Not so hot Marble Bar

The Pilbara is a section of mid western western Australia, and to start off with we’ll look at the northern section of it. It’s very dry there. I once went to marble bar, and there’s not a lot going on. In fact, the only thing to do there was drink beer, which is what everyone naturally does. It’s very hot, all year around, and the amount of rain is very very low. So any increase would be a natural godsend. But lets look at the temperature first.

We’ve got stations with adequate data from 3 different places. Marble Bar, Port Headland Airport and Roebourne. Marble Bar is located a little inland whilst the latter two are on the coast.

Our analysis of maximum temperature suggest no significant increase or decrease in temperature (t = 1.1, p=0.27), however there was a significant increase in minimum temperatures (t = 4.1, p <>

Unfortunately for this area time related temperature data is somewhat sparse, and only good reliable data is found at 9am and 3pm, which were the original 2 standard temperature taking times for the BOM. Out analysis shows no significant temperature difference at 9am (t = 1.51, p = 0.13) and also no significant difference in temperature at 3pm (t = -0.7, p 0.48).

However what is interesting is the difference in temperatures that Marble Bar has compared to its two ocean dwelling counterparts. Marble Bar actually recorded a significant decrease in maximum temperatures going back as far as 1910 (t = -2.56, p = 0.01). This being at a rate of 0.72 degrees Celsius degrease per century. In fact Marble Bar recorded its lowest average maximum temperature on record last year and in the last 8 years has recorded 4 of it’s lowest 8 recorded maximum temperatures. This place is cooling down (the residents are thankful). Whereas both Port Headland airport since 1949 (t = 2.8, p < 0.05) and Roebourne airport (t = 5.8, , p < 0.05) showed significant increases in temperature.

This is quite strange considering that, aside from the ocean dwelling difference, they are located in near proximity to each other. What makes this more compelling is that that the temperatures per year are highly correlated with each other from station to station. For example at 9am correlations vary from 0.65 to 0.79 (p <>

Well, lets point out that those correlations given above were only from 1957, since this is where specific time recorded data was first recorded with all three stations. If you look closely at the maximum temperature for Marble Bar pre this time you will note a lot of large positive anomalies. If Port Headlands maximum temperature data had gone back that far could it have reached similar large positive maximum anomalies, especially considering that the two are largely and significantly correlated? Quite possibly.

Two things are for sure however, is that it is very difficult to compare data from stations where there are differences in the amount of sampled years, and I believe we need more data before we can make concrete evaluations.

Oh and has rainfall increased in the northern Pilbara? For Marble Bar fortunately yes! But Port Headland and Roebourne, unfortunately no.