Environmental, regulatory and stakeholder plan

Daniel's other assignment, the one with four weeks of float, is the part of the plan that many engineering teams leave until last and regret. A development that is technically sound and economic can still be delayed for years by an approval that was not applied for in time, a community that was not consulted, or an emissions problem that was not designed out. Lena Haugen's advice when the team met her in week 3 was blunt: “The committee will read your economics. The regulator and the neighbors will read this chapter. Write it as if they were the audience, because they are.” This lesson covers the approvals the project needs, how sour gas shapes safety and emergency planning, the greenhouse gas inventory and how to reduce it, and the plan for working with the people affected.

The approvals

In North Dakota, the Industrial Commission's Oil and Gas Division regulates wells, spacing, unitization, underground injection and flaring, while the Department of Environmental Quality issues air permits and the Public Service Commission has a role for some pipelines. Federal air rules from the Environmental Protection Agency also apply to the battery. Daniel lists the approvals the project needs and the order in which they must be obtained, because some depend on others.

The longest lead item is unitization, if the pool's mineral ownership is divided among several owners and leases, which is likely for a pool of several square miles. Secondary recovery across lease lines generally needs a unit, so that owners share in the production in proportion to agreed tract factors rather than according to which wells the water happens to push oil toward. Forming a unit takes negotiation with every working-interest and royalty owner and a hearing before the Industrial Commission. Daniel puts the unitization work at the start of the define stage, so that it can be completed during phase 1 of option C and does not delay the phase 2 decision.

Sour gas: safety and emergency planning

Hydrogen sulfide is the main safety hazard of the project. At low concentrations it smells of rotten eggs, but at higher concentrations it deadens the sense of smell, and a few hundred parts per million can be lethal in minutes. The design measures described in Lesson 9, sour-service materials, H₂S detection, emergency shutdown and a safe flare, reduce the chance of a release. Emergency planning deals with what happens if one occurs anyway.

North Dakota's rules require wells and facilities producing gas with significant hydrogen sulfide to post warning signs, install detection and have an H₂S contingency plan, and federal OSHA standards protect workers. The team also applies the industry practice of estimating a radius of exposure from the potential release rate and H₂S content, which at Spruce Ridge comes to roughly 1 mile for the producing wells and includes three farmsteads. The contingency plan identifies those residents and sets out how they will be notified and protected, and the county's emergency manager is involved from the start.

The greenhouse gas inventory

Every development plan now needs an estimate of its greenhouse gas emissions, both for regulation and because investors and lenders ask for it. Hannah and Daniel build a source-by-source inventory for the battery at plateau. The main sources are fuel combustion in compressors and the heater treater, flaring of pilot, purge, upset and tank vapors, venting from gas-driven pneumatic controllers, and fugitive leaks from the thousands of valves, flanges and connections. Methane is counted with its 100-year global warming potential of 28, the IPCC's fifth assessment value; reporting programs specify their own values, which the define stage will apply.

The inventory compares two designs. The conventional design uses gas-engine compressors, gas-driven pneumatic controllers, routes tank vapors to the flare and has a basic leak inspection program. The low-emission design uses electric motors for compression, instrument air for the controllers, a vapor recovery unit on the tanks, and a leak detection and repair program with frequent optical gas imaging surveys.

# Greenhouse gas inventory for the Spruce Ridge battery at plateau: a conventional design and a low-emission design.
SI = False     # True: Canadian case (m3, Alberta grid), False: US case (bbl, North Dakota grid)
GWP = 28.0                                       # 100-year global warming potential of methane (AR5)
if SI:
    oil, gas = 320.0, 48000.0                    # m3/d oil, m3/d raw solution gas to the plant
    ch4_frac, rho_ch4 = 0.70, 0.678              # methane mole fraction of the gas, kg/m3 at standard conditions
    comp_kw, grid = 450.0, 0.48                  # compression load; grid intensity t CO2e per MWh (illustrative)
    co2_per_m3_fuel, fuel_per_kwh = 2.25, 0.28   # kg CO2 per m3 of fuel gas; m3 of fuel per kWh from a gas engine
    boe = oil * 6.2898 + gas * 35.3147 / 6000    # barrels of oil equivalent per day (6 Mcf per bbl)
else:
    oil, gas = 1600.0, 1.44e6 / 35.3147          # bbl/d oil, raw gas converted to m3/d for the calculation
    ch4_frac, rho_ch4 = 0.70, 0.678
    comp_kw, grid = 520.0, 0.62
    co2_per_m3_fuel, fuel_per_kwh = 2.25, 0.28
    boe = oil + gas * 35.3147 / 6000
pilot = 100.0                                    # m3/d of pilot and purge gas to the flare
upset = 0.005                                    # fraction of produced gas flared in upsets and maintenance
eff = 0.98                                       # combustion efficiency: 2 percent of the methane passes through unburned

def inventory(electric, air_pneumatics, vru, ldar):
    """Daily emissions by source, t CO2e."""
    t = {}
    fuel = 0.0 if electric else comp_kw * 24 * fuel_per_kwh
    t["compression"] = comp_kw * 24 / 1000 * grid if electric else fuel * co2_per_m3_fuel / 1000
    t["heater treater"] = 900.0 * co2_per_m3_fuel / 1000      # 900 m3/d of fuel gas in both designs
    tank = 0.0 if vru else oil * (1 if SI else 1 / 6.2898) * 1.8  # sour tank flash gas, m3 per m3 of oil: flared, never vented
    flared = pilot + upset * gas + tank
    t["flare CO2"] = flared * 2.6 / 1000                         # kg CO2 per m3 of rich gas burned (illustrative)
    t["flare methane slip"] = flared * ch4_frac * rho_ch4 * (1 - eff) * GWP / 1000
    pneu = 0.0 if air_pneumatics else 30 * 0.17 * 24             # 30 intermittent gas-driven controllers, m3/h each
    t["pneumatic venting"] = pneu * ch4_frac * rho_ch4 * GWP / 1000
    fug = gas * (0.0005 if ldar else 0.0015)                      # fugitive leakage as a fraction of throughput
    t["fugitives"] = fug * ch4_frac * rho_ch4 * GWP / 1000
    return t

for name, args in (("conventional", (False, False, False, False)), ("low-emission", (True, True, True, True))):
    t = inventory(*args); total = sum(t.values())
    ch4 = (t["flare methane slip"] + t["pneumatic venting"] + t["fugitives"]) / GWP   # t CH4 per day
    print(f"{name} design: {total:.1f} t CO2e per day, {total * 365 / 1000:.1f} kt per year; intensity {1000 * total / boe:.1f} kg CO2e per boe" +
          (f" ({1000 * total / (boe / 6.2898):.0f} kg per m3 OE)" if SI else ""))
    gas_ch4_t = gas * ch4_frac * rho_ch4 / 1000
    print(f"   methane {ch4:.3f} t/d = {100 * ch4 / gas_ch4_t:.2f} percent of the methane produced")
    for k, v in sorted(t.items(), key=lambda kv: -kv[1]):
        print(f"   {k:20s} {v:7.2f} t CO2e/d")

conventional design: 14.5 t CO2e per day, 5.3 kt per year; intensity 7.9 kg CO2e per boe
   methane 0.094 t/d = 0.49 percent of the methane produced
   compression             7.86 t CO2e/d
   heater treater          2.02 t CO2e/d
   flare CO2               1.98 t CO2e/d
   pneumatic venting       1.63 t CO2e/d
   fugitives               0.81 t CO2e/d
   flare methane slip      0.20 t CO2e/d
low-emission design: 10.9 t CO2e per day, 4.0 kt per year; intensity 5.9 kg CO2e per boe
   methane 0.013 t/d = 0.06 percent of the methane produced
   compression             7.74 t CO2e/d
   heater treater          2.02 t CO2e/d
   flare CO2               0.79 t CO2e/d
   fugitives               0.27 t CO2e/d
   flare methane slip      0.08 t CO2e/d
   pneumatic venting       0.00 t CO2e/d

The conventional design emits about 14.5 metric tons of CO₂ equivalent a day, 5.3 kilotons a year, an intensity of 7.9 kg per barrel of oil equivalent, with methane losses of 0.49 percent of the methane produced. The low-emission design cuts this to 10.9 metric tons a day, 5.9 kg per barrel of oil equivalent, and reduces methane losses almost to zero, 0.06 percent. The largest reductions come from eliminating pneumatic venting, recovering the tank vapors instead of flaring them, and the leak program. Electric compression barely helps, about 0.1 metric tons a day, because the regional grid's intensity is high enough that burning gas on site emits almost as much as buying electricity.

The low-emission design costs about 1.2 million dollars more in capital, mostly for the vapor recovery unit and the electrical supply, and that amount is already in the battery estimates of Lesson 9. The team includes it in every option, for three reasons. Venting sour vapors is unacceptable anyway, so the vapor recovery unit replaces a flare duty rather than adding a new one. Methane rules in both jurisdictions are tightening: they would require most of these measures within a few years in any case, and retrofitting is more expensive than building in. And the recovered gas has value. The emissions intensity is therefore almost the same for all three options, and it is not a major differentiator between them, though option A produces slightly fewer barrels for similar fixed emissions and so has a slightly higher intensity.

Land, water and reclamation

The development's footprint is small: two well pads, the battery site, the pipeline rights of way and an access road. The team routes the pipelines along existing rights of way where possible, avoids wetlands and watercourses, and plans to bore under the one creek that the gas line must cross rather than trench through it. The battery is sited to keep the emergency planning zone away from as many residences as possible and to allow a buffer from the nearest farmstead. Water for injection make-up comes from a saline aquifer, and produced water is reinjected into the reservoir, so the project uses no fresh water for injection. At the end of the field's life, the wells will be abandoned and the sites reclaimed to equivalent land capability, and the cost of doing so is included in the economics.

Working with the people affected

The stakeholders are the surface owners whose land the wells, battery and lines will occupy; residents near the sour facilities; the mineral owners, who will share in any unit; the county, which maintains the roads and provides emergency services; and the midstream company. North Dakota law requires operators to notify surface owners before drilling and to compensate them for damage and disruption, and the county has its own road-use requirements for heavy traffic. Daniel's plan starts with early, informal meetings before any permit is filed, so that concerns about routes, truck traffic, flaring and sour gas can shape the design rather than arrive as complaints.

The plan identifies what the project can offer and what it cannot. It can offer early information, routes and site locations adjusted where feasible, a pipeline instead of truck traffic, no routine flaring, and a named contact who answers calls. It cannot offer to remove all risk, and it should not pretend to. Lena's experience is that people affected by a sour gas development want to be told honestly what could happen and what will be done about it, and that they remember whether the company kept its word.