THE GREAT PATAGONIAN OIL SPILL (METULA), 41 YEARS LATER

This paper concentrates on followup surveys conducted in February 2015 at two principal areas where Metula oil deposited in August 1974 was most likely to remain. These areas are commonly known as Puerto Espora and East Espora Marsh. Figure 1 provides an overview of the Strait of Magellan showing a representation of historic shoreline oiling and the study area. The shoreline indicated as oiled is measured (this report) as 204 km. Figure 2 shows an aerial photograph of the study area with oiling in February 1976.

The spill site climatically has low annual rainfall (250–300 mm) and low temperatures (mean monthly average between 2° C in winter and 10° C in summer), falling into the Köppen classification (BSk) for a cold semi-arid climate. Tides in the study area are ~7–10 m. Shoreline ice is not present. A brief history of scientific activities at the Metula spill site since 1974 to 2015 is provided below.

Studies during the Spill: Initial documentation of the vessel grounding, oil releases, photographs and shoreline impacts at the time of the spill are provided in Baker et al. (1976), Gunnerson and Peter (1976), Hann (1977), and Hann and Young (1979). The VLCC Metula grounded on the northwest side of the First Narrows (Primera Angostura) in the Strait of Magellan on 9 August 1974. The subsequent rupture of several tanks until refloating on 25 September 1974 resulted in a loss of 51,500 to 52,300 tons of light Saudi Arabian crude oil plus an estimated 2,000 tons of Bunker C fuel oil. Very strong currents and winds distributed the oil over much of the shoreline within the two embayments between the Second Narrows (Segunda Angostura) and the eastern opening of the Strait of Magellan to the Atlantic Ocean (Figure 1). Mousse (water-in-oil) formation occurred rapidly and shoreline surface deposits of several cm were common. Thicker deposits reached up to 20–30 cm along Puerto Espora beaches but likely included oil/sediment mixtures (Baker et al, 1976; Guzman and Campodonico, 1980). No cleanup was undertaken.

Studies 1975–1980: The author first participated in Metula studies beginning in August 1975 focusing on understanding the interactions of physical processes and spilled oil behavior on various shoreline types (mixed-sand and gravel/boulder, tidal flats and marshes; Hayes and Gundlach (1975), with an intensive study in February–March 1976 followed by a shorter site visit in August 1976 (Blount, 1976; Gundlach et al., 1978). Extensive areas of asphalted pavement (oil mixed with gravel) remained. Fresh oil, particularly within thick deposits in East Espora Marsh was evident.

Colewell et al. (1978) collected field samples in 1976 finding that biodegradation of the oil was proceeding relatively slowly even at low temperatures (3°C) but was ineffective at degrading thick and/or buried deposits. They predicted that complete oil removal “will not occur for years, perhaps decades”. Straughan (1976) conducted a biological survey in January 1975 finding that physical impact (smothering) was the most important factor related to species mortality. Guzman and Campodonico (1980) report physical and ecological observations of spill impacts extending to 1980, including references to studies by other Chilean and international scientists.

Studies 1981–1990: With colleagues, the author completed a field survey of 12 sites in 1981 that were previously classified as moderately-to-heavily oiled in 1975/1976 (Gundlach et al., 1982). Persistence was most obvious at Puerto Espora (along both the outer spit and sheltered interior beach) and at East Espora Marsh, with the prediction that oil in East Espora Marsh “may persist for more than 100 years”. Some Salicornia plants had re-established in the marsh including by growing on top of the oil deposit. Owens et al. (1987) found oil remaining at nine surveyed sites. Oil persistence was based on exposure to wave energy such that oil was found in sheltered locations and in areas where oil was deposited above the active swash area during high tidal levels at the time of the spill. Chemical analyses showed relatively fresh oil in Espora Marsh and at Puerto Espora. The hard pavement surface was highly weathered. The fresh oil was characterized as “almost fluid and apparently unweathered”.

Studies 1991–2000: Baker et al. (1995) report on 1990/1991 and 1993 surveys in the outer portion of East Espora Marsh focusing on the recovery of vegetation. Natural recovery was indicated by Salicornia growth in cracks of weathered oil and where oil depths were less than an average of 2.4 cm. Other areas having thick oil concentrations (average 4.1 cm) showed no recovery. Oil chemistry completed on two samples show a high degree of weathering in thin oil and little degradation in the middle of a thick mousse layer. With these results, Baker (1995) predicted, “natural recovery for such an extreme scenario can be predicted as substantially more than 20 years”. Baker et al. (1993 and 1994) report on remediation plots placed in outer East Espora Marsh. In January 1993, sites were tilled with and without the addition of fertilizer and compared to controls. Short term results show heterogeneity and that a single tilling was likely inadequate due to the thickness of the deposits. Followup site visits were conducted by Owens et al. (1999) in May 1998 (described below) and by the author in 2015 (this paper).

Bragg and Owens (1995) report on the lack of mineral-fines interaction (flocculation) on two Metula mousse samples and sediments. The oil with water removed was high viscous (23,000 cp). Because the extracted fine sediments did interact with weathered Exxon Valdez oil, the interaction of Metula oil with fine-grained material may have occurred during the earlier stages of the spill.

In December 1995, the author returned to the spill site and documented that fresh mousse appeared under a thin coating of silt/clay sediments in East Espora Marsh, that asphaltic remnants remained on the gravel beach fronting the marsh, and that pavement within the interior Puerto Espora area had eroded less than 1 m since 1981 (Gundlach, 1997).

Shigenaka and Henry (1999) visited Puerto Espora in 1995 measuring the length of oil pavement as 240 m and chemical analyses of the pavement showed very high degradation (<2% extractable hydrocarbons).

Owens et al. (1999a, b) conducted site visits in May 1998 at East and West Espora Marshes and Puerto Espora finding only limited recolonization in East Espora Marsh with some areas basically 100% covered by weathered oil. The review of the experimental plots set by Baker et al. (1994) showed substantially better recovery of the tilled plots compared to controls and tilling with the application of fertilizer, perhaps due to infrequent tidal coverage that was unable to dissolve the pelletized fertilizer. Some seeding had also taken place in cow hoof prints and cracks in the oil cover. The oil chemistry of samples taken in this area ranged from highly weathered to relatively fresh with n-alkanes as light as C-12. Plant recovery at two plots in West Espora Marsh was limited to the landward side of the oil deposits and a growth in plant size between 1977 and 1987 surveys. At Puerto Espora, the pavement was measured as 242 m in length with a width reaching 15.6 m. Some erosion was evident along the upper edges and the pavement was discontinuous in places. Chemistry showed this pavement was highly weathered. Wang et al. (2001) estimated that the East Marsh untilled plot area was 25–32% weathered as compared to the original oil. In the tilled plot, weathering was 40–55%.

Studies 2001–2015: Owens and Sergy (2005) visited the same three sites in 2005, finding a maximum loss of 60 cm along the pavement west of the concrete block at Puerto Espora and notable erosional loss across the pavement in two locations. The East Espora Marsh showed an increase in colonization (2× to 5×) at two large plot areas from 1998 to 2005, but the area remained unvegetated in most sections. A review of the Baker et al. (1994) test plots showed a reduction in the number of plants due to the dominance of fewer larger plants but they had increased in surface area coverage from 1998 to 2005. The West Espora Marsh showed an increase in recolonization as plants spread over the oil.

Primary data sources for this report are photographs and field survey data principally noting oil location, condition and thickness from 1975 (August), 1976 (February, March, August), 1981 (February), 1995 (December) and 2015 (February). Aerial photographs taken in February 1976 assist in providing oil location for later comparison in East Espora Marsh. Additional photographs from Dr. Jenifer Baker taken in December 1991 and March 1993 related to the remediation test plots in East Espora Marsh conditions are also utilized. The location of photographs in 2015 was noted by GPS in the field and plotted in Google Earth imagery from 23 October 2014. QGIS software was used to make report maps and measure the length of oiled shoreline. Field photographs were also located by comparing reference points (manmade objects and local topography (channel configuration and adjacent hills). The studies described in the Introduction also provided many relevant observations and chemical analyses to understand the changes occurring at these locations.

Three areas are described: Puerto Espora and two areas (inner and outer) within East Espora Marsh (Figure 2).

The erosion of the asphalt pavement from 1976 to 2015 at Puerto Espora is shown in Figure 3. Gundlach (1997) estimated <1 m erosion of the upper edge of pavement from 1976 to 1981 and an addition regression of 0.5 to 1 m from 1981 to 1995 (total loss ~1.5 to 2 m). In 2015, Figure 3 illustrates a 2.5–3.0 m regression from 1976 to 2015 using the concrete block in the background for comparison.

In 2015, the spill pavement was very discontinuous and very highly weathered (mineralized) such that it was commonly difficult to differentiate remnant oil from background sediments. Table 1 shows measurements of the remaining Metula oil as measure westward from the concrete block showing the band becoming discontinuous (e.g. complete erosion at 73.4 and 105 m) and ending at 180 m. Shigenaka and Henry (1999) and Owens et al. (1999a) previously measured the length as 240/242 m. The width of remnant oil has a maximum of 8 m but is highly variable.

An aerial view of oiling conditions in East Espora Marsh in 1976 is shown in Figure 4. From 1976 to 1981, little-to-no change in marsh recovery nor in oil quantity was observed (Figure 5).

Continuing from 1981 to 1996 and to 2015, recovery progressed substantially (Figure 6). Figure 7 illustrates the common types of remaining oil in the area including: (a) 1–2 cm of surface tar / asphalt, (b) 4–5 cm of surface tar with brown mousse evident, and (c) a layer of oil buried 4–10 cm below surface clay.

This area is an embayment at the entrance to East Espora Marsh (Fig. 4 box A) and received heavy accumulations of oil during the spill. Essentially no recovery of vegetation was noted during my 1995 survey (Figure 8 top). A thin silt/clay coating was present over much of the surface, with fresh mousse present below the surface. In 2015, more vegetation is present, covering approximately 35% of previously vegetated areas within the embayment (Fig. 4, insert A). Mousse is still present where thick accumulations remain.

The staked plots set in 1993 (Baker et al. 1994) are still evident (Figure 8) and provide a basis for a further review of experimental oil treatments (tilling, fertilizer, tilling and fertilizer, and control). There is no obvious difference between the treatments in 2015. An example (sites C1 to C4) is shown in Figure 9. Vegetation regrowth is common in all plots. As described by Gundlach (1982, Fig. 10) and others, new plant growth can occur directly over the oily substrate, which seems to be the case with the experimental plots. Seeds can also settle on the thin clay/silt layer evident on the oil’s surface and in cracks in the oily pavement. It appears that vegetation has changed from earlier plot analyses but this was not reviewed during the 2015 survey.

The Metula oil spill site continues to provide valuable information concerning the longevity of spilled oil and serves as a model for the long-term effects of large crude oil concentrations. In February 2015, almost 41 years after the incident in August 1974, oil remains evident in the partially sheltered shoreline behind a large spit at Puerto Espora and even more so within the very sheltered East Espora Marsh area. From the survey in 1981, the author predicted that Metula “oil may persist for more than 100 years” (Gundlach et al, 1982). It appears that this prediction is well on the way to confirmation.

At Puerto Espora, remnants of the previously continuous asphalt pavement remain as highly weathered, discontinuous material with a length of 180 m and varying in thickness from less than a cm to a maximum of 10 cm. The maximum width is now reduced to 8 m. In many cases, the remaining degraded material is difficult to differentiate from natural sediments. Continued weathering and physical processes are expected to further reduce the quantity of remaining material.

In East Espora Marsh, oil remains common as a surface tar still containing mousse where oil is thick (~4 to 8 cm). Thin layers, as along the upper shoreline, are highly weathered and in the process of breaking up. Other areas show oil as a thin but hard surface tar (as along channel banks) and as oil covered by silt/clay ranging from a thin surface layer to a depth of ~10 cm. Cow hoof prints are common and may help plant recolonization by providing breaks in the surface tar. The inner marsh area shows greater vegetative recovery as compared to the outer marsh found at marsh entry (roughly 75% inner versus 35% outer), likely due to the thicker concentrations and less sedimentation found in the outer area. The amount of remaining oil has been visually reduced by weathering, biological processes and some physical breakup, perhaps being 25 % of the original deposit.

Experimental plots testing the efficacy of tilling and fertilizer application placed in the area in 1993 show no obvious differentiation in 2015. All plots show substantial vegetative regrowth occurring directly on top of oil deposits several cm in thickness.

It is noted that this survey only covered two areas that were most likely to have remaining oil. Most other areas, except for West Espora Marsh, are more exposed to wave action and are less likely to have remaining oil but cannot be confirmed as being oil free.

I would like to thank John Bauer and Judith Story for field survey participation, comments and photographs, and Dr. Jenifer Baker for providing historic reports on the experimental plots and advice concerning the revegetation observed in 2015.