Scientific Evidence & Data Tables – Detailed Study Notes

Definition of PPM (and related units)
- “Parts per million” = out of every $1{,}000{,}000$ equal‐sized mass units, one unit is the substance of interest; the remaining $999{,}999$ are something else (soil, water, tissue, etc.).
- Analogous units: parts per thousand (ppt), parts per billion (ppb), etc.
- Mass unit is arbitrary: grams, kilograms, pounds, ….
Conceptual ratio
- Written generically as
  $\text{PPM value} : 1{,}000{,}000$
  e.g.
  $5 \; \text{ppm Fe} \;\Longrightarrow\; 5\ \text{g Fe} : 1{,}000{,}000\ \text{g soil}$
Dimensional-analysis approach
1. Write the total mass you have.
2. Multiply by unit conversions so that unwanted units cancel.
3. Introduce the PPM ratio as a fraction $\displaystyle \frac{\text{mass of solute}}{\text{mass of medium}}$ .
4. Cancel common units & numbers; the remainder is the desired mass.
Fully worked classroom example (Iron in soil)
- Given: $1{,}000\ \text{kg soil}$ , iron concentration $5\ \text{ppm}$ .
- Convert $1{,}000\ \text{kg} \times \frac{1{,}000 \ \text{g}}{1 \ \text{kg}} = 1{,}000{,}000\ \text{g}$ of soil.
- Apply PPM ratio:
  $1{,}000{,}000\ \text{g soil} \times \frac{5\ \text{g Fe}}{1{,}000{,}000\ \text{g soil}} = 5\ \text{g Fe}.$
- Kilograms, grams of soil, and the numeric $1{,}000{,}000$ cancel, leaving only grams of Fe.
- $m_{\text{solute}}$ = mass of contaminant (g).
- $C_{\text{ppm}}$ = concentration in ppm.
- $m_{\text{total}}$ = total mass of medium (g).

Figure in video illustrates biomagnification of toxic mercury (Hg) up the Arctic food web.
- Organic matter & water: <0.02\ \text{ppm} Hg.
- Copepods: $0.06\ \text{ppm}$ Hg.
- Arctic cod: $0.19\ \text{ppm}$ Hg.
- Harp seal: $2.25\ \text{ppm}$ Hg (highest in chain).
Exercise prompt
- Harp seal body mass $=120\ \text{kg}$ .
- Students asked to compute Hg mass (g) via dimensional analysis.
- Solution path (left to student on Canvas):
  $120\ \text{kg} \times \frac{1{,}000\ \text{g}}{1\ \text{kg}} \times \frac{2.25\ \text{g Hg}}{1{,}000{,}000\ \text{g seal}} \approx 0.27\ \text{g Hg}.$

Alternatives: embed data in text or use graphical figures.
Advantages of tables
- Compact & efficient display of many numbers.
- Precise values easily re-used by others (vs. estimating from a graph).
- Natural grouping allows side-by-side comparison (rows/columns invite comparison).
- Handy for summary statistics (mean, median, range, SD) split into categories.
- Can juxtapose data with reference criteria or standards.
Disadvantages
- Less immediate visual impact than a good figure.
- Dense tables can overwhelm casual readers.

Context
- Maryland had a moratorium on unconventional oil & gas development (hydraulic fracturing, “fracking”).
- Neighboring Pennsylvania allowed fracking in the Marcellus Shale region.
- Research question: Does fracking increase metal contamination in streams?
Field work
- Streams sampled in southwestern Pennsylvania (fracked) and adjacent Maryland counties (non-fracked).
- Lead author + undergraduate researcher Alexandra Masker collected water and measured metal concentrations.
Hypotheses & predictions
- H₁ (contamination hypothesis): Fracking causes elevated pollutant concentrations.
- Prediction: Metals higher in Pennsylvania streams than Maryland streams.
- H₀ (null / alternate given in lecture): No difference between fracked and non-fracked watersheds.
- Students asked to consider hidden mechanisms whereby fracking might pollute yet no concentration difference is observed (e.g., dilution, sampling timing, unmeasured pollutant forms).

Summarize existing knowledge
- Example Table: Chemistry of produced/flowback water from five prior studies (Barbot et al., Hayes et al., …)
- Columns list pH, specific conductance, radium, numerous metals.
- Allows quick comparison of typical contaminant ranges.
Present summary statistics of new data
- Watershed characteristics table
- Mean watershed area (Maryland $90\ \text{km}^2$ vs Pennsylvania $92\ \text{km}^2$ ).
- Oil & gas exposure metrics: $0$ in MD, significantly greater in PA (asterisk indicates p<0.05).
- Sensitivity indices, background pollution sources, well densities (unconventional & conventional), coal-mine density.
Compare observations to criteria/standards
- EPA water-quality criteria table
- Acute aquatic life limits, chronic limits, human drinking-water limits.
- Counts of samples exceeding limits: presented as $n<em>{MD}, n</em>{PA}$ for each analyte.
- Sample size differences captured (e.g., 10 MD samples, 19 PA; subset analyzed for certain metals).

Metals & example numeric thresholds
- Arsenic: drinking-water $10\ \mu\text{g L}^{-1}$ .
- Barium: drinking-water $2{,}000\ \text{mg L}^{-1}$ .
- (All criteria listed in table; values illustrated verbally in lecture.)
Acute criteria = immediate toxicity; chronic = long-term exposure.
Chronic limits are always < acute limits.

Clear labels & units on every column/row.
Compact: minimal empty space, no redundancy.
Logical structure aligns with research questions.
- Comparable groups placed adjacent.
- Hierarchical or categorical ordering intuitive.
Contains more quantitative detail than text alone can provide efficiently.
Referenced explicitly in manuscript (e.g., “…see Table 1”).
Formatting (fonts, borders, light shading) used sparingly to enhance clarity, not distract.

Principle: The simplest explanation that accounts for all facts is preferred until evidence dictates otherwise.
Practical application
- Begin with parsimonious hypotheses.
- Retain openness to complex or “unlikely” mechanisms if (and only if) simpler ones fail to match data.