Ecology basics and capstone: biodiversity, evolution and conservation

Evolution explains where the diversity of life came from; ecology explains how organisms interact with each other and their environments today. The two are tightly linked: ecological interactions are the source of natural selection, and evolutionary history shapes which species live where. This final lesson introduces the basic ideas of ecology, from populations to ecosystems, and then brings the course together in a capstone: analysing an invented biodiversity survey and placing it in the context of conservation law and practice. BIOL-208 (Principles of Ecology), BIOL-367 (Conservation Biology) and BIOL-440 (Boreal and Northern Ecology) build on this lesson.

Levels of ecology

Ecology is studied at nested levels. A population is the members of one species in an area. A community is all the populations of different species living and interacting in an area. An ecosystem is the community together with its physical environment: soil, water, climate and nutrients. A biome is a major type of ecosystem defined by climate and dominant vegetation, such as tundra, boreal forest, temperate deciduous forest, grassland or desert. The biosphere is all life on Earth.

Population growth

Populations grow when births and immigration exceed deaths and emigration. With unlimited resources, a population grows exponentially, at a rate proportional to its size: \(\frac{dN}{dt} = rN\), where \(r\) is the per-capita growth rate. Exponential growth cannot last, because resources run out. Populations then tend toward the carrying capacity, \(K\), the largest population the environment can sustain. The logistic model captures this slowing:

The fraction \((K - N)/K\) is the share of the carrying capacity still unused. In words: growth is fast when the population is small relative to \(K\), and slows to zero as the population approaches \(K\).

# Exponential versus logistic growth, stepped one year at a time
r, K = 0.4, 1000.0
n_exp = n_log = 20.0
print("year  exponential  logistic")
for year in range(0, 21, 4):
    print(f"{year:4d}  {n_exp:11,.0f}  {n_log:8,.0f}")
    for _ in range(4):
        n_exp += r * n_exp
        n_log += r * n_log * (1 - n_log / K)

year  exponential  logistic
   0           20        20
   4           77        74
   8          295       245
  12        1,134       603
  16        4,356       902
  20       16,734       985

Species differ in their life-history strategies. Some, called r-selected, produce many small offspring with little care and boom in unstable environments: mosquitoes, dandelions, mice. Others, called K-selected, produce few large offspring with much care and live near carrying capacity: elephants, whales, oaks, humans. Most species fall between these extremes. K-selected species recover slowly from losses, which is why large, slow-breeding animals such as whales, caribou and sea turtles are often the most vulnerable to overharvest.

Interactions between species

Species interact in several ways, classified by whether each partner gains (+), loses (−) or is unaffected (0).

Predator–prey relationships can produce cycles. The classic case is the snowshoe hare and the Canada lynx: records of pelts bought by the Hudson's Bay Company over nearly two centuries show both populations rising and falling on a cycle of about ten years, with lynx peaks lagging behind hare peaks. The Lotka–Volterra equations, developed in the 1920s, model this: prey grow on their own but are eaten in proportion to encounters with predators; predators grow by eating prey and die at a constant rate.

# Lotka-Volterra predator-prey model, Euler steps (illustrative parameters)
a, b = 1.0, 0.02      # hare growth rate; predation rate per encounter
c, d = 0.01, 0.8      # lynx growth per hare eaten; lynx death rate
hares, lynx = 60.0, 20.0
dt = 0.01
peaks = []
prev_h, rising = hares, True
for step in range(1, 3001):
    dh = (a * hares - b * hares * lynx) * dt
    dl = (c * hares * lynx - d * lynx) * dt
    hares, lynx = hares + dh, lynx + dl
    if rising and hares < prev_h:
        peaks.append(round(step * dt, 1))
    rising = hares >= prev_h
    prev_h = hares
    if step % 500 == 0:
        print(f"t = {step*dt:4.1f}: hares {hares:6.1f}, lynx {lynx:5.1f}")
print("hare peaks at times:", peaks)

t =  5.0: hares   26.7, lynx  60.4
t = 10.0: hares  179.4, lynx  62.5
t = 15.0: hares   57.0, lynx  19.0
t = 20.0: hares   25.0, lynx  62.7
t = 25.0: hares  191.3, lynx  55.1
t = 30.0: hares   49.8, lynx  18.6
hare peaks at times: [2.3, 9.8, 17.4, 24.9]

The model is far simpler than reality. Real hare cycles also depend on food plants, stress hormones in hares and other predators such as owls and coyotes, and studies in Yukon's Kluane region showed that food and predation together drive the cycle. Simple models are still useful because they reveal the core mechanism: delayed feedback between predator and prey.

Energy and nutrients in ecosystems

Energy flows through ecosystems in one direction, entering as sunlight captured by producers (plants, algae and cyanobacteria) and passing to primary consumers (herbivores), secondary consumers (carnivores) and decomposers. At each step, most energy is lost as heat through respiration; on average only about 10 per cent passes to the next level. That is why food chains rarely have more than four or five links, and why there are far fewer wolves than moose and far fewer moose than plants.

Nutrients, by contrast, cycle. Carbon is fixed by photosynthesis and released by respiration, decomposition and burning. Nitrogen gas, abundant in air but unusable by most organisms, is fixed by bacteria (lesson 7), taken up by plants, passed through food webs and returned to the air by denitrifying bacteria. Phosphorus cycles mostly through rocks, soils and water. Human activity has greatly altered these cycles: burning fossil fuels adds carbon dioxide to the atmosphere, and fertilizer runoff overloads lakes with nitrogen and phosphorus, fuelling algal blooms.

Some species have effects far out of proportion to their abundance. These keystone species include sea otters, which eat sea urchins and so protect kelp forests along the Pacific coast; beavers, which create wetlands that support many other species; and wolves, whose return to Yellowstone National Park in 1995 changed how elk used the landscape, though the size of the effects on vegetation is still debated.

Measuring biodiversity

Biodiversity includes genetic diversity within species, species diversity and ecosystem diversity. Species diversity has two parts: richness, the number of species, and evenness, how equally individuals are spread among them. Two indices combine both. Simpson's index gives the probability that two individuals picked at random belong to different species: \(D = 1 - \sum p_i^2\). Shannon's index, \(H = -\sum p_i \ln p_i\), comes from information theory. In both, \(p_i\) is the proportion of individuals belonging to species \(i\).

Capstone: an invented biodiversity survey

The capstone brings together ideas from the whole course. Imagine a field team surveys birds along standardized routes in two habitats near a boreal town: an intact older forest and a forest regenerating after logging 15 years ago. The counts below are invented for teaching, but the methods are those used in real surveys.

import math
# Invented bird counts from equal survey effort in two habitats
old = {"boreal chickadee": 22, "red-breasted nuthatch": 18, "brown creeper": 12, "golden-crowned kinglet": 25,
       "black-backed woodpecker": 6, "Canada jay": 9, "bay-breasted warbler": 11, "white-winged crossbill": 14}
regen = {"white-throated sparrow": 48, "chestnut-sided warbler": 20, "American robin": 15,
         "dark-eyed junco": 12, "Canada jay": 4, "golden-crowned kinglet": 3}

def indices(counts):
    n = sum(counts.values())
    p = [c / n for c in counts.values()]
    simpson = 1 - sum(x * x for x in p)
    shannon = -sum(x * math.log(x) for x in p)
    evenness = shannon / math.log(len(p))
    return n, len(p), simpson, shannon, evenness

for name, counts in (("older forest", old), ("regenerating", regen)):
    n, s, d, h, e = indices(counts)
    print(f"{name:<13} birds {n:3d}, species {s}, Simpson {d:.3f}, Shannon {h:.3f}, evenness {e:.2f}")
shared = set(old) & set(regen)
print("species shared:", sorted(shared))

older forest  birds 117, species 8, Simpson 0.853, Shannon 1.991, evenness 0.96
regenerating  birds 102, species 6, Simpson 0.702, Shannon 1.439, evenness 0.80
species shared: ['Canada jay', 'golden-crowned kinglet']

The older forest has more species and a more even community, so both indices are higher; the regenerating stand is dominated by one species, the white-throated sparrow, typical of young, shrubby habitat. But the regenerating stand is not worthless: it supports species, such as chestnut-sided warblers, that the older forest lacks. A landscape with a mix of habitat ages supports more species overall than either alone, an idea ecologists call beta diversity, the turnover of species between habitats.

Next, the team estimates the population of a species of concern, using mark–recapture. In a first session they capture, band and release a number of birds; later they capture a second sample and count how many are already banded. If marked birds mix randomly, the proportion marked in the second sample estimates the proportion marked in the whole population, giving the Lincoln–Petersen estimate \(N \approx \frac{M \times C}{R}\), where \(M\) is the number marked, \(C\) the number caught in the second session and \(R\) the number of those already marked.

import numpy as np
# Lincoln-Petersen estimate, checked against a simulated population of known size
rng = np.random.default_rng(12)
true_N = 400
M, C = 60, 80
estimates = []
for trial in range(1000):
    marked = np.zeros(true_N, dtype=bool)
    marked[rng.choice(true_N, M, replace=False)] = True
    second = rng.choice(true_N, C, replace=False)
    R = marked[second].sum()
    if R > 0:
        estimates.append((M + 1) * (C + 1) / (R + 1) - 1)   # Chapman's bias-corrected version
est = np.array(estimates)
print(f"one field result: M = 60, C = 80, R = 12 -> N = {60 * 80 / 12:.0f}")
print(f"simulation: mean estimate {est.mean():.0f}, 95% of estimates between "
      f"{np.percentile(est, 2.5):.0f} and {np.percentile(est, 97.5):.0f} (true N = {true_N})")

one field result: M = 60, C = 80, R = 12 -> N = 400
simulation: mean estimate 398, 95% of estimates between 273 and 617 (true N = 400)

The simulation shows that a single mark–recapture study can easily be off by a quarter or more, so real studies repeat sessions and report confidence intervals. Finally, the team asks how much habitat must be protected. The species–area relationship, one of ecology's oldest patterns, says that the number of species \(S\) found in an area \(A\) grows as \(S = cA^z\), with \(z\) typically between about 0.15 and 0.35. It implies that losing habitat loses species, though not in direct proportion.

# Species-area relationship S = c * A^z: expected species remaining after habitat loss
z = 0.25
for remaining in (1.0, 0.5, 0.25, 0.10):
    fraction_species = remaining ** z
    print(f"{remaining:5.0%} of habitat left -> about {fraction_species:5.1%} of species expected to persist long term")

 100% of habitat left -> about 100.0% of species expected to persist long term
  50% of habitat left -> about 84.1% of species expected to persist long term
  25% of habitat left -> about 70.7% of species expected to persist long term
  10% of habitat left -> about 56.2% of species expected to persist long term

Losing half the habitat is expected eventually to cost about 16 per cent of species, and losing 90 per cent to cost about 44 per cent. The losses often arrive with a delay, called an extinction debt, as small isolated populations dwindle through drift, inbreeding (lesson 3) and chance events. Keeping habitat connected, so that gene flow and recolonization can continue, is therefore as important as its total area.

Conservation law and species at risk

Both countries protect threatened species by law, with independent scientific bodies assessing status and governments deciding on legal listing and protection.

In Canada, the Species at Risk Act (SARA, 2002) protects listed wildlife species. The independent Committee on the Status of Endangered Wildlife in Canada (COSEWIC) assesses species as extinct, extirpated, endangered, threatened or of special concern, and the federal government decides whether to add them to Schedule 1 of the Act. Listing triggers prohibitions on killing or harming individuals and their residences, and requires recovery strategies that identify critical habitat. The Act applies directly on federal lands and to aquatic species and migratory birds; elsewhere, provinces and territories have the main responsibility, and the federal minister can recommend an emergency order when a species faces imminent threats. The monarch butterfly was listed as endangered under SARA in December 2023. Woodland caribou are listed in several populations, with boreal caribou threatened; in 2024 the federal environment minister formed the opinion that boreal caribou in Quebec face imminent threats, but in a statement published on 30 June 2026 the government chose to pursue a collaborative approach with Quebec and First Nations rather than an emergency order at that time.

Conservation increasingly draws on Indigenous knowledge alongside Western science. Indigenous peoples have observed and managed ecosystems for thousands of years, and their knowledge of species, seasons and landscapes often extends further back in time than scientific records. In Canada, the Mi'kmaw concept of Etuaptmumk, or Two-Eyed Seeing, articulated by Elder Albert Marshall, describes learning to see with the strengths of Indigenous knowledge in one eye and Western knowledge in the other, and Indigenous Guardians programs employ community members to monitor lands and waters. In the United States, many tribal natural resource departments manage fisheries and wildlife, and federal agencies have issued guidance on including Indigenous knowledge in decisions. Respectful collaboration means communities decide how their knowledge is shared and used.

Bringing it together

The capstone shows how the course's ideas connect. The bird species in the survey are products of natural selection (lesson 2) and speciation (lesson 4) over millions of years, related as branches of a tree (lesson 6). Their populations are shaped by competition, predation and habitat. Their persistence depends on keeping populations large and connected enough to resist drift and inbreeding (lesson 3). And the law and ethics of conservation decide what is protected. Evolution is not just the history of life: it is a process continuing around us, in antibiotic-resistant bacteria, changing flowering times, and the species we are trying to save.