure or holding procedures, and where control is required for the safety of air traffic.
for VFR operations, where required to ensure that appropriate safety levels are met.

I find the policy to be incredibly non descriptive about the airspace classifications, levels of ATC service and that there essentially are no ‘trigger points’ to which allude to levels of safety.

In Australia we are having a ongoing debate about airspace classifications in general then whether a UNICOMs/CAGRO, ATC tower services ranging from GAAP (similar to the VFR TWR model), to Class D (regional towers), to Class C tower services (provided at capital cities). See the recent debates about AVALON.

In Australia we have 3 prominent examples of UNICOM/CAGRO services, Ayres Rock, Broome and Jandakot Tower (outside tower hours). These have all proved to be successful in increasing safety, based on the lack of available statistics to prove otherwise; but are they "safe enough"? Is the Unicom/Cagro as safe and as cost effective as a GAAP or regional Class D tower? Is there another way? Is it safer, see below about vigilance.

Australia has much Class E airspace ‘rolled out’ progressively since about 1998; the majority of this Class E is outside radar or ADS-B surveillance.

The benefit of Class E for IFRs is like Class D/C/A they are separated from other IFRs; however there is no requirement for VFRs to be known to the system when flying in Class E. It is up to the pilot of a VFR to comply with the Visual Flight Rules and remain in VMC, and adjust course as necessary to avoid IFR operations; this may include avoiding IFR routes; this doesn't always happen, but to what consequence?

How does a VFR avoid an IFR route; obviously if operating to a particular aerodrome there is no opportunity to avoid the destination; if however flying via GPS (or VFR direct) point A to point B, cross referencing to the IFR routes is next to impossible; similarly many VFR points ‘closely replicate’ IFR tracking points; thus encourage the VFRs to be at the same place as IFRs.

Thinking back to the IFATCA policy above, “ensure that appropriate safety levels” and “is required for the safety of air traffic” whilst both of these are great motherhood statements they don’t describe the when, how’s and why’s about safety.

Safety is about managing risk. Risk is present in everything we do; minimising the effect of that risk essentially becomes managing safety. In simple terms lowering the speed limit on a road is reducing risk, but only if you have compliance to the limit. A without "cause" reduction is more likely to increase non-compliance thus not altering the risk; enforced compliance helps.

Statistically the most dangerous phase of flight is landing and departure; this is because this is the phase of flight (logically) where contact with the ground is present; improper contact causes injury and damage.

In terms of airspace risk the most dangerous area of sky is close to the runway, because that is where the concentration of traffic is likely to occur; i.e. the planes want to use the runway; and often it’s only one runway.

That does not mean no risk in other areas of sky.

The risk of collision in the middle of Australia at 5500 feet (in class G) is “vanishingly small”; that doesn’t mean no risk just really really low; it assumes that there is compliance; i.e. VFR aeroplanes will ‘see and avoid’ other aircraft and that due to the density and size of the airspace available compared to the ‘footprint’ of the aeroplanes involved it is statistically improbable to think that two random events (trajectories of aeroplanes) could intersect at the collision point together. This is often referred to as "BIG SKY THEORY".

Things that conspire against BIG SKY THEORY; GPS accuracy, height locks in aeroplanes, Restricted Area avoidance, terrain avoidance, bad weather, towns, aerodromes, topographical features, ERC routes. All these things add up to increase the risk; yet at 5500 feet in the ‘middle of nowhere’ the risk of hitting another aircraft is still “vanishingly small”; this is because compliance still applies, aircraft should be operated by ‘trained’ pilots; who have received training in VFR and/or IFR flying, there are rules designed to segregate aircraft flying on the same track (or hemispheric track); crossing tracks makes it trickier; TCAS, visual reference, radio calls all make flying in Class G safer; avoiding IFR routes or way-points (if VFR) etc.

Which brings me back to when is ATC ‘required’; Air Traffic Control improves safety of flight, because conflictions should be avoided (ground and aircraft conflictions); but that is dependant on airspace classification Class E doesn’t provide a VFR with any form of protection above that of Class G; in fact it could be argued that Class G is safer as IFRs (probably the most likely confliction) are more vigilant in Class G without ATC ‘protection’ than in Class E with ATC ‘protection”, they conflict with “unknown unknowns”.

Surveillance greatly mitigates against that risk, because the VFRs can be seen “known unknowns” and thus the IFRs alerted, the VFR may hear the IFR getting the information and thus themselves become more aware; the controller is working in less difficult conditions when surveillance is available and is less concentrating on “proving the standards” being used between IFRs and providing a service to them.

Class D and Class C airspace provide extra protection; as the only elements able to use the airspace are known to the system (air traffic controller). This doesn’t mean no risk, just less risk; “known knowns”; but that assumes compliance.

Aircraft often ‘penetrate’ restricted or controlled airspace without clearance; this is usually due to pilot error, but sometimes it is deliberate non-compliance. The only place where these ‘penetrations’ are recorded and reported is where surveillance in available; or where a controller or pilot visually sees the non-compliance and reports it. There are approximately 1000 penetrations of control or restricted areas each year; many of these lead to conflictions with other aircraft, where the ATC or pilot may not even be aware of the other aircraft; or it's simply avoiding action only.

It’s not as simple in saying then make everything Class D or Class C or higher; there is a trade-off in capacity when increasing safety; this also comes at great cost. Equipment, controllers etc. all add up to increase the costs associated with increasing risk mitigation. Capacity is often reduced when increasing service as 'standards must be used'.

Remember that Class G in the “middle of nowhere” the risk is “vanishingly small”; but that doesn’t mean no risk.

It's my managers money right?

Yesterday at work I was extremely discouraged by two events in particular.

Number 1:

We run a roster of 5 staff in the morning, with staggered start times:

The shifts run 0500-1100, 0600-1300, 0630-1430 X 2 and 0700-1500; these shifts are replaced by 4 afternoon shifts, a 1200-2000, 1400-2100, 1400-2200 and 1500-2300.

So our staffing requirement in the afternoon is reduced by 1. This effectively is amended when bad wx is forecast and we get in someone on Overtime to cover the peak from 1600-2000 (4 hours minimum OT payment).

The reason for the disparity in staffing levels is traffic demand; the morning peak is always busier than the afternoon peak. I guess businesses always start their day early; but finish meetings etc. at different times; at least this is how is seems.

So what was the issue: The Supervisor decided that the 0500-1100 did not need replacing when the person called in sick. So this would put us below minimum staffing; which we often do due to not having an option. Why in the case? Lack of staff to fill the vacant shift, no, simply management discretion.

So what mitigation is offered to overcome the 'stupid management decision'? Traffic Management strategies; so for the cost of about $500 (gross, the cost of the call out) We force the traffic to be delayed and avoid, enter the airspace in a particular manner so as to not overload the 4 controllers working the 4 sectors without breaks; or the 3 sectors when there should be four, so that some toilet breaks etc can occur; did someone say breakfast?

I think that the cost to industry in this scenario would exceed $5,000 (much higher if holding delays resulted), due to all the delays created in managing the traffic; for what reason, so the company didn't spend a lousy $500. Was this in the name of safety? Hell no, management stooge economics 101.

Number 2:

Almost 7 months ago a controller was "promised in writing" a pay rise on achieving their next set of ratings; this happened 5 months ago and guess what, no pay rise.

The individual in question was given a BS excuse initially about high sick-leave, 7 days in the year all consecutive, when an immediate relative was effectively on his death bed.

The individual has been 'strung along' with multiple meetings, e-mails, phone calls about a 'decision pending'; the decision now rests with a manager at head office; who apparently has being "mulling over" the decision for 2 months now; in my experience it's most unlike that individual to not make a snap call.

All the while the controller involved has been 'required' to perform extra non console duties in all this time to prove his 'worth' whilst the decision is pending.

Makes me wonder if the manager has being BSsing him the whole time; or whether the administration really is that incompetent and can't approve a pay rise ($5K or so per annum) without the say-so of someone 1 wrung below the CEO; hell by the time it is approved he'll have achieved the annual increment anyway.

It does actually come out of the purse or wallet of the relevant manager right, that's why that are so reluctant to pay; well perhaps not; it's a disgrace.