Discriminating between aa and pahoehoe lava flows is important in terms of understanding cooling and eruption rates and thus in deciphering the evolution and the history of the volcanic pile.
The physical and magnetic properties in Hole 990A, together with the flow type differentiation in aa, pahoehoe, and transitional form flows, are shown in Figure 2. In general, the massive and homogeneous flow interiors show the highest density and velocity values, whereas the flow tops and bottoms are characterized by extremely low values. Despite the large intra-flow variations in physical properties, an overall change with depth is also obvious. There is a general uphole decrease in susceptibility and a downhole increase in magnetic intensity, resulting in generally higher magnetic intensities for the pahoehoe flows and higher susceptibilities for the aa flows.
Not only are the absolute values of the physical and magnetic properties interesting, but also their variations and ranges. On average, the pahoehoe flows show the lowest variations in density and velocity and the highest variations in magnetic intensity (Fig. 2, Fig. 3). The variations in density and velocity for the aa flows are primarily caused by the brecciated and altered flow tops. The magnetic intensity shows only some enhanced peaks, mainly in the top parts. The susceptibility is, in general, high for the aa flows, not only in the top or bottom parts, but also in the massive inner zones of the flows. Although the differences in magnetic properties behavior have several single or combined reasons (carrier of magnetization, concentration, grain size, grain shape, stress, and others), they might be potentially helpful in distinguishing flow types.
Both flow types show very low density and velocity values in their top parts as a result of the high vesicularity and scoriaceous or brecciated material, respectively, resulting in similar log or core measurement responses in the top part of the flows for both lava types. Although the different flow types are clearly visible in the flow tops (vesicular vs. brecciated zones), it is not possible to use the physical properties of the flow tops for lava type differentiation.
To demonstrate the impossibility of distinguishing between flow types by using the overall physical and magnetic properties, the average values are shown together with their statistical variation in Figure 3. The downhole plots on the left side show only values with densities >2.7 Mg/m3. This cut-off density value will serve as a discriminator as described later. The plots on the right side (considering all data) clearly show that all the average values are nearly identical in the 50% range.
A density value of 2.7 Mg/m3 is suggested as a discriminator between the altered and the unaltered flow sections based on the loss of ignition (LOI) values derived from X-ray fluorescence analysis (S. Planke et al., unpubl. data). The LOI is a parameter that is strongly controlled by alteration. All samples with a density >2.7 Mg/m3 have a LOI of <0.5%, pointing to a very low alteration state. Only the parts of the core with densities >2.7 Mg/m3 were used for further investigations. This means that only the massive and homogeneous interior parts of the lavas are considered. The advantage of this filtering is to make the data comparable, because the massive parts of the flows are either not influenced or less influenced by alteration, and thus should reflect the intrinsic properties of each unit. The cut-off density value of 2.7 Mg/m3 still provides sufficient data points from all flow units except for Unit 13 (see Fig. 2, Fig. 3).
Table 2 shows the average values and the standard deviations of the physical and magnetic properties of the flow types for densities higher than 2.7 Mg/m3. Discriminant values for the flow types are stippled, and it can be seen that aa flows are characterized by high velocity and density and a high susceptibility, whereas the pahoehoe flows are characterized by high magnetic intensities. But the average susceptibility difference (1694-1260 × 10-5 SI) is not as high as the average magnetic intensity difference (387-1346 × 10-3 A/m) (aa compared to pahoehoe). Note that the difference in average magnetic intensity between the flow types is about a factor of 3.5. The natural gamma-ray values are identical within one standard deviation (mean values of 8.1 to 8.6 counts).
The variation of physical and magnetic properties for the flow types is shown in Figure 4 in the same manner as in Figure 3, but here only samples with densities higher than 2.7 Mg/m3 are considered for the plots. The density and velocity values are clearly different, with the higher values for the aa flows (mean density of 2.91 Mg/m3), and the lower values for the pahoehoe flows (mean density of 2.86 Mg/m3). No big differences can be seen in the natural gamma-ray values. In contrast, the magnetic properties are well suited to distinguish between the flow types: pahoehoe flows show high magnetic intensities, whereas aa flows show generally higher susceptibilities. Although the whiskers in the plots (small horizontal bars describing the 95% level) at the right side of Figure 4 demonstrate that all flow types seem to have the same minimum values in magnetic properties, high maximum values are only recorded for pahoehoe or aa flows, respectively. There is only a little overlap in the boxes containing 50% of all samples. The same trend is reflected by the downhole core logs at the left side of Figure 4 for the density and susceptibility. Except for the natural gamma ray, all physical and magnetic properties show an intermediate behavior for the transitional flow type.
Table 3 is a summary of a suite of general diagnostic features for the various lava types. The combination of a low eruption rate and generally thin flow thicknesses makes it tempting to suggest a fast cooling rate for the pahoehoe flows, resulting in relatively small magnetic particles having a high magnetic intensity but a low susceptibility.
Until now, only the data from Hole 990A have been analyzed in detail with respect to changes in flow morphologies. Exploratory investigation of data from Holes 917A and 642E show similar clusters for the correlations between the physical properties and the flow types. Unfortunately, no magnetic data are available from Hole 642E, and only limited data exist from Hole 917A. For further investigations on subaerial volcanic rocks from large igneous provinces, it will be very important to get a complete as possible suite of downhole logs. The new downhole Geologic High-Resolution Magnetic Tool, with the measurement of induced and remanent magnetization, will be extremely useful for the interpretation of volcanic piles.