Looking into the CT200H mileage trends

What with work and family, sometimes I don't have as much time as I'd like to devote to authoring posts. And some of them take a significant amount of time. In this case, I've made a couple of posts regarding the the possibility of sequestering CO₂ from power plants in carbonate rocks, pavers, bricks, etc. I still owe my audience an analysis of this process in terms of energetics and economics. Those are taking some time.

In the mean time, I want to take a look at some of the data from my records of mileage in the Lexus CT200h that's my daily driver. To the left is a plot of the mileage at each fill-up since my acquisition of the car. It raises some questions.

Of course, the very low and very high numbers are related to the profile of driving done during the applicable tank. The lowest, for example, involved climbing into the mountains above Los Angeles for an outing with my son.

But I noted a trend, beginning at the end of the third quarter of 2012 and extending to the end of the first quarter of 2013, of declining mileage. I took the vehicle in for scheduled service in mid-March and told the service crew about the declining fuel economy. When they returned the car, they said they'd checked all applicable parameters in the fuel delivery system, the engine control unit, etc. and found no anomalies and adjusted nothing. But the mileage increased noticeably and is still appearing to be on that upward trend.

Now, to the best of my ability, I always drive in the same way (much to the frustration of my passengers - those who will still ride with me anyway - and the vehicles that share that road with me). And, to the extent possible, I try to always purchase gasoline from the same station. Assuming that it's not related to my route or driving techniques, what could explain it?

One possibility is "winter blend" vs. "summer blend" gasoline. In summer, particularly in California, refiners must use gasoline blends with lower vapor pressure to minimize vaporization due to warmer temperatures (and more driving). In winter, refiners add butane to blends because it's cheaper (thereby partially explaining winter's lower gas prices) and the higher vapor pressure of butane containing blends isn't as harmful due to the lower temperatures and lesser total vehicle miles driven. And butane has a lower specific energy content, thus possibly explaining my reduced fuel economy.

Is there such an annual "signal" in my fuel economy data? I've got 90 data points, and so ran a Fast Fourier Transform of the data. Such a process is used to transform data from the "time domain" (as in a time series) to the frequency domain (showing periodic components in the data). If the winter/summer blend switch, which happens annually, is a significant part of the fuel economy changes I've noted, my theory is that such periodicity should appear in the frequency domain. If you squint, you can even convince yourself that it's there - lower in winter and higher in summer.


Alas, there's no such peak apparent in the Fourier Transform. It's back to the drawing board. I can't imagine that the dealer fixed  or adjusted something and didn't charge me for it!