Capstone: a management plan for an invented boreal region

This capstone brings the course together. The Sable River team has been asked to help a joint planning board draft a 50-year ecological management plan for the watershed around their station and the caribou range to its north. The board brings together the local Indigenous government, the provincial, territorial or state government, a forestry company, an energy company that holds exploration rights, and the station's scientists. The task is to describe the system, identify what people value in it, project how it might change under different choices, and propose management that is robust to an uncertain future. Every number here is synthetic, but the reasoning is the kind used in real northern planning.

The planning area

The planning area covers 6,000 square kilometers (about 2,300 square miles) of the invented Sable Creek watershed in interior Alaska: black spruce over permafrost on north slopes and lowlands, aspen, birch and white spruce on south slopes, fens and bogs over 30 percent of the area, and many lakes. The northern 3,500 square kilometers overlap the winter range of a migratory caribou herd and the range of a small resident woodland herd. State forest land covers the southern half, a highway and powerline run east–west across the middle, and an energy company holds exploration leases in the northeast. Alaska Native village and regional corporations own several large parcels, and village residents hunt, trap and fish across the area under state and federal subsistence regulations. The plan must coordinate state, federal and Native corporation landowners, which in Alaska is often the hardest part of landscape planning.

Step 1: what matters

Planning starts with values, not models. Through workshops and interviews, the board identified valued components: a self-sustaining caribou population; moose, fish and furbearers for harvest; clean water and healthy fish; peatland and forest carbon; a sustainable timber supply and local jobs; safety of the community and infrastructure from wildfire; cultural sites and travel routes; and room for economic development. Each valued component needs an indicator that can be measured and projected, and a target or threshold that signals concern. Some targets come from law or science, such as the 65 percent undisturbed habitat threshold for boreal caribou; others are set by the board, such as maintaining at least 20 percent of the forest older than 120 years.

Step 2: the current state

The team assembled the evidence. Fire history from lesson 4 gives a fire cycle of roughly 85 to 90 years under recent climate. The stand-age map shows 34 percent of the forest older than 120 years. In the caribou range, burns under 40 years old cover 18 percent and human features buffered by 500 meters cover 24 percent, with some overlap, so total disturbance is about 39 percent and the undisturbed share about 61 percent, just below the threshold. Collared caribou show adult female survival of 0.88 and recruitment of 25 calves per 100 cows, a slowly declining population with a growth rate of about 0.99. Peatlands store an average of about 150 kilograms of carbon per square meter, and permafrost persists only in scattered peat plateaus, which have shrunk over the past 50 years. Guardians report that moose have increased near the road and that wolves now use the seismic lines in winter.

Step 3: scenarios

The future is uncertain, so the team projected the system under alternative management options and climate futures, rather than trying to predict a single outcome. Three management options were defined. Business as usual continues current harvest and allows exploration to proceed with standard mitigation. Conservation focus protects the northern peatlands and caribou range from new industrial disturbance, restores 1,500 kilometers of seismic lines over 20 years and concentrates harvest in the south. Balanced uses designated zones for development with strict limits on new linear features, restores half as many lines and reduces harvest in the caribou range. Each option was run under a moderate and a hot climate future, the latter with more frequent fire.

# 50-year scenario model for the Sable caribou range (synthetic, illustrative).
import numpy as np
def project(option, climate, runs=400, years=50, seed=1212):
    rng = np.random.default_rng(seed)
    fire_rate = {'moderate': 0.0045, 'hot': 0.012}[climate]          # fraction of range burning per year
    new_dist = {'business as usual': 0.006, 'balanced': 0.0015, 'conservation': 0.0}[option]
    restore = {'business as usual': 0.0, 'balanced': 0.003, 'conservation': 0.006}[option]
    out = []
    for _ in range(runs):
        burned, human = 0.18, 0.24                                  # starting disturbed shares (with overlap below)
        for t in range(years):
            fire = rng.gamma(0.3, fire_rate / 0.3)                    # variable annual area burned
            burned = min(burned + fire * (1 - burned) - burned / 40, 0.95)   # burns recover after ~40 years
            human = max(human + new_dist - (restore if t < 20 else 0), 0.05)
        total = 1 - (1 - burned) * (1 - human)
        recruit = 36 - 0.24 * 100 * total                            # calves per 100 cows (lesson 7 relationship)
        lam = 0.88 + 0.5 * recruit / 100 * 0.88                     # female survival plus female recruits
        out.append((total, lam))
    a = np.array(out)
    return a[:, 0].mean(), np.mean(a[:, 0] <= 0.35), np.median(a[:, 1]), np.mean(a[:, 1] >= 1.0)
print(f"{'option':<18}{'climate':<10}{'disturbed':>10}{'P(meets 65%)':>14}{'lambda':>8}{'P(lambda>=1)':>14}")
for opt in ('business as usual', 'balanced', 'conservation'):
    for clim in ('moderate', 'hot'):
        d, p, l, pl = project(opt, clim)
        print(f"{opt:<18}{clim:<10}{100 * d:>9.0f}%{p:>14.2f}{l:>8.3f}{pl:>14.2f}")

option            climate    disturbed  P(meets 65%)  lambda  P(lambda>=1)
business as usual moderate         61%          0.00   0.974          0.00
business as usual hot              68%          0.00   0.967          0.00
balanced          moderate         37%          0.10   0.999          0.33
balanced          hot              48%          0.00   0.987          0.00
conservation      moderate         26%          1.00   1.011          1.00
conservation      hot              39%          0.21   0.997          0.33

The projections show the logic of robust decision-making. Under business as usual, disturbance rises well above the threshold in both climates and the caribou population almost certainly declines. Conservation focus brings disturbance well below the threshold under a moderate climate and gives a stable or growing herd in nearly every run, but in the hot future it meets the threshold in only about a fifth of runs and gives a stable herd in about a third, because fire alone pushes disturbance up. The balanced option sits between them, failing the threshold in most runs even under a moderate climate. No option guarantees success under the hot climate, which tells the board two things: managing human disturbance is necessary but not sufficient, and the plan must include actions for fire, such as protecting key lichen winter ranges where feasible, and monitoring that would trigger further action.

Trade-offs must be made explicit. The conservation option reduces the timber available from the northern half of the license and defers some exploration, with costs to jobs and revenue that the board must weigh, perhaps offset by restoration employment, guardian programs and carbon finance for protected peatlands. Uncertainty must be communicated honestly: the projections rest on a disturbance–recruitment relationship borrowed from other ranges, on assumed fire rates and on a highly simplified population model. The team therefore presented ranges and probabilities rather than single numbers, ran the model again with different fire rates and recruitment slopes to check whether the ranking of options changed, and explained the limits of the model in plain language at community meetings, alongside the guardians' observations of the same landscape.

Step 4: the plan

Drawing on the projections, the board's draft plan combines measures across the course. It protects the northern peatlands and core caribou range from new industrial disturbance, consistent with the proposed Indigenous protected area; restores seismic lines in the caribou range by mounding, planting and line blocking, prioritizing lines that wolves use most according to collar data; and concentrates harvest in the south using natural disturbance-based patterns that retain old-forest patches and snags. It sets a cap on new linear features and requires shared roads and narrow, low-impact seismic methods in the development zone. For fire, it supports FireSmart or Firewise treatments around the community and infrastructure, revives cultural burning in spring around the community with the Indigenous government's fire program, and designates priority caribou winter range for initial attack when fires threaten it. For fish and water, it requires baseline and ongoing mercury monitoring before any new impoundment or road crossing, and for carbon it adopts no net loss of intact peatland.

Implementation depends on the state forest plan and permitting, the cooperation of Native corporations as major landowners, federal agencies for subsistence and migratory caribou management, and agreements with the energy leaseholder. Tribal councils and the regional caribou working group would share in monitoring and in setting harvest recommendations, and federal and state fire agencies would assign protection levels consistent with the plan's priorities.

Step 5: monitoring and adaptive management

The plan is adaptive: it treats management as an experiment, monitors outcomes and adjusts. The monitoring program follows the design principles of BIOL-430. Satellite imagery tracks disturbance and fire annually; guardians and scientists jointly survey caribou recruitment each March and collar a sample of cows; wolf collars and trail cameras on restored and unrestored lines test whether restoration reduces wolf use, using a before–after–control–impact design; stand plots track regeneration after fire and harvest; permafrost and peatland plots track thaw; and fish mercury is sampled every three years. The board agreed on triggers in advance: if recruitment stays below 25 calves per 100 cows for three consecutive years, or disturbance exceeds 45 percent, the board will consider further measures, such as maternal penning or temporary harvest closures. Data are shared under agreed governance principles, OCAP or CARE, with results reported to the community each year.