GEOMORPHOMETRIC DESCRIPTIVE PARAMETERS OF SCORIA … · GEOMORPHOMETRIC DESCRIPTIVE PARAMETERS OF SCORIA CONES FROM DIFFERENT DTMS: A RESOLUTION INVARIANCE STUDY Fanni Vörös, Benjamin

GEOMORPHOMETRIC DESCRIPTIVE PARAMETERS OF SCORIA CONES FROM DIFFERENT DTMS: A RESOLUTION

INVARIANCE STUDY

Fanni Vörös, Benjamin Van Wyk de Vries, Balázs Székely

Fanni Vörös BSc Department of Cartography and Geoinformatics, ELTE Eötvös University, Budapest, Hungary address for contact: H-1518 Budapest, Pf. 32., Hungary e-mail: [email protected] Benjamin Van Wyk de Vries PhD Université Clermont Auvergne, CNRS, IRD, OPGC, Laboratoire Magmas et Volcans, Clermont-Ferrand, France address for contact: Campus Universitaire des Cézeaux, 6 Avenue Blaise Pascal, TSA 60026 - CS 60026, 63178 AUBIERE Cedex, France e-mail: [email protected] Balázs Székely PhD Department of Geophysics and Space Science, ELTE Eötvös University, Budapest, Hungary address for contact: H-1117 Budapest, Pázmány Péter sétány 1/C., Hungary e-mail: [email protected]

Abstract Digital terrain models (DTMs) are used in geomorphometry. Coupled with field validation, their use makes the studies more comprehensive, and opens the way towards automation. DTMs are influenced by the acquisition technology, accuracy, and resolution and questions arise about whether results depend on the type of DTMs. To study this, we have chosen a simple landform: the scoria cone, a typically symmetrical, monogenetic volcanic edifice. Study areas are San Francisco Volcanic Field (DTM resolutions 30 and 10 m) in Arizona (USA) and Chaîne des Puys (resolutions 90 and 0.5 m) in the Auvergne (France). Descriptive parameters were calculated and comparison has been carried out via grouping by study area, by age, and the Mann–Whitney test has been applied. The by-area comparison shows statistically significant differences for some groups, though overlapping occurs. In addition, such behaviour can be observed in the same field grouped by the DTM resolution, but they are statistically insignificant.

Keywords: San Francisco Volcanic Field, Chaîne des Puys, DTM, statistical comparison, Mann–Whitney test

PRESENTING THE AREAS

Previous research on morphometry of volcanic cones

The first important work on scoria cone morphometry is COLTON’s (1967) study. In his pioneering work, the basalt cones and lava flows were classified into different classes based on their degree of degradation and mounding. PORTER (1972) – as a result of researche in Hawaii—defined volcanic morphometric parameter relationships, which were later used for several comparative analyses in other areas as well. The comparative studies of WOOD (1980a, b) are milestones in volcanic morphometry. He stated that the cone’s approximate age can be concluded from some simple morphometric studies. He studied the destruction processes of cones that were due to erosion or other mass movements. After studying volcanic fields such as San Francisco Volcanic Field (SFVF), Nevada, Oregon, Italy, Réunion, he claimed that cone height (Hco), height and diameter (Wco) ratio and slope angle (α) decreased over time, crater diameter (Wcr) and cone diameter (Wco) ratio (Figure 1) does not change with decay.

Proceedings, 7th International Conference on Cartography and GIS, 18-23 June 2018, Sozopol, Bulgaria ISSN: 1314-0604, Eds: Bandrova T., Konečný M.

603

Figure 1. Relationships between volcano parameters [PORTER (1972)]

In a detailed study of the SFVF more than 650 volcanic cones were distinguished by TANAKA et al. (1986). The paleomagnetism of the cones was studied, K/Ar ages were determined, and the investigation of the rock and stratigraphic studies was carried out. During these, the migration of volcanic activity was recognized. A further step in the geomorphometric research have been made by HOOPER and SHERIDAN (1998) who modelled surface erosion processes on ideal scoria cones. According to them, the relative determination of age with comparative morphology is based on the examination of decreasing morphological parameters over time. From the 2000’s on, digital terrain model (DTM) became the basis for morphometric research. With the help of the DTMs FAVALLI et al. (2009) concluded that the height of scoria cones on volcanic slopes should be calculated in different way.

Scoria cones

Scoria cones are the simplest volcanic structures (WOOD 1980a, b), made of loose pyroclastic segments. They are monogenetic, which means they erupt once, and during the eruption the whole volcanic edifice is created. They are about from 30 to 300 meters high and a few hundred meters wide, so relatively small. Their slope angle is 20-30°, which decreases with the aging of the cone. The youngest—ideal—cones have approximately regular circular symmetric shapes (Figure 2). Any kind of asymmetry can be caused by erosion, formation of a volcanic origin (e.g., lava flow), or if the cone is not circumscribed in the eruption, it is created by some breaks along the line. In this case, the cone itself is intact but elongated, elliptical.

Figure 2. Chain of scoria cones is the Chaîne des Puys area

San Francisco Volcanic Field

The San Francisco Volcanic Field (SFVF) is located in North America, Arizona, near Flagstaff, in the south of the Colorado Plateau. In the area of approximately 4700 km2, there are about 600 scoria cones, lava domes, lava flows and extensive scoria and ash sediments (PRIEST et al., 2001). The dominant rock of the SFVF is the alkaline basalt. Because of the low content of silicon dioxide (<54%), the lava is rich in fluids and spreads over a large area due to its


604

low viscosity. This type of volcanic activity is characterized by explosive outbursts. In addition to the basalt, there are also andezitic forms: the central form of the area, the San Francisco volcano’s dominant rock is andesite (52 to 63% of SiO2 content). Because of its medium viscosity, it is a builder of scoria cones and lava domes. SFVF has some higher dacitic dome with higher SiO2 content (WOLFE et al., 1987). The most typical forms of the area are volcanic forms, and the volcanic history of the SFVF is young, dating to only 5 - 6 million years. Within this interval, the age of structures is on a wide scale. There are cones with ages of a few thousand years (Sunset Crater), million years (San Francisco Mountain: 0.4-1 million years) and millions of years (Bill Williams Mountain: 5-6 million years) (TANAKA et al., 1986). The following table shows the age groups defined by the aforementioned studies (COLTON, 1967 and MOORE & WOLFE, 1976). Older Pliocene cones are located in the western and southwestern parts of SFVF, while eastward there is a younging trend (TANAKA et al., 1986). The easternmost cones are Late Pleistocene—Holocene in age. This younging trend can prove that the area has developed over a hot spot (PRIEST et al., 2001).

Based on the classifications of COLTON (1967), MOORE & WOLFE (1976) and USGS 1: 50,000 scale maps, BATA et al., (2007) defined the following age groups that will be used in this study:

• 1. Holocene—Late Pleistocene (0 - 0.16 Ma)

• 2. Middle Pleistocene (0.16 - 0.73 Ma)

• 3. Early Pleistocene (0.73 - 2.0 Ma)

• 4. Early Pleistocene – Late Pliocene (2.0−2.48 Ma)

• 5. Late Pliocene (2.48−5.0 Ma)

Figure 3. Location of the cones by age in the area (BATA et al. 2008) gray: 1st age group, green:2nd age group, blue: 3rd age group, red: 4th age group, white: 5th age group

Chaîne des Puys

Chaîne des Puys (CdP) is located in central France, in the Auvergne region, near Clermont-Ferrand, in the south of the Massif Central. In this predominantly Quaternary volcanic area there are about 100 vents: 48 scoria cones, 8 lava domes and 15 maars are distributed in North-South direction for about 40 kilometers. The highest point is Puy de Dôme (Figure 4), which was formed by two domes 10,800 years ago (MIALLIER et al., 2010, 2012, VAN WYK DE VRIES et al., 2014).


605

Figure 4. Puy de Dôme from Puy de Côme

Basalt, trachybasalt and trachyandesite are found in the area, while the other structures (domes) are predominantly trachitic. The youngest cone is ~ 4.4 ky old, while the oldest is ~ 100 kz old. Based on these, it can be said that, compared to SFVF, CdP cones are much closer together in time (BOIVIN, 2009).

DATA AND METHOD

SRTM

The DEM data of SRTM (Shuttle Radar Topography Mission) can be downloaded for free for the 80 % of the world. A more detailed model applies to the continental United States of America (approx. 30*30 meter resolution), while the rest of the world has made the aprox. 90*90 m version available. We used the 30 meter (for SFVF) and 90 meter (for CdP) resolution SRTM from the CGIAR (2004) website., and the 10 meter resolution from USGS 3D Elevation Program.

LiDAR

One of the most accurate methods of surface mapping and terrain modeling is laser scanning based on LiDAR (Light Detection and Ranging) technology that emits laser light (UV, visible or near infrared wavelengths) and determines the distance between the instrument and the reflecting surface from its reflection time. We had 5 meter resolution LiDAR data for the whole Chaîne des Puys, and 0.5 metres for the center part, which is our study area.

Figure 5. Shaded relief map of the San Francisco Mountain in 30 m (on the left) and 10 m (on the right) resolution


606

Figure 6. Shaded relief map of the Puy de Côme in 90 m (on the left) and 0 .5m (on the right) resolution

Mann–Whitney test

The most common statistical test is t-test that can only be applied to a standard data set. In the present case, however, most of the data series to be compared are considerably different in size, so we chose the nonparametric equivalent of the two-sample t-test–used for not standard distribution and ordinal (sorted) variables. The Mann–Whitney test is especially used to compare groups with a small number of elements. It can be used to determine whether two independent samples were selected from populations having the same distribution. We calculated value p; if the significant level is under 0.05, the null hypothesis of the different originating distributions is accepted.

RESULTS

Average relief and slope values were calculated for 308 cones in SFVF, 26 cones/domes in CdP, and the values were grouped by age. The above mentioned 5 groups (first: 11 pcs, second: 131 pcs, third: 26 pcs, fourth: 113 pcs, fifth 24 pcs) were used in SFVF. We had ages between 8 and 64 ky for CdP’s cones, so 3 groups were created. During determinations the main point was the equal number of cones of the groups because of the statistical test, so we have 9,8, and-9 cones, respectively, in each group.

San Francisco Volcanic Field

Relief

Figure 7. Boxplot diagram of average relief values of the scoria cones in SFVF in 30 m (on the left) and 10 m (on the right) resolution


607

The results of the analysis of the relief is shown in boxplot diagram (Fig. 7). The figure suggests that the height of the cones decreases with time, and this trend does not depend on the DTMs’ resolution. It is more important, that the maximum height is higher and the relief values in one group are less scattered with the 10 meter resolution DTM. These values are equal with previous researches (HOOPER and SHERIDAN (1998)), unlike 30 meter resolution’s values.

Figure 8. Relief relationships and p values of the age groups of SFVF, significant differences in red (p <0.05) in 30 m (on the left) and 10 m (on the right) resolution

To show significant differences Mann–Whitney test was used. We can see in Figure 9, of the 10 groups 6 pairs 6 differ significantly. Although the altitudes in the groups are higher in the case of 10 meters resolution, there is no change in their proportions.

Slope

Figure 9. Boxplot diagram of average slope values of the scoria cones in SFVF in 30 m (on the left) and 10 m (on the right) resolution

The most common parameter for describing volcanic edifices is slope. The youngest cones have the steepest shape and with age it becomes more and more shallow-sloped – regardless of resolution. We have same trend with relief: with better resolution values have less deviation and are higher. Here it is more typical: according to HOOPER & SHERIDAN (1998) the average slope is between 26° and 8°. 30 meter resolution DTMs give us non-realistic results.

Figure 10. Slope relationships and p values of the age groups of SFVF, significant differences in red (p <0.05) in 30 m (on the left) and 10 m (on the right) resolution


608

Although there is a big difference between data provided by different resolution DTMs, this does not appear on the Mann–Whitney test.

Chaîne des Puys

Relief

Figure 11. Boxplot diagram of average relief values of the scoria cones in CdP in 90 m (on the left) and 0.5 m (on the right) resolution

Because of the small number of cones and the different lithology it is a little bit different with the CdP’s cones – not very representative –, but both method can be used. We can see in Figure 11 the difference is ca. 50 meters in the elevation values – which represent the 20% of the full height.

Figure 12. Relief relationships and p values of the age groups of CdP, significant differences in red (p <0.05) in 90 m (on the left) and 0.5 m (on the right) resolution

Slope

Figure 13. Boxplot diagram of average slope values of the scoria cones in CdP in 90 m (on the left) and 0.5 m (on the right) resolution


609

Figure 14. Slope relationships and p values of the age groups of CdP, significant differences in red (p <0.05) in 90 m (on the left) and 0.5 m (on the right) resolution

Just like SFVF’s cones the CdP slopes are higher in better resolution (the difference is about 5° in every age group).

DISCUSSION

It is a common problem in the statistical evaluation of DTM studies that the results might be dependent on the resolution of the input grid. This is especially true if the evaluation contains calculation of slope angles, because derivative of the lower resolution DTMs tends to underestimate steeper slopes. As in the case of the scoria cones the slope distribution (average, standard deviation, etc.) is the key feature describing the cone, the statistical studies always face the aforementioned problem whether the resolution is influencing the results or not. In our study we intended to point out that there could be a threshold resolution that is already suitable for defining age groups that give statistically significant separation.

In Figure 15 we can see the relationships between the relief and slope values of the SFVF’s cones. In axis x there are the result of the worse DTM, in axis y the better. According to the relief there is no difference in altitude, the values are located along a nearly 45 ° straight line (y=x). In the slope values there is a larger spread along the steeper straight. The 0.5 meter resolution DTM has higher values (Figure 8); in most cases the difference is around 5°.

Figure 15. Cross-plot diagram of average relief and slope values in SFVF; black is a fitted and the gray is a 45° line

In case of CdP the resolution difference is bigger (0.5-90 m). Nonetheless the relief values are still located in a same 45° straight line, but - due to its resolution - it is shifted slightly along the y axis. The previously highlighted difference in the slope values is also shown here: the average slope of each scoria cone shall be at least 5 ° in different resolutions.


610

Figure 16. Cross-plot diagram of average relief and slope values in CdP; black is a fitted and the gray is a 45° line

In the two presented areas the total number of cones are very different, however, the behaviour has been found similar. In order to double check our conclusions we randomized the values of the SFVF (unfortunately in case of CdP the total number of cones is too low for that) and grouped randomly into groups. Of the 10 such random tests 8 showed p>0.05 values for every (10) group pairs and only 2 cases provided p<0.05 (significant difference). The Mann–Whitney test results clearly indicate that the age groups have a strong relationships with the degradation of the cones.

CONCLUSIONS

Based on the results obtained from the comparisons of the two study areas we can conclude:

1. In accordance with our expectation using a better resolution DTM will provide more realistic results and it is more straightforward to compare scoria cone groups.

2. The significance of the differences can be highlighted most prominently by analysing the slope angle distributions. If the number of cones is high (like in the case of SFVF) the age-based grouping shows clear differences; this observation is supported by the fact that randomization of the grouping in many cases does not show statistically significant differences.

REFERENCES

Bata T, Székely B., Karátson D. (2008): Determination of morphometric parameters of scoria cones in San Francisco Volcanic Field. Geophysical Research Abstracts 10:05631.

Boivin, P., Besson, J. C., Briot, D., Camus, G., de Goër de Herve, A., Gougaud, A., ... & Miallier, D.: Volcanologie de la Chaîne des Puys, 5° édition bilingue. Editions du Parc Régional des Volcans d’Auvergne, Aydat, 179. (2009)

CGIAR (The CGIAR Consortium for Spatial Information, SRTM Data Search) [online] srtm.csi.cgiar.org. Available at: http://srtm.csi.cgiar.org/SELECTION/inputCoord.asp [Access: 25 April 2018]. (2004)

Colton, H.S.: The basaltic cinder cones and lava flows of the San Francisco Mountain volcanic field: Museum of Northern Arizona Bulletin 10, p. 58. (1967)

Favalli, M., Karátson, D., Mazzarini, F., Pareschi, M.T., Boschi, E.: Morphometry of scoria cones located on a volcano flank: A case study from Mt. Etna (Italy), based on high-resolution LiDAR data, Journal of Volcanology and Geothermal Research 186:(3-4) pp. 320-330. (2009)

Hooper, D.M., Sheridan M.F.: Computer-simulation models of scoria cone degradation. Journal of Volcanology and Geothermal Research 83: 41 - 267 pp. (1998)

Miallier, D., Boivin, P., Deniel, C., Gourgaud, A., Lanos, P., Sforna, M. & Pilleyre, T.: The ultimate eruption of Puy de Dôme volcano (Chaîne des Puys, French Massif Central), about 10,700 yr ago. Comptes Rendus Géoscience, 342 (11): p. 847-854 (2010)


611

Miallier, D., Pilleyre, T., Sanzelle, S., Boivin, P., Lanos, P.: revised chronology of the youngest volcanoes of the Chaîne des Puys (French Massif Central). Quaternaire, 23, (4): p. 283-290 (2012)

Moore, R.B., Wolfe, E.W.: Geologic map of the eastern San Francisco Volcanic Field, Arizona: USGS Miscellaneous Investigations Series Map I -953, scale 1:50,000. (1976)

Porter, S.C.: Distribution, morphology, and size-frequency of cinder cones on Mauna Kea volcano, Hawaii. Geological Society of America Bulletin 83, p. 3607-3612. (1972)

Priest, S.S., Duffield, W.A., Malis-Clark, K., Hendley, J. and Stauffer, P.: The San Francisco Volcanic Field, Arizona, USGS Fact Sheet 01701, Flagstaff, Arizona, p. 1-2. (2001)

Tanaka, K.L., Shoemaker, E.M., Ulrich, G.E. and Wolfe, E.W.: Migration of volcanism in the San Francisco volcanic field, Arizona. Geological Society of America Bulletin, 97: 129-141. (1986)

van Wyk de Vries, B., Marques, A., Herrera, R., GranjaBruña, J.L., Llanes, P. Delcamp, A.: Craters of elevation revisited: forced folds, bulging and uplift of volcanoes. Bulletin of Volcanology 76(11): 1-20. (2014)

Wood, C.A., Morphometric evolution of cinder cones. J. Volcanol. Geotherm. Res. 7, p. 387-413. (1980a)

Wood, C.A., Morphometric analysis of cinder cone degradation. J. Volcanol. Geotherm. Res. 8, p. 137-160. (1980b)

Wolfe, E.W., Ulrich, G.E., Holm, R.F., Moore, R.B. and Newhall, C.G.: Geologic map of the central part of the San Francisco Volcanic Field, North Central Arizona. USGS Misc. Field Studies Map, MF-1959, scale 1:50,000. (1987)

FV was supported by the ÚNKP-17-2 New National Excellence Program of the Ministry of Human Capacities, Hungary. BSz contributed as an Alexander von Humboldt Research Fellow.

BIOGRAPHY

Fanni Vörös is from ELTE Eötvös Loránd University, Department of Cartography and Geoinformatics. She received her BSc degree in Earth Sciences in 2017, now she is doing MSc in cartography. Her interest in volcano morphometry started in 2015. Since then she presented her work three times at the European Geosciences Union, at Student V4 Geoscience Conference and Scientific Meeting GISÁČEK conference (ISBN: 9788024840338). She won first prize in the National Scientific Students' Associations Conference in the Geomorphology session and the Outstanding Student of the Faculty of Informatics award. Parallel to her studies she acts as a teaching assistant in GPS and cartography.

Assoc. Prof. Dr. habil. Balázs Székely (MSc in geophysics and astronomy (Eötvös University), PhD in geology (Tübingen University)) is an interdisciplinary geoscientist at the Department of Geophysics and Space Science, Eötvös University. He spent 6+ years at Tübingen University, 10 years at Technische Universität Wien as research fellow, and 4 semesters at Interdisciplinary Ecological Center of TU Bergakademie Freiberg as an Alexander von Humboldt Research Fellow. He is active in areas like archaeometry, tectonic geomorphology, geomorphometry, and shallow geophysics and projects of tectonic geomorphology, LiDAR mapping of active landslides, crop yield estimation, volcanic and Martian geomorphometry, LiDAR studies on lacunarity of forests. He is a (co-)author of 70 scientific papers (1000+ citations, h=17) and participated in publishing of georeferenced historical map data (Military Surveys of the Habsburg Empire). He is a founding member of the editorial board of Archeometriai Műhely, and a co-convener of numerous sessions at EGU General Assemblies.

Professor Benjamin van Wyk de Vries (BSc Geology, London, PhD Open University, HDR, Université Blaise Pascal) worked for Instututo Nicaraguense de Estudios Territoriales (1985-1990), then Open University, and Blaise Pascal University (now Clermont Auvergne). Generalises in natural risk, scientific communication, geoheritage, geomorphological, tectonics and volcanology problems. Has published about 90 papers (with healthy citations and h factor). He coordinates INVOGE – International Geological Masters in Geotechniques and Volcanology, and and ERASMUS+ capacity building project 3DTelc (www.3DTelC.com). He is currently half seconded to the Conseil Departmental du Puy de Dome for a UNESCO nomination 'Tectonic area of the Chaine des Puys and Limagne fault'.


612

Documents

GEOMORPHOMETRIC DESCRIPTIVE PARAMETERS OF SCORIA … · GEOMORPHOMETRIC DESCRIPTIVE PARAMETERS OF SCORIA CONES FROM DIFFERENT DTMS: A RESOLUTION INVARIANCE STUDY Fanni Vörös, Benjamin