OR/14/005 Mt Meru case study

From MediaWiki
Jump to navigation Jump to search
Vye-Brown, C, Crummy, J, Smith, K, Mruma,A and Kabelwa H. 2014. Volcanic hazards in Tanzania. Nottingham, UK, British geological Survey. (OR/14/005).

Mt. Meru is a 4565m high stratovolcano in northern Tanzania, approximately 70 km west- southwest of Kilimanjaro. It is Africa’s fourth highest peak and is historically active, with its last eruption occurring in 1910 from the summit ash-cone. The summit of Mt Meru is just 14 km from the centre of the city of Arusha, which has a permanent population of just over 400 000 (2012 census). Arusha city lies in the Arusha region which has a population of approximately 740 000. It is estimated that 75% of the total number of tourists entering Tanzania, pass through the Arusha region which is on the Northern Circuit tourist route (Mwunga Pers. Comm.).

Previous work

The first study of Mt Meru was carried out in 1953 as part of the Sheffield University Expedition to Kilimanjaro (Guest and Leedal, 1953[1]). The authors climbed to the summit of Mt Meru and made observations and some interpretations of the geological features. They also described the recent volcanic activity and observed two main areas of fumaroles within the crater. Wilkinson et al. (1983)[2] produced a detailed geological map of Mt Meru and a description of the geology. Through whole-rock K-Ar dating of lava flows and stratigraphic correlation, Wilkinson et al. (1986)[3] published a volcanic chronology for Mt Meru and Mt Kilimanjaro.

Further detailed work was carried out by Roberts (2002)[4] who studied the geochemical and volcanological evolution of Mt Meru for his PhD. No other work has been carried out on Mt Meru.

Geological summary

Mt. Meru sits on a faulted terrain of flood basalts associated with the Neogene Rift Valley volcanic province (Wilkinson et al., 1986[3]; Roberts, 2002[4]). These flood lavas have a thickness of several hundred metres and overly the Usagaran basement gneisses (Wilkinson et al., 1986[3]). In the northwest of the Meru-Kilimanjaro area, faulting has a west-northwest — east-southeast trend; and in the southwest of the area, the trend is north-south (Wilkinson et al., 1986[3]).

K-Ar dating of volcanic rocks from the area reveal the flood lavas were erupted approximately 2.37±0.11 Ma (Wilkinson et al., 1986[3]). Lavas, which overflow the fault scarps, have been dated at 1.71±0.06 Ma therefore bracketing the faulting to between 2.37 and 1.71 Ma (Wilkinson et al., 1986[3]). Volcanic activity at Mt. Meru commenced approximately 1.5 Ma (Wilkinson et al., 1986[3]).

Volcanic Evolution of Mt Meru

Mt. Meru is the main volcanic centre, with the smaller centres of Meru West and Little Meru on its flanks, and small parasitic cones and craters on the lower slopes and surrounding plains (Dawson, 2008[5]). The earliest activity of Mt. Meru was the extrusion of nephelinite lava flows at Meru West approximately 1.5 Ma (Wilkinson et al., 1986[3]). Due to subsequent ash and lahar deposits, the extent and relationship of these lavas to the main volcanic centre is unknown (Wilkinson et al., 1986[3]).

Alkaline lava flows and clasts from volcanic breccia from the lower north-western slopes of Mt. Meru yield K-Ar dates of between 0.28±0.05 and 0.38±0.009 Ma (Wilkinson et al., 1986)[3]. Wilkinson et al. (1986)[3] therefore suggested that at around 0.35 Ma extensive alkaline lavas accumulated to the northwest of the current cone, and that much of the northern part of the volcano is underlain by these lavas. The date of 0.28 Ma is from an angular lava block typical of those from Little Meru. Little Meru is on the north-eastern flank of Mt Meru, and is a steep- sided symmetrical volcano with a height of 3800m (Roberts, 2002[4]). Wilkinson et al. (1986)[3] suggested that this block represents the lavas on which Little Meru is constructed; however, Roberts (2002)[4] noted that this block is characteristic of volcanic breccias that dominate Little Meru and therefore proposed that this age likely represents the age of Little Meru. Little Meru is constructed from moderately to poorly sorted volcanic breccias overlain by lava flows near the summit. Roberts (2002)[4] proposed that the construction of Little Meru was through numerous strombolian explosive events, with a final phase of lava extrusion.

A date of 0.17±0.01 Ma from one of the cones on the volcano’s northern flank is thought to represent one of the earliest events in the construction of the main volcanic cone of Mt. Meru (Wilkinson et al., 1986)[3]. This age is much younger than that of Little Meru, therefore the activity at Little Meru ceased well before the construction of Mt Meru.

Exposure of the main Mt Meru cone is limited due to densely vegetated lower slopes, and later lahar and pumice deposits on the middle and lower slopes (Roberts, 2002[4]). The main cone deposits are exposed in the walls of the crater and comprise intercalated lava flows and volcaniclastic tuffs and breccias (Wilkinson et al., 1986[3]; Roberts, 2002). Roberts (2002)[4] estimated a ratio of 70:30 for the volcaniclastic deposits to lava flows, showing that Mt Meru had a largely explosive history with intermittent effusive lava flows.

Roberts (2002)[4] divided the main cone deposits into three groups: the Lower Group; the Mid- cone Group and the Summit Group. The Lower Group comprises a series of sub-horizontal lava flows at the bottom of the western wall of the crater, which Roberts (2002)[4] suggested could have resulted from a series of lava lake flows. The Mid-cone Group form about 800m of the 1400m crater wall, comprising mainly volcaniclastic deposits (Roberts, 2002[4]). The ages of these deposits range from approximately 0.111 to 0.102 Ma (Wilkinson et al., 1986[3]). Intruded into the Mid-cone Group are numerous, vertical to near horizontal, cross-cutting dykes, estimated to be 3 to 20 m thick (Roberts, 2002)[4]. Mid-cone Group lavas are also exposed between Little Meru and Mt Meru (Roberts, 2002)[4].

The Summit Group consists of interbedded volcaniclastic breccias, tuffs and lava flows and represents the final phase of construction of the Mt. Meru cone (Roberts, 2002[4]). The Summit Group deposits erupted between ~0.08 and 0.067 Ma (Wilkinson et al., 1986[3]). The dominance of breccias and tuffs suggests the final constructive phase was explosive, comprising strombolian and sub-plinian eruptions (Roberts, 2002[4]).

Superficial Eruption deposits

LAHARS

Four major lahar deposits were initially identified on the north, south and eastern flanks of Mt. Meru by Wilkinson et al. (1983)[6]. Roberts (2002)[4] identified further lahar deposits on the western and north-eastern slopes of Mt Meru, and describes the lahars as being dominated by angular to sub-angular blocks of nephelinite and phonolite up to 3 m in diameter, supported in a pale-brown ash and/or clay matrix which contains lapilli-sized vesicular phonolite. He observes no internal structures in the deposits and describes the surface of the lahars as undulating.

The two lahars on the northern flank of Mt Meru are confined to drainage valleys and spread out laterally onto the plains. The flow fronts are lobate, comprising multiple pulses (Roberts, 2002)[4]. The thickness of the lahars, near their termini, is approximately 20–30 m (Roberts, 2002[4]). As this is where the lahars have spread laterally, their thicknesses, where confined to the river valleys, should be greater; therefore Roberts (2002)[4] estimated 20–30m to be a minimum thickness. Roberts (2002)[4] estimated the minimum volume of the north-western and north- eastern lahars as 1.9 to 2.9 km3 and 2 to 3.1 km3, respectively, based on the minimum thickness and aerial extent.

A lahar deposit to the east of Mt Meru has been described as the main Meru lahar (Wilkinson et al., 1983[6]; Roberts, 2002[4]). It has an estimated aerial extent of ~1500 km2, and extends as far as the base of Mt. Kilimanjaro, 60 km distant (Wilkinson et al., 1983[6]; Roberts, 2002[4]). Based on the absence of well-preserved flow termini, Roberts (2002)[4] suggested that this lahar was more fluid than those on the northern flanks of Mt. Meru. Such a fluid flow would produce a thinner flow (Roberts, 2002[4]). The only thickness estimate of this deposit comes from a river exposure near the base of Mt. Meru. The lahar thickness here is 20 m; however, the base of the lahar is not observed; therefore this is a minimum proximal thickness (Roberts, 2002[4]). The deposit volume could not be measured as there is no measured thickness; however, based on the assumption that the flow was fluid and therefore thinner than the northern lahar deposits, and taking the aerial extent into account, Roberts (2002)[4] estimated that for a thickness of 3 m the volume would be ~4.5 km3, making it one of the biggest lahars ever described. Roberts (2002)[4] does however concede that this may be a series of lahars rather than one massive deposit.

Roberts (2002)[4] suggested that the lahar deposits resulted from the presence of an ice cap on Mt Meru during the Last Glacial Maximum (~21 ka), either by melting of the ice, or indirectly by hydrothermal water. Hydrothermal systems can develop on glaciated volcanoes, and result in hydrothermal alteration at the summit (Vallance, 2000[7]), which in turn results in an increase in porosity, trapping water in the edifice.

Although there are abundant unconsolidated tephra deposits blanketing the slopes of Mt Meru, Roberts (2002)[4] suggested that the likelihood of future lahars has been reduced as a result of the collapse of the edifice to the east, which lowered the summit to below the permanent snow/ice level.

PUMICE AND ASH DEPOSITS

Pumice and ash deposits blanket much of the western slopes, and to a lesser extent, the northern and southern flanks of Mt Meru (Wilkinson et al., 1983[6]; Roberts, 2002[4]). They consist of a sequence of pumice fall, ash fall and reworked pumice and ash deposits (Roberts, 2002)[4]. The pumice and ash fall deposits are visible in drainage valleys where rivers have cut through the overlying reworked ash and soil layers (Roberts, 2002)[4].

Roberts (2002)[4] divided the pumice and ash deposits into three groups: near-vent/block facies pumice deposit; early/proximal pumice fall deposits; and the main pumice fall deposits. The near-vent/block facies pumice deposit is described at one locality, ~9 km north of the summit of Mt Meru. The deposit is ~20m thick comprising poorly bedded, clast-supported units of sub-angular to sub-rounded pumice and lithics (Roberts, 2002)[4]. The lithic clasts are phonoloite, nephelinite and black obsidian, and comprise ~15 vol. % of the deposit. The pumice and lithics are comparable in size, with maximum diameters of 30 and 25 cm, respectively (Roberts, 2002)[4]. Due to the comparable clast sizes, and the clast-supported nature of the deposit, Roberts (2002)[4] suggested that this deposit is associated with the initial stages of the Mt Meru pumice forming eruptions i.e. vent widening.

The early/proximal pumice deposits are also only described at one locality, on the west-southwest slope of Mt Meru. This sequence is exposed in an inaccessible river valley, and comprises at least six pumice fall units interbedded with dark-grey reworked units (Roberts, 2002)[4]. The bases of the reworked units are erosional.

The main pumice deposits are widespread, with exposures over 23 km west of Mt Meru (Roberts, 2002[4]). The main pumice deposits are a sequence of four primary pumice fall units (A, B and C) and one ash fall unit (F) interbedded with reworked units (Roberts, 2002[4]). The pumice fall deposits comprise angular pale-grey pumice with amphibole, sanidine and nepheline phenocrysts (Roberts, 2002[4]). Clast sizes vary depending on the thickness of the deposit, with the thicker deposit (A) containing the larger clasts (up to 6.5 cm; Roberts, 2002[4]).

Roberts (2002)[4] interpreted the main pumice deposits to represent the fallout from two explosive events: the first producing the ash fall deposits of unit F; and the second producing the sequence of pumice fall deposits of units A, B and C. Based on the eruption volume calculations of Pyle (1989[8], 1999[9]) and Fierstein and Nathenson (1992)[10], Roberts (2002)[4] estimated an eruption volume of 2.1 km3 for the second explosive event which produced units A, B and C. Using maximum clast size data and the model of Carey and Sparks (1986)[11], Roberts (2002)[4] also estimated a maximum column height of the eruption of 23 km.

DEBRIS AVALANCHE DEPOSIT

The Mt Meru debris avalanche deposit resulted in the 5 km wide and 8 km long horse-shoe shape crater open to the east (Wilkinson et al., 1983[6]). The collapse occurred approximately 8600 yrs BP (calibrated 14C; Hecky, 1971[12]; Wilkinson et al., 1983[6], 1986[3]), and produced a deposit covering an area of ~390 km2. Initially, this deposit was thought to have formed from a lahar (Wilkinson et al., 1983[6]); however, the hummocky nature of the deposit and the horse-shoe shape crater are characteristic features of a debris-avalanche deposit (Roberts, 2002[4]). Roberts (2002)[4] estimated the volume of the debris-avalanche deposit, based on the volume of the missing sector, to be 28 km3, making it one of the largest ever recorded (Socompa was ~36 km3, and Mt Shasta was 26 km3). There is no clear evidence for the cause of the collapse which resulted in the Mt Meru debris-avalanche (Roberts, 2002[4]).

Wilkinson et al. (1983[6], 1986[3]) proposed that the pumice and ash deposits were associated with the collapse event. However, based on field evidence (no young pumice deposits inside the Meru crater) Roberts (2002)[4] suggested that the pumice and ash deposits preceded the collapse event. There is no reliable dating for these deposits. Roberts (2002)[4] obtained Ar-Ar ages from sanidine phenocrysts from unit A pumice of >1 Ma. Roberts (2002)[4] suggested that significant atmospheric argon affected the dating.

POST-COLLAPSE DEPOSITS

Volcanic activity that occurred after the collapse has been confined to inside the crater, and included the ash-cone, a lava dome and lava flows (Roberts, 2002[4]). The ash cone is ~1.5 km wide at its base, rises 200 m above the crater floor and comprises dark ash and vesicular pumiceous blocks up to 30cm in diameter (Roberts, 2002[4]). The dominant style of eruptive activity is thought to be strombolian due to the wide range in clast sizes.

At the western base of the ash cone is a nephelinite lava dome, which is 350 m in diameter and ~70 m high (Roberts, 2002[4]). Numerous later lava flows originated from the lava dome, and flowed down the collapse scar extending up to 7 km distant (Roberts, 2002[4]). The most recent flows were in 1877 and 1886 (Padang, 1954; IAVCEI Catalogue of Volcanoes; referenced by Roberts, 2002[4]).

Roberts (2002)[4] suggested that these lava flows represent effusive activity in between explosive strombolian activity associated with variations in dissolved gas content.

The most resent eruption of Mt Meru was in 1910 with explosions on 26th October and 13th, 18th and 22nd December (Roberts, 2002)[4]. The explosions resulted in much of the crater being filled with ash. Since 1910, fumarolic activity has been recorded in the ash cone (in 1911, 1926, 1936 and 1953; Guest and Leedal, 1953[1]). Although no activity has been recorded since 1953, Roberts (2002)[4] reported that in the last few years several local guides working in the Arusha National Park have reported smelling sulphur whilst walking along the north crater wall. In March 1999, a 20 cm wide, 100 m long fissure opened on the southern slopes of Mt Meru and emitted steam for about a week (Roberts, 2002[4]). This was attributed to the result of heated ground water as it occurred in the wet season.

Exposure

The population of the area surrounding Mt Meru is concentrated in Arusha town located on the southern flanks of the volcano although there is a distributed rural population on all the flanks with the exception of the National Park boundaries on a sector of the eastern flank. Population is lowest in the arid western flank which coincides with the area inundated by tephra deposits from the most recent Holocene eruptions of Meru. Whilst the population figures provided by the Arusha Regional Government in the September 2013 census (Table 1) capture the permanent residents in the area, there are a significant number of tourists that use Arusha as a base for safaris and expeditions to Meru prior to an ascent of the neighbouring Kilimanjaro. There is also a significant expatriate community that reside in Arusha and who may not be captured in the current census data.

Table 1    Population data for the Arusha Region.
Population by District Households Male Female Total
Arusha Jiji 97540 199524 216918 416442
Arusha V 72150 154301 168897 323198
Karatu 42469 117769 112397 230166
Longidp 23494 60199 62954 123153
Meru 59499 131264 136880 268144
Monduli 31903 75615 83314 158929
Ngorngoro 33815 82610 91668 174278

Of the population data available, there is limited information about the age distribution of the population. However, the distribution and number of schools within the Arusha Region (Table 2) suggests that a significant proportion of the population is under 16.

Table 2    Number and type
of school in the Arusha Region.
Schools by District Preparatory Primary Secondary
Arusha City

108

111

47

Arusha V

108

116

48

Karatu

103

103

31

L ongidp

36

41

8

Meru

138

138

55

Monduli

44

60

19

Ngorngoro

64

64

11

Total

601

633

219

Information was provided on the water sources for the Arusha Region (Table 3). Water access is not a particular problem in the region due in part to the rains induced by the topography of Mt Meru and Kilimanjaro. However, a significant number of surface water storage and schemes that use surface runoff of meteoric waters may be subject to contamination in the event of an eruption through ash deposition.

Table 3    Type of water sources in the Arusha Region.
Water sources by District Shallow boreholes Deep boreholes Dams Surface water storage

Gravity schemes (source from rivers)

Juli Arusha

0

17

0

4

2

Arusha District

2

10

16

15

48

Karatu

16

9

10

50

18

Longido

2

13

10

37

39

Meru

7

17

3

32

45

Mondoli

6

18

68

158

17

Ngogoto

2

13

53

7

25

TOTAL

35

97

160

303

194

If further work to assess the volcanic hazard and risk is conducted this data may be used to provide population exposure indices. Further information on the location of critical services and infrastructure such as hospitals and medical facilities will be sought.

GIS and remote data

A GIS (Geographic Information System) was built in ArcGIS 10.1 using vector and raster data acquired for the whole of the Mt Meru and surrounding area. This system enables comparative analysis of datasets as well as providing a better understanding of the spatial and topological context of features within the study area.

GIS Vector Data

A series of vector shapefiles was obtained from a free-of-charge online source (www.diva- gis.org) for the whole of Tanzania. The files provided spatial and contextual information on the following types of features:

  • point features: named places (e.g. town, village, city, suburb, etc..) and other point features
  • linear features: railways (e.g. used, disused, abandoned, etc...), roads (e.g. dirt road, track, primary, etc...) and waterways (e.g. stream, river, drain, etc...)
  • polygon features: buildings (e.g. church, industrial, residential, etc...), landuse (e.g. farmland, landfill, reservoir, etc...) and natural features (e.g. water, riverbank, forest, etc...)

Remote Sensing Image Data

Remote Sensing image data were obtained from the USGS Global Visualisation Viewer (http://glovis.usgs.gov) when minimal cloud conditions prevailed over Mt Meru and the surrounding area. This data source provides global image data free-of-charge from specific satellite sensors, such as the TM (Thematic Mapper) onboard the Landsat-4/-5 satellite platforms and the ETM+ (Enhanced Thematic Mapper Plus) onboard the Landsat-7 satellite platform. These are spectral imaging sensors that measure the amount of solar radiation reflected by the ground surface across specific spectral wavebands. Data from both sensors were used in this study and their spectral and spatial characteristics are detailed in Table 4.

Table 4    Spectral and spatial characteristics of the Landsat TM and ETM+ sensors
(from http://landsat.usgs.gov/band_designations_landsat_satellites.php).
Sensor Band Spectral Characteristic (µm) Spectral Region Spatial Resolution (m)
Landsat TM

1

2

3

4

5

6

7

0.45–0.52

0.52–0.60

0.63–0.69

0.76–0.90

1.55–1.75

10.40–12.50

2.08–2.35

VIS VIS VIS NIR SWIR TIR SWIR

30

30

30

30

30

120

30

Landsat ETM+

1

2

3

4

5

6

7

8

0.450–0.515

0.525–0.605

0.630–0.690

0.750–0.900

1.550–1.750

10.400–12.500

2.090–2.350

0.520–0.900

VIS VIS VIS NIR SWIR TIR SWIR PAN

30

30

30

30

30

60

30

15

Level 1G terrain-corrected TM data were obtained for scene Path 168/Row 062 for 25 February 1987 (NASA, 1987a[13]), 1 June 1987 (NASA, 1987b[14]) and 17 February 1993 (NASA, 1993[15]). Level 1G terrain-corrected SLC-on (Scan Line Corrector-on) ETM+ data were obtained for the same scene for 14 September 1993 (NASA, 1999[16]) and 21 February 2000 (NASA, 2000[17]). A series of images were required in order that the peak and flanks of Mt Meru were fully exposed at some point during the time series as well as to minimise vegetation cover for mapping purposes (Figure 2).

Figure 2    Overview natural colour composite imagery for calibrated Landsat Path 198/ Row 062 scenes for a) 25/02/1987, b) 01/06/1987, c) 17/02/1993, d) 14/09/1999 and e) 21/02/2000. Data available from the U.S. Geological Survey.

These images were pre-georeferenced to UTM zone 37-South projection, with WGS84 horizontal datum. Each spectral band of TM and ETM+ data were converted from DN (digital number) to spectral radiance using the Landsat Calibration Pre-processing utility in ENVI Version 4.7 software (EXELIS Visual Information Solutions). This process uses published post-launch gains and offsets specific to the sensor for the user-defined date of image acquisition, in conjunction with parameters on sun elevation angle, band minimum DN value and band maximum DN values, each of which are specified in the associated image metadata file.

Figure 3    RGB band combinations for calibrated Landsat scene from 21 February 2000 a) 321, b) 432, c) 742, and d) 457.

Digital image processing was performed on each calibrated scene to highlight particular surface features. A series of images were generated using specific RGB display channel band combinations with histogram manipulation in order to enhance visualisation of the study area. RGB band combinations visualised (Figure 3) were the natural colour composite 321 (similar colouring to that observed through a digital camera) and the false colour-composites for 432 (healthy vegetation appears red), 742 (exposed lithologies appear in blue hues), and 457 (enhances variations in exposed lithologies). All image enhancements were exported as GeoTIFF files and incorporated into the GIS for future analysis.

The calibrated TM and ETM+ data can be further manipulated in future work to help distinguish specific surface compositions and thus enhance mapping.

Topographic Data

Topographic data was obtained from the USGS Earth Explorer online facility (http://earthexplorer.usgs.gov) over Mt Meru and the surrounding area. This data source provides remotely sensed image data free-of-charge, including topographic data from the SRTM (Shuttle Radar Topography Mission) as used in this study. The SRTM was flown on the Endeavour Space Shuttle during February 2000 acquiring C-band (5.6 cm) radar data from which to create a near-global data set of land elevations.

A 3-arc-second (90m) resolution GeoTIFF tile for 036°-037°N and 001–002°S (SRTM3S04E036V2) was downloaded from Earth Explorer. The elevation data was reprojected to the UTM zone 37-South projection, with WGS84 horizontal datum, and a hill-shade generated to enhance visualisation of topographic features (Figure 4). Both elevation images were incorporated into the GIS.

Figure 4 SRTM a) colour-coded elevation for tile SRTM3S04E036V2 over the Mt Meru study area and visualised b) draped on hill-shade image. Data available from the U.S. Geological Survey.

Field study

A day was spent in the field looking at the tephra fall deposits described by Roberts (2002)[4], to the west and northwest of Mt Meru. In northern Tanzania, the prevalent wind direction is from the east, therefore tephra fallout deposits occur on the western slopes of Mt Meru, and beyond. Roberts (2002)[4] reported the locations of exposures of tephra fallout deposits (Figure 5). The aim of the day was to validate his observations, explore potential exposures along river gullies, and to sample pumice from fallout units to determine the ages of the eruptions. Dating of the pumice fallout deposits has been attempted; however, Ar-Ar dating on a sanidine-bearing pumice fall deposit yielded an age of 1 Ma, and was discarded (Roberts, 2002)[4]. Through improved dating techniques, we hope to accurately determine the ages of these large explosive eruption.

Figure 5    Mt Meru sample location map.

Yellow triangles show the locations of samples collected by BGS on October 2013 as part of this study. Blue circles are sample locations from Roberts (2002)[4].

Roberts (2002)[18] proposed that the tephra fallout deposits form part of a large ‘climatic’ eruption, based on the similarities in field characteristics of the deposits, and the whole-rock geochemistry. When plotted on a TAS diagram the deposits reveal a clear fractionation trend, which Roberts (2002) interpreted as the emptying of a large, fractionated magma chamber. He calculated the volume of such an eruption, taking the combined thicknesses of the units, as 2.1 km3.

Tephra Sections

As part of this study, tephra deposits were described and samples were collected at five localities to the west of Mt Meru (Figure 5). Locality 1 is in a dry river bed at the village of Kiushian, just off the main Arusha–Nairobi road. Within the dry riverbed, a ~1.5 m thick consolidated ash deposit with abundant pumice and lithics is exposed (Figure 6). Average clast sizes are between 1 and 5 cm, with rare larger clasts up to 30 cm. The deposit has internal structure with layers of more concentrated lithic and pumice fragments.

Figure 6    Field photographs of Locality 1 and Locality 2.

Locality 2 is further up the same riverbed towards Mt Meru. The river valley is much deeper here (~5 m). Exposed near the top of the channel walls is a ~50 cm thick, unconsolidated ash-supported deposit that contains abundant rounded cream pumice fragments. Overlying this deposit is a ~1.5 m thick reworked deposit, possibly a lahar, with very large blocks (up to 1 m) of lithics and consolidated pyroclastic flow material (Figure 6).

Figure 7    Field photographs of Locality 3: the road-cut reveals numerous interbedded ash-rich reworked, and pumice fall deposits.

Locality 3 is on the edge of the north-western slope of Mt Meru, through a Masaai village just south of the main Arusha-Nairobi road. A series of tephra fallout and reworked pyroclastic deposits are exposed in a road-cut through a topographic high (or mound) (Figure 7). It appears as though the pyroclastic material is draping a topographic feature such as a monogenetic cone, of which there are many surrounding and on the flanks of Mt Meru. At least three tephra fall deposits were identified, all of which are pumice-supported with abundant lithics (Figure 7). The pumice is very light grey/cream in colour, and up to 2 cm along the long-axis. These deposits are all between 10 and 15 cm thick. Interbedded between the pumice fall deposits are ash-supported, pyroclastic flow deposits with reworked pumice and lithic fragments. These deposits vary in thickness from <1 m to >2 m.

Locality 4 is in a steep-sided river valley, ~50 m along the road from Locality 3. The gorge is ~20 m deep with clearly visible grey tephra fall deposits interbedded with yellow and pale cream ash-rich units (Figure 8). Near the top of the gorge is a rubbly a’a lava flow. The lava comprises a very dark glassy groundmass with abundant phenocrysts of feldspar, quartz and hornblende.

Figure 8    Locality 4 stratigraphic section with field photographs. Sample descriptions are in Table 1.

The final locality, 5, is located on the main Arusha–Nairobi road in another river gorge. The gorge is ~10 m deep, and exposes a thick (>2.7 m) tephra fallout deposit, overlain by a series of tephra and ash fall layers (Figure 9). These are then overlain by a ~70 cm thick pumice-rich deposit. The thick fallout unit is clast-supported with predominantly light-grey pumices. The deposit is lithic-poor. Phenocrysts are clearly visible in the pumice clasts. Clasts are typically 3 cm, with common clasts up to 5 cm across.

Figure 9    Locality 5 stratigraphic log with field photographs. Sample descriptions are in Table 1.

Tephra Samples

Pumice clasts were collected at each locality in order to try to date the large magnitude eruptions that produced the tephra fallout deposits. Roberts (2002)[4] previously tried to date a sanidine crystal from a pumice fall deposit; however, the results were questionable. Based on field evidence and whole-rock geochemistry, Roberts (2002)[4] described the fall deposits as resulting from one, climatic, plinian eruption. Dating of individual deposits would enable this hypothesis to be tested.

In total 12 samples were collected. The sample descriptions and locations are presented in Table 1. Nine of the samples comprise pumice clasts, while samples TANZ_2b and 2c are lithic fragments from a reworked lahar(?) deposit, and sample TANZ_4a is taken from the rubbly lava flow at Locality 4. Samples are labelled with the country (TANZ) followed by the locality and a letter (a, b, c) depending on the number of samples taken at a single locality.

Table 5    BGS sample descriptions.
Sample Locality Easting Northing

Description

TANZ_1

1

236029 9631733 Pumice fragments from consolidated ash-supported deposit
TANZ_2a

2

239314 9635172 Reworked rounded pumice fragments from lower ash-rich deposit
TANZ_2b

2

239314 9635172 Block of consolidated pyroclastic flow material. Light grey colour with lithic and pumice fragments. Overlying TANZ_2a
TANZ_2c

2

239314 9635172 Lithic block from same unit as 2b.
TANZ_3a

3

237963 9645158 Pumice from lower fall deposit
TANZ_3b

3

237963 9645158 Pumice, very similar to TANZ_3a. From fall deposit higher up section
TANZ_3c

3

237963 9645158 Pumice from fall deposit near the top of the road-cut
TANZ_4a

4

238268 9645325 Vesicular lava, very dark glassy groundmass with abundant phenocrysts of hornblende, feldspar and quartz.
TANZ_4b

4

238268 9645325 Pumice is pale grey in colour. Very friable — possibly highly altered. Angular clasts, average size 1–2 cm, max 4 cm.
TANZ_4c

4

238268 9645325 Very similar to TANZ_4b from a fall unit below.
TANZ_5a

5

240467 9647438 Pumice from thick fall deposit
TANZ_5b

5

240467 9647438 Pumice from upper pumice-rick deposit

Fieldwork recommendations

There are common drainage gullies along the north-western slopes of Mt Meru. In many of these, tephra fallout deposits are exposed. Clear variations in thickness can be seen between the exposed deposits in the river gullies, therefore a useful exercise would be to map along the gullies and correlate the deposits between the gullies. This would enable detailed study of the lateral continuity of the deposits, which would feed in to tephra dispersion modelling and enable estimations of eruption volume and column height (i.e. Pyle, 1989[19]; Carey and Sparks, 1986[11]; Connor and Connor 2006[20]). If no dates can be yielded from the phenocryst phases in the pumice samples, detailed mapping and investigation of the individual units would lead to a better understanding of the relationship between these units.

In order to determine eruption recurrence rates, the volcanic stratigraphy must be well understood. Detailed stratigraphic logging would enable this work to be carried out.

If exposure is limited, it may be possible to dig some trenches on the western slopes of Mt Meru. This would give a better spatial distribution of sample sites which reduces the uncertainty of tephra dispersal modelling.

References

  1. 1.0 1.1 GUEST, N J and LEEDAL, G P. 1956. The volcanic activity of Mount Meru. Records of the Geological Survey of Tanganyika 3, 40–46.
  2. WILKINSON, P, DOWNIE, C, CATTERMOLE, P J and MITCHELL, J G. 1983. Arusha. Geological Survey of Tanzania, Quarter Degree Sheet 143.
  3. 3.00 3.01 3.02 3.03 3.04 3.05 3.06 3.07 3.08 3.09 3.10 3.11 3.12 3.13 3.14 3.15 3.16 3.17 WILKINSON, P, MITCHELL, J G, CATTERMOLE, P J and DOWNIE, C. 1986. Volcanic chronology of the Meru-Kilimanjaro region, Northern Tanzania. Journal of the Geological Society, London 143, 601–605.
  4. 4.00 4.01 4.02 4.03 4.04 4.05 4.06 4.07 4.08 4.09 4.10 4.11 4.12 4.13 4.14 4.15 4.16 4.17 4.18 4.19 4.20 4.21 4.22 4.23 4.24 4.25 4.26 4.27 4.28 4.29 4.30 4.31 4.32 4.33 4.34 4.35 4.36 4.37 4.38 4.39 4.40 4.41 4.42 4.43 4.44 4.45 4.46 4.47 4.48 4.49 4.50 4.51 4.52 4.53 4.54 4.55 4.56 4.57 4.58 4.59 4.60 4.61 4.62 4.63 ROBERTS, M A. 2002. The geochemical and volcanological evolution of the Mt Meru Region, Northern Tanzania. PhD Thesis, University of Cambridge.
  5. DAWSON, J B. 2008. The Gregory Rift Valley and Neogene-recent volcanoes of Northern Tanzania. Geological Society Memoir 33.
  6. 6.0 6.1 6.2 6.3 6.4 6.5 6.6 6.7 WILKINSON, P, DOWNIE, C, CATTERMOLE, P J and MITCHELL, J G. 1983. Arusha. Geological Survey of Tanzania, Quarter Degree Sheet 143.
  7. VALLANCE, J W. 2000. Lahars. 601–606 Encyclopedia of Volcanoes 143, SIGURDSSON, H. (editor). (London: Academic Press.).
  8. PYLE, D M, 1989. The thickness, volume and grainsize of tephra fallout deposits. Bulletin of Volcanology 51, 95–112.
  9. PYLE, D M, 1999. Widely dispersed Quaternary tephra in Africa. Global and Planetary Change 21, 1–15.
  10. FIERSTEIN, J and NATHENSON, M., 1992. Another look at the calculation of fallout tephra volumes. Bulletin of Volcanology 54, 156–167.
  11. 11.0 11.1 CAREY, S N and SPARKS, R S J. 1986. Quantitative models of the fallout and dispersal of tephra from volcanic eruption columns. Bulletin of Volcanology 48, 109–125.
  12. HECKY, R E. 1971. The palaeoclimatology of the alkaline, saline Lakes on the Mt Meru lahar. PhD Thesis, Duke University.
  13. NASA, 1987a. Landsat TM scene LT51680621987056XXX01, Orthorectified, Terrain-Corrected. Image courtesy of U.S. Geological Survey. 25/02/1987
  14. NASA, 1987b. Landsat TM scene LT51680621987152XXX01, Orthorectified, Terrain-Corrected. Image courtesy of U.S. Geological Survey. 01/06/1987
  15. NASA, 1993. Landsat TM scene LT41680621993048XXX02, Orthorectified, Terrain-Corrected. Image courtesy of U.S. Geological Survey. 17/02/1993
  16. NASA, 1999. Landsat ETM+ scene LE71680621999257SGS03, SLC-on, Orthorectified, Terrain-Corrected. Image courtesy of U.S. Geological Survey. 14/09/1999
  17. NASA, 2000. Landsat ETM+ scene LE71680622000052EDC00, SLC-on, Orthorectified, Terrain-Corrected. Image courtesy of U.S. Geological Survey. 21/02/2000
  18. ROBERTS, M A. 2002. The geochemical and volcanological evolution of the Mt Meru Region, Northern Tanzania. PhD Thesis, University of Cambridge.
  19. PYLE, D M, 1989. The thickness, volume and grainsize of tephra fallout deposits. Bulletin of Volcanology 51, 95–112. PYLE, D M, 1999. Widely dispersed Quaternary tephra in Africa. Global and Planetary Change 21, 1–15.
  20. CONNOR, C B and CONNOR, L J. 2006. Inversion is the key to dispersion: understanding eruption dynamics by inverting tephra fallout. 231–243 in Statistics in Volcanology. MADER, H M, COLES, S G, CONNOR, C B and CONNOR, L J. (editors). (London: Geological Society.)