Scholarly article on topic 'Quantifying the potential exposure hazard due to energetic releases of CO2 from a failed sequestration well'

Quantifying the potential exposure hazard due to energetic releases of CO2 from a failed sequestration well Academic research paper on "Earth and related environmental sciences"

CC BY-NC-ND
0
0
Share paper
Academic journal
Energy Procedia
Keywords
{}

Abstract of research paper on Earth and related environmental sciences, author of scientific article — Roger D. Aines, Martin J. Leach, Todd H. Weisgraber, Matthew D. Simpson, S. Julio Friedmann, et al.

Abstract Wells are designed to bring fluids from depth to the earth’s surface quickly. As such they are the most likely pathway for CO2 to return to the surface in large quantities and present a hazard without adequate management. We surveyed oil industry experience of CO2 well failures, and separately, calculated the maximal CO2 flow rate from a 5000 ft depth supercritical CO2 reservoir. The calculated maximum of 20,000 tonne/day was set by the sound speed and the seven-inch well casing diameter, and was greater than any observed event. We used this flux to simulate atmospheric releases and the associated hazard utilizing the National Atmospheric Release Advisory Center (NARAC) tools and real meteorology at a representative location in the High Plains of the United States. Three cases representing a maximum hazard day (quiet winds <1 m s−1 near the wellhead) and medium and minimal hazard days (average winds 3 m s−1 and 7 m s−1) were assessed. As expected for such large releases, there is a near-well hazard when there is little or no wind. In all three cases the hazardous Temporary Emergency Exposure Levels (TEEL) 2 or 3 only occurred within the first few hundreds of meters. Because the preliminary 3-D model runs may not have been run at high enough resolution to accurately simulate very small distances, we also used a simple Gaussian plume model to provide an upper bound on the distance at which hazardous conditions might exist. This extremely conservative model, which ignores inhomogeneity in the mean wind and turbulence fields, also predicts possible hazardous concentrations up to several hundred meters downwind from a maximal release.

Academic research paper on topic "Quantifying the potential exposure hazard due to energetic releases of CO2 from a failed sequestration well"

Available online at www.sciencedirect.com

ScienceDirect

Energy Procedía

ELSEVIER

Energy Procedía 1 (2009) 224021-2429

www.elsevier.com/locate/procedia

GHGT-9

Quantifying the potential exposure hazard due to energetic releases of CO2 from a failed sequestration well

Roger D. Aines*, Martin J. Leach, Todd H. Weisgraber, Matthew D. Simpson, S. Julio

Friedmann and Carol J. Bruton

Wells are designed to bring fluids from depth to the earth's surface quickly. As such they are the most likely pathway for CO2to return to the surface in large quantities and present a hazard without adequate management. We surveyed oil industry experience of CO2 well failures, and separately, calculated the maximal CO2 flow rate from a 5000 ft depth supercritical CO2 reservoir. The calculated maximum of 20,000 tonne/day was set by the sound speed and the seven-inch well casing diameter, and was greater than any observed event. We used this flux to simulate atmospheric releases and the associated hazard utilizing the National Atmospheric Release Advisory Center (NARAC) tools and real meteorology at a representative location in the High Plains of the United States. Three cases representing a maximum hazard day (quiet winds <1 m s-1 near the wellhead) and medium and minimal hazard days (average winds 3 m s-1 and 7 m s-1) were assessed. As expected for such large releases, there is a near-well hazard when there is little or no wind. In all three cases the hazardous Temporary Emergency Exposure Levels (TEEL) 2 or 3 only occurred within the first few hundreds of meters. Because the preliminary 3-D model runs may not have been run at high enough resolution to accurately simulate very small distances, we also used a simple Gaussian plume model to provide an upper bound on the distance at which hazardous conditions might exist. This extremely conservative model, which ignores inhomogeneity in the mean wind and turbulence fields, also predicts possible hazardous concentrations up to several hundred meters downwind from a maximal release.

© 2009 Elsevier Ltd. All rights reserved. Sequestration

1. Introduction

Release to the atmosphere constitutes a significant safety concern in the handling of any large volume of gas. While there are established measures for evaluating the risk from chemical facilities and pipelines, the specific hazards

* Corresponding author. Tel.: +1-925-423-7184.

E-mail address: aines@llnl.gov.

Lawrence Livermore National Laboratory, Livermore, California, USA 94550

Abstract

doi:10.1016/j.egypro.2009.02.003

associated with the geologic storage are only beginning to be addressed. Wells are designed to bring fluids from depth to the earth's surface quickly. As such they are the most likely pathway for CO2to return to the surface in large quantities and present a hazard without adequate management. In order to quantify the hazard, we constructed a hypothetical release scenario involving a supercritical CO2 reservoir at 1500 m (5000 ft) depth at a site in the central plains of the United States where the states of Colorado, New Mexico, Oklahoma, and Kansas join. We chose a specific site (deliberately not identified in this publication) with flat terrain and reasonably good local meteorological data.

2. Energetic Releases - Well Failure

Release of carbon dioxide from the underground reservoir could occur by failure of an injection well or the failure of an improperly sealed or compromised well. In order to maintain operational integrity, deep wells are cased and cemented and, ultimately, plugged and abandoned (Jarrell et al. 2002). Despite the long, successful history of well engineering, there are many potential failure mechanisms that could potentially allow CO2 to escape from deep reservoirs (Scherer et al. 2004, Gasda et al. 2006, Lewicki et al. 2007). Many conditions control the likelihood of well effectiveness, including the age and plugging mechanism, quality of completion, and post-closure history (Ide et al. 2006). Table 1.2 lists some well failures involving CO2 for which there is documentation or first-hand information available. Several other examples are cited without name or location by Skinner (2003).

Table 1. Examples of CO2 blow outs in the oil and energy extraction industries.

Location CO2 release rate (original units) CO2 release rate (kg/sec) Date Reference

Sheep Mt., CO At least 200x106 scf/ day (2.5 x106 m3/d) 120 March 17-April 3, 1982 Lynch et al. (1983, 1985) Holloway (2007)

Torre Alfina geothermal field, Italy 300 tons/hour 76 1973 Lewicki et al. (2007)

Travale geothermal field, Italy 450 t fluid/hr 113 Jan. 7, 1972 Geothermics Lewicki, et al. (2007)

Leroy Gas Storage Facility, WY 3e6 m3/year 0.2 1976-1981 Lewicki et al. (2007)

Edmund Trust #1-33 well, Kingfisher, OK 45 million cubic feet of gas/month 0.9 Dec. 2005-Jan. 2006 Lewicki et al. (2007)

Crystal Geyser, UT 2.6 to 5.8 kg/sec 2.6 to 5.8 Continuing Gouveia and Friedmann (2006)

Well failures in a CO2 field have some hazards common to any oil field blow out. A major exception is that CO2 is not flammable or caustic. However, there is the suffocation hazard presented by large concentrations of CO2 (IPCC 2005). Experience in the oil field (Skinner 2003) indicates that the CO2 is not particularly hazardous in most instances because it vents high into the air where it mixes, ameliorating dangerous conditions at ground level.

In order to determine whether the release rates in Table 1 are representative of the worst case, we conducted a theoretical analysis of the possible release rate from cased wells open to deep reservoirs containing supercritical CO2. The largest release would result from the complete removal of the wellhead leaving an open casing which would be a direct pathway to the CO2 at depth. Such an accident could occur, for instance, if a bulldozer accidentally sheared off a well head. In such a case the amount of CO2 released is a function of the pressure at depth, the size of the pipe, and to a lesser extent the friction and heat transfer in the pipe (the CO2 naturally cools off as it expands, which reduces the pressure and driving force - heat from the walls of the pipe reduces that effect). The velocity of the gas leaving the top of the pipe is limited by the speed of sound. This allows us to calculate the maximum CO2 that can be released.

The oil industry uses a variety of

Table 2. Calculated maximum CO2 well failure mass flow rates. ,. , , ,

casing diameters. We calculated

the possible release using two typical oil field configurations: a 7 inch ID casing with and without a concentric 4 inch OD injection tube. A range of depths (and therefore reservoir pressures) corresponding a site in the central plains of the United States was used. The resulting flows are somewhat higher than the estimated flows at the Sheep Mountain failure (the largest in Table 1) which is in accord with our conservative estimates of a completely open pipe. Any restriction (e.g., the injection tube) tends to reduce the velocity of the gas, and the overall flow rate.

In this scenario, the CO2 was confined to the casing along the entire depth of the well bore. There was no structural failure of the casing and no fissures were created as a result of the failure. The flow was assumed to be one dimensional and steady with wall friction losses and heat transfer. The supercritical CO2 pressure and temperature at the bottom of the casing were estimated for a given well depth based on a hydrostatic pressure assumption and a 25 K/km temperature gradient from the surface. Inherent in the steady flow assumption is that conditions of the CO2 in the well do not change as the gas escapes through the bore. Instead of an equation of state, an interpolation of National Institute of Standards and Technology tables (NIST) provided CO2 properties (density, viscosity, specific heat, enthalpy, and sound speed) as a function of temperature and pressure. With the tabulated data, the calculation allowed for phase changes, but was limited to a single phase at each depth due to the one-dimensional approximation.

The one-dimensional equations describing mass, momentum, and energy transport were numerically integrated from the given conditions at the bottom of the bore to the surface. Since the velocity at the well depth was unknown, the calculation began with an initial estimate and proceeded to determine the pressure, temperature, and velocity at points along the bore. If the surface CO2 pressure at the top of the bore was less than atmospheric or the exit velocity was slower than the sound speed, the initial velocity was increased and this process was iterated until either the pressure continuity or sonic flow condition was satisfied. The mass flow rate was determined using the velocity that satisfies these bore exit conditions.

Momentum losses due to wall shear stress were incorporated with a friction factor that depended on local CO2 properties, velocity, and casing wall roughness (0.046 mm) along the bore (White, 1986). The rate of heat transfer from the casing wall to the CO2 was incorporated into the energy equation and was proportional to the temperature difference between the wall and CO2. To simplify the calculation, an analogy between the heat transfer coefficient and friction factor was assumed (Shapiro 1953). Since the coupling is only one-way, the casing wall acted as a heat bath with a depth dependent temperature corresponding to the geological gradient of 25 K/km.

3. Atmospheric Hazard Assessment

In order to evaluate the maximum hazard represented by a well failure, we evaluated the historical meteorology in the vicinity of our hypothetical site over the last 11 years and chose three cases: a maximum hazard day (quiet winds), a medium hazard day (average winds) and a typical windy day which presents minimal hazard. The maximum hazard occurs in still conditions because the carbon dioxide is not diluted as it blows downwind, but accumulates in the vicinity of the wellhead. We then selected actual days from last year representing these conditions and used the National Atmospheric Advisory Release Capability (NARAC) facilities at LLNL to evaluate the consequences of the maximal release.

The National Atmospheric Release Advisory Center (NARAC) provides tools and services that map the spread of hazardous material accidentally or intentionally released into the atmosphere. Located at the Lawrence Livermore National Laboratory (LLNL), NARAC is a national support and resource center for planning, preparedness, real-

Release From 7 Inch ID Casing Release From 7 Inch ID Casing

without Injection Tube with 4 Inch OD Injection Tube

Reservoir Depth (m) Flow Rate Flow rate Flow Rate FlowRate

(kg/s) (tonne/day) (kg/s) (tonne/day)

1406 217 18,749 87 7,517

1488 224 19,354

1535 225 19,440 91 7,862

1555 226 19,526

time emergency response, and threat assessments involving nuclear, radiological, chemical, biological or natural emissions. NARAC predictions provide information on affected areas and populations, potential casualties, health effects, and protective action guides to assist decision makers and responders (for more information, see http://narac.llnl.gov). On April 15, 2004, the Homeland Security Council designated NARAC as the interim provider for the new DHS-led Interagency Modeling and Atmospheric Assessment Center (IMAAC), whose role under the National Response Plan is to serve as the single source of federal atmospheric dispersion predictions during Incidents of National Significance.

In NARAC's suite of three-dimensional models, the ADAPT model (Sugiyama and Chan, 1998) assimilates data from observations (e.g., from surface stations, rawinsondes, profilers) and/or weather forecast models, as well as land-surface data, for use in the NARAC dispersion model, LODI. ADAPT constructs meteorological fields (mean winds, pressure, precipitation, temperature, turbulence quantities, etc.) based on a variety of interpolation methods and atmospheric parameterizations (Chan and Sugiyama, 1997; Sugiyama and Chan, 1998). ADAPT produces nondivergent wind fields using an adjustment procedure based on the variational principle and a finite-element discretization. A finite-element representation is used for spatial discretization because of its effectiveness in treating complex terrain and its flexibility in dealing with variable resolution grids. The solution is obtained via a choice of conjugate gradient solvers, using a stabilization matrix to improve computational efficiency. The turbulent diffusivities, Kx, Ky, and Kz, are calculated as a function of height and horizontal location using scaling parameters and similarity-theory relationships described by Nasstrom et al. (2000). LODI is a Lagrangian particle model, but for the purposes of calculating the concentration, an Eulerian grid is used. In the cases that we simulated, the grid spacing near the source was 10 m x 10 m or 100 m2. Until the plume width approximately exceeds the length of 2 grid cells, LODI does not properly resolve the peak concentration, and therefore may underestimate the concentration near the source. A more detailed discussion of the NARAC modeling system and its validation is available in Nasstrom et al. (2006).

High concentrations of material injected into the atmosphere are most likely to occur with low wind speeds and a thermally stable atmosphere (i.e., a temperature inversion). Material disperses in the atmosphere due to the mean wind (transport) and turbulent mixing (diffusion). Obviously, transport will be small with slow wind speeds. Turbulence is primarily generated by wind shear and buoyancy forces. Wind shear is the changing of the wind vector (speed and direction) in the atmosphere. When the wind speeds are low, wind shear is small. The effects of buoyancy forces are suppressed in a stable atmosphere, resulting in less generation of turbulence. Therefore with low wind speeds and a stable atmosphere, dispersion is minimal and high concentrations of air-borne material are most likely to occur.

The predicted consequences from a maximal release of 225 kg/sec on a quiet day (November 25), a wind day (May 10), and an average day (March 1) are shown in Figures 1, 2, and 3. The concentrations shown are 15 minute averages representing the greatest levels of toxicity during the simulations. As expected for such large releases, there is a hazard in the immediate vicinity of the release when there is little or no wind. Regardless of the wind direction, toxic levels of CO2 would occur only in the near vicinity of the source. All three cases were selected as representative. The wind direction may vary over the entire compass for high, average or low winds, but is better defined for the average and high wind cases. In none of the three cases do concentrations associated with hazardous TEEL levels 2 or 3 exist far from the source, always occurring within the first few hundreds of meters. The maximum extent for TEEL level 2 predicted on the quiet day is 274 meters from the well head.

The calculations for figures 1, 2 and 3 are conservative because they require that all the released carbon dioxide stay near the ground. In actual well blowouts, much of the carbon dioxide vents high in the air where it mixes with the atmosphere and presents no hazard. This was the case for the Sheep Mt. CO2 failure event. It is important to note that releases of this size are not clandestine (Skinner 2003). The gas is leaving the well at the speed of sound and often snow condenses from the air - it is unlikely to go unnoticed. This was true of the Sheep Mt. events, equivalent events in the Otway Basin, Australia, and much smaller events such as those at Crystal Geyser (Bogen et al. 2006). Skinner describes some of the recent oil industry experience in controlling blowouts of this kind - no special safety measures are taken. This is in accord with our calculations, which indicate that it requires special considerations to generate hazardous conditions except right at the wellhead.

It is of interest to evaluate if weather conditions other than our hand picked minimum, average, and maximum days might theoretically result in larger hazard areas. In order to evaluate this we used a simple Gaussian model with assumed winds of 1, 3, and 7 m s -1, approximating the LODI simulations with low and moderate wind speeds (Figure 4). These models are conservative because they ignore turbulence and inhomogeneity in the wind field, often over-predicting the concentration. Gaussian Plume Models are frequently used in air pollution studies and for regulatory purposes. A simple GPM (Arya, 1999) was formulated and run for the six Pasquill-Gifford (PG) classes, where class A is the most unstable and class F is most stable. With source strength of 225 kg s-1, the source is assumed to have momentum such that the effective source height is 10 meters. The thin black lines represent the center-line concentrations for the PG classes as predicted by the GPM. The thick red line represents TEEL-3 (40000 ppm) and the thick orange line TEEL-2 (30000 ppm). The high wind speed case (7 m s-1 ) is not reproduced here as toxic levels of CO2 are never reached, even very near the source. Results from the low wind case (1 m s-1) indicate that TEEL-3 levels are exceeded for all stability classes. The interesting point to note is that for the more unstable cases (A through C) the distances at which the levels are first toxic are near the source, whereas for the more stable cases that distance is farther downwind. The distances at which the levels recede to be again below toxic levels are also farther downwind for the more stable cases. For example, TEEL-3 levels are exceeded up to about 750 m downwind for PG stability class F (most stable). For the moderate wind case, Figure 2, the wind speeds are assumed to be 3 m s-1. The TEEL-3 level is exceeded only very near the source for the most unstable case while TEEL-2 levels are exceeded up through PG stability class D. The more stable cases, E and F never exceed toxic levels.

According to the Gaussian Plume Model, a narrow plume with toxic levels of CO2 may exist several hundred meters downwind from the source, due to slow growth of the cross-wind spread of the plume. As stated above, this is a conservative estimate. Inhomogeneous wind or turbulence fields are ignored in the GPM. Near the source, the LODI estimates are most likely too low and the area where toxic levels occur may be several tens of meters greater than that model estimates. However, farther from the source, LODI, using the ADAPT generated winds provides a better estimate of the dispersion.

The results of both types of analysis indicate that, as has been observed in the oil industry, releases of this size represent a significant local hazard but do not affect a large area. Operational wells or abandoned wells in the immediate vicinity of populated areas may warrant additional monitoring or controls, but there is minimal or no risk presented at significant distances from wells.

4. References

Arya, S. P, 1999 Air Pollution Meteorology and Dispersion, Oxford University Press, New York, N.Y.

Chan, S.T. and G. Sugiyama (1997) A New Model for Generating Mass-Consistent Wind Fields over Continuous Terrain, Preprint, ANS Sixth

Topical Meeting on Emergency Preparedness and Response, San Francisco, CA (April, 1997), 375-378. Gasda, S.E., Bachu, S., and Celia, M.A. (2004). The potential for CO2 leakage from storage sites in geological media: Analysis of well

distribution in mature sedimentary basins. Environmental Geology 46 (6-7), 707-720. Gouveia, F.J. and Friedmann, S.J. (2006) timing and prediction of CO2 eruptions from Crystal Geyser, UT: Lawrence Livermore National

Laboratory Report UCRL-TR-221731, 16 p. Holloway, S., J.M. Pearce, V.L. Hards, T. Ohsumi, and J. Gale, (2007). "Natural emissions of CO2 from the geosphere and their bearing on the

geological storage of carbon dioxide," Energy 32, 1194 Ide, S.T., S.J. Friedmann, and H. Herzog (2006) CO2 leakage through existing wells: current technology and regulations. Proceedings of the 8th

International Conference on Greenhouse Gas Control Technologies, Trondheim, Norway 19-22 June, 2006. IPCC (2005) Intergovernmental Panel on Climate Change, IPCC Special Report on Carbon Dioxide Capture and Storage. Interlachen, http://www.ipcc.ch/

Jarrell, P.M., C.E. Fox, M.H. Stein, S.L. Webb (2002) Practical Aspects of CO2 flooding. Monograph 22. Society of Petroleum Engineers, Richardson, TX, USA.

Leone Jr., J.M., J.S. Nasstrom, D.M. Maddix, D.J. Larson, G. Sugiyama and D.L. Ermak,(2005) Lagrangian Operational Dispersion Integrator

(LODI) User's Guide, Report UCRL-AM-212798, Lawrence Livermore National Laboratory, Livermore, CA Lewicki, J.L., Birkholzer, J. and Tsang, C-F. (2007) Natural and industrial analogues for leakage of CO2 from storage reservoirs: identification of

features, events, and processes and lessons learned: Environ. Geology, v. 52, p. 457-467. Lynch, R.D., McBride, E.J., Perkins, T.K. and Wiley, M.E. (1983) Dynamic kill of an uncontrolled CO2 well. Soc. Pet. Eng. AIME, Paper

IADC/SPE 11378; IADC/SPE drilling conference; 20 Feb 1983; New Orleans, LA, USA Lynch, R.D., McBride, E.J., Perkins, T.K. and Wiley, M.E. (1985) J. of Pet. Technology, v. 37, no. 7, p. 1267-1275

Nasstrom, J.S., G. Sugiyama, J.M. Leone, Jr., and D.L. Ermak (2000) A real-time atmospheric dispersion modeling system, Eleventh Joint

Conference on the Applications of Air Pollution Meteorology, Long Beach, CA, Jan. 9-14, 2000. American Meteorological Society, Boston, MA, 84-89

Nasstrom, J.S., G. Sugiyama, R. L. Baskett, S.C. Larsen, M.M. Bradley (2006) The National Atmospheric Release Advisory Center (NARAC) Modeling and Decision Support System for Radiological and Nuclear Emergency Preparedness and Response. International Journal of Emergency Management,vol 4, no.3. NIST http://webbook.nist.gov/chemistry/fluid/

Scherer, G.W., M.A. Celia, J.H. Prevost, S. Bachu, R. Bruant, A. Duguid, R. Fuller, S.E. Gasda, M. Radonjic, and W. Vichit-Vadakan, "Leakage of CO2 through Abandoned Wells: Role of Corrosion of Cement", in The CO2 Capture and Storage Project (CCP), Volume II, D.C. Thomas and S.M. Benson (Eds.), 823-844, 200 Shapiro, A.H. (1953) The Dynamics and Thermodynamics of Compressible Fluid Flow: Volume I (Ronald Press Company, New York). Skinner, Les (2003) CO2 blowouts, an emerging problem. World Oil Magazine January, 2003.

Sugiyama, G. and S. T. Chan (1998) A New Meteorological Data Assimilation Model for Real-Time Emergency Response, Preprint, 10th Joint Conference on the Applications of Air Pollution Meteorology, Phoenix, AZ (11-16 January, 1998), Am. Met. Soc., Boston, MA. 285-289 U.S. DOE, 2007, available at www.hss.energy.gov/HealthSafety/WSHP/chem_safety/teel/TEELsRev-23Introduction.pdf White, F.M. (1986), Fluid Mechanics (McGraw-Hill, New York).

This work performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344.

5. Disclaimer

This document was prepared as an account of work sponsored by an agency of the United States government. Neither the United States government nor Lawrence Livermore National Security, LLC, nor any of their employees makes any warranty, expressed or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States government or Lawrence Livermore National Security, LLC. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States government or Lawrence Livermore National Security, LLC, and shall not be used for advertising or product endorsement purposes.

Figure 1. Health effects of a hypothetical release of carbon dioxide on November 25, a day with low wind.

Figure 2. Health effects of a hypothetical release of carbon dioxide on a day with average winds.

Figure 3 Health effects of a hypothetical release of carbon dioxide near on May 10, 2007, a windy day.

Set 1: Average Air Cone 10 May 2006 Release (Short-Term Population Effects) Automated Report- Testing

Acute (Short-Term) Effects

' H9 Description (ppm) Extent Area Population

jet! ¡.f; 2 1 >TEEL-3: Death or irreversible rtealth effects posaöle. >-10,000 108 m 5,436 m£ 0

HHH - h H SrmmmKi . >TEEL-2 ard TEEL-1: Ssious health effects or impaired ability to take protective action. >30,000 123 m 7,110 m2 0

! 1 Wm - f 1 yf* ÜHh ■ * \' Wistj m ^BHlf Vote: Areas and counts in the table are cumulative. Casualties include both Fatal and NonFatal effects.

1 11 flr^^ TC to May 10. £006 19:00 LlTC at or rear ground level. Release Location: 36.133333 N. 102-669170 W Material: CARBON DIOXIDE Generated On: November 06.2000 00:14 UTC Model: ADAPTA.OOI

IB*'1!!

JR BMÜ-: I En ■, B ..... ■i^HPpil^. . fl

m : JMHt. ■ ■ ■

Figure 4. Centerline concentration values for the six Pasquill-Gifford stability classes, labeled A through F, with a wind speed of 3 m/s (left) and 1 m/s (right). Above the red line, TEEL-3 conditions exist, above the orange line TEEL-2 conditions exist.