We Need More Science - and Less Spin - on Selenium and Coal Mining
A recent study by Cooke et al raised alarms that fish in Alberta remain contaminated from coal mining that ended decades ago, and that reopening those mines would push the ecosystem “beyond recovery.” But the study explicitly states that no clinical symptoms of selenium poisoning were observed. And what it really shows is that reopening these legacy mines under modern regulations may actually improve water quality and reduce ecosystem risk.
Moreover, the title and press framing use the term "mountain top removal" mining as a term to strike fear into readers rather than a normal term like "above ground mining" or "open pit mining". The study showed that the sediments actually stopped increasing in concentrations when the mining method switched from underground to above ground. The inclusion of that term "mountain top removal" adds sensationalism without scientific relevance, and is a red flag for technical scientific publications.
Fear is the enemy of curiosity, and curiosity is the foundation for science. I’m not here to defend bad mining. I’m here to defend good science. Preprints should spark discussion, not panic.
Let’s start by looking at how this study was reported. Fourteen outlets published stories… all on the same day. Scratch the surface, and it’s the same article syndicated across all of them. The full repetitive outlet list is included at the bottom of this post.
Now, back to the study. As a PhD environmental scientist with 17 years of experience working directly on selenium monitoring, treatment, and regulation (and as a member of the North American Selenium Working Group) this is my field. I approached this paper the way I would peer-review an academic article.
What the media doesn’t show you, and what the authors don’t explain (line items and figures refer to the un-reviewed pre-print article):
The sediment selenium concentration increases between 1930–1950 when there was only underground mining, but it stops increasing once the above ground mining started (referred to in this paper as “mountain top removal”) (Figure 1 ).
There is no historic water quality information to correlate to the sediments or impacts (lines 171 – 173).
The study clearly states that “clinical expression of selenium poisoning was not observed in our assessment” (lines 203-204).
As recently as 2021, inputs to the lake were as high as 24 µg/L selenium (Line 183 and Figure 1B). This is crucial context when discussing historic mine sites. If reopened today, these sites would be held to today’s stricter standards, which require much lower concentrations, thereby improving water quality. This is achieved through proven water treatment process (an example list from 2020, which has grown significantly since).
This matters because if these historic mines were reopened today, they’d need to meet strict discharge criteria (typically 2-10x lower than the concentrations reported in this study) as outlined in the Canadian Council of Ministers of the Environment (CCME) Water Quality Guidelines and British Columbia Water Quality Guidelines (BCWQG), as well as CMER technical guidance for selenium. So reopening could actually decrease selenium concentrations in the system. Additionally, the Alberta Energy Regulator (AER) has published selenium management frameworks and expectations for coal projects that align with national guidelines, reinforcing the regulatory oversight that would be applied to any future activity.
The idea that “any new coal development will push fish beyond recovery” is not supported by the study’s own data. Nor do the authors account for fish migration, stocking, angling pressure, the presence of whirling disease, or other confounding stressors they themselves list.
A deeper look at the study’s shortcomings, modelling gaps, misused comparisons, and peer-review commentary follows.
Peer review is when other experts in the field scrutinize the methods, data, and conclusions before a study is accepted as scientific evidence. Without peer review, speculative conclusions can enter the public domain without being properly challenged, undermining both science and public trust.
This analysis is based on standard peer-review practice: line-by-line review, examination of methodology, and contextual scientific comparison. While the authors have raised important issues, the framing, extrapolations, and omissions risk undermining public trust in environmental science.
What we need now is more data, better modelling, and clearer distinctions between legacy impacts and future risk under modern regulation.
This is why we need strong, peer-reviewed science to guide environmental decision-making—not fear-based headlines drawn from a single unreviewed manuscript. Other scientists are also now reviewing the Cooke et al. preprint and reached similar conclusions about its scientific flaws. For example, a detailed review by Gilron (2025). You will see some similarities and some differences between our reviews, as Gilron's background is more deeply into ecotoxicology, while I'm more focused on biogeochemistry, so we have different realms of expertise to rely on.
Peer-Review Style Commentary Line items and figures refer to the un-reviewed pre-print article:
1. No Observable Toxic Effects
o Lines 203–204: The study notes that clinical symptoms of selenium toxicity were not observed. This aligns with what is seen in naturally selenium-elevated ecosystems (e.g., parts of Colorado).
2. Population Effects Not Proven
o Lines 199–200, 212–215: There are no comparisons provided for relevant reference lakes or stocking records. Low observed fish numbers may reflect stocking practices, migration patterns, or recreational angling pressure—not necessarily selenium-related mortality. This is especially relevant considering the absence of toxicity symptoms.
o Lines 228–230 mention other stressors (e.g., Whirling Disease, drought, angling pressure), but lines 231–232 leap to a broad, unvalidated claim: “Any new development of coal mining… may well push the fishery beyond recovery.” This lacks evidentiary support given that reopening brownfield mines today would trigger strict regulatory controls and treatment requirements.
3. Data vs. Assertions
o While the tissue and sediment data appear valid, the conclusions drawn overreach the evidence and ignore potential confounding variables.
o The fish tissue work is not my area of expertise, and the Gilron review goes into this in more detail (link above).
4. Lack of Water Chemistry Context
o Lines 171–173 and 176-177: The authors state that lake water wasn’t measured regularly and note the absence of longitudinal water chemistry data
o Line 183 and Figure 1B: Inputs up to 24 µg/L were recorded as recently as 2021. If the mines were reopened today, treatment standards would require lower discharges.
5. Sediment Background Levels and Reference Locations
o Lines 178–179: Sediment selenium exceeded 2 µg/g pre-mining, indicating natural background levels likely tied to local geology.
o No control or reference lakes with similar coal-bearing formations were included, making interpretation difficult.
o Figure 2: It’s unclear whether comparator lakes were exposed to similar geologic conditions. If not, they may be inappropriate baselines.
6. Sediment Stability Over Time
o Sediment selenium increased 1930–1950 (underground mining), then plateaued after surface mining began (called “mountaintop removal” in this study).
o Figure 1 shows stability through the operational and post-operational period, despite no treatment.
o The paper does not address redox-driven redistribution of selenium in sediments - a known factor in selenium mobility and concentration profiles.
7. Misuse of “Mountaintop Removal” Framing
o Despite the title and press framing, this is not a study of MTR, and it showed that the sediments actually stopped increasing in concentrations when the mining method switched from underground to above ground. The inclusion of that term adds sensationalism without scientific relevance.
Media coverage of the original article. All are actually the same one article, ‘syndicated’ to publish repeatedly at different outlets. Check out the links, or ask your favourite AI to check into it.
Vancouver Is Awesome, 06 Jun 2025 (now removed)
Squamish Chief, 06 Jun 2025 (now removed)