One year in one landscape
The project is ongoing: in this publication 0 of 12 months have a photograph. Before one exists, check what has to match between two visits for two frames to be comparable.
The blue band is the selected feature: 12% of the frame width, in the same place in both frames. The pole stands at the distance set below, the willow at 80 m, the spruce at 1500 m — these are model assumptions, not measurements taken at the site.
Shift in frame 2 — three layers at once, one scale (% of frame width)
-
Pole in the foreground 15 m
3.7% of frame width to the left — outside tolerance
-
Willow in the middle ground 80 m
0.7% of frame width to the left — within tolerance
-
Spruce on the treeline 1500 m
0.0% of frame width — within tolerance
Empty marker: position in frame 1. Filled marker: position in frame 2. Tolerance is the permitted error as a % of frame width.
Return settings
Without JavaScript you see a single state of the model: tripod shifted 40 cm to the right, focal length 50 mm, sensor 36 mm, pole at a distance of 15 m, tolerance 1.0%. General rule: a shifted tripod throws off mainly the foreground, while a rotated axis moves all layers by almost the same amount.
Selected feature — pole in the foreground (15 m): 3.7% to the left. The object is still within the crop, but outside the 1.0% tolerance — the comparison will carry an alignment error.
Model parameters
- Horizontal angle of view
- 39.6°
- Margin for tripod shift (pole, no rotation)
- ± 10.8 cm
- Margin for axis rotation (frame centre, no shift)
- ± 0.4°
Model: a rectilinear view without distortion, horizontal error only, no camera tilt. This illustrates the relationships; it is not a calibration of equipment.
Twelve months, none yet with a photograph
The tiles are of similar width; the number of days in a month is not shown by the drawing but given in the description of the selected month. Each tile contains an empty window with the feature selected above — the one that will have to be found in that month's frame. No tile has a photograph, so none is filled in, generated or copied from a neighbour.
Without JavaScript: all 12 months (January–December) are in the same state — no photograph in this publication. There is not yet any visit date.
January — no photograph in this publication
There is no photograph, visit date or description of the water here. We do not use neighbouring months (December, February) as a substitute. The empty window in the tile shows what will have to be found in that month's frame: pole in the foreground.
31 days · from here “Previous” leads to December of the previous cycle.
Off the axis: a contextual photograph with no month
Rozlewiska Biebrzy [Biebrza floodplain wetlands] near Brzostowo, photo by Adikpl, file on Wikimedia Commons, CC BY 3.0. This is contextual material carried over from an earlier version; it has no month assigned and does not come from a site covered by the protocol described here, which is why it is not on the strip. It is not embedded in the page, because the page is meant to work without a network connection.
Return protocol — five field records
- Site. Choose a point that is legally accessible and safe, with a permanent reference mark in the foreground. The frame is no reason to enter flooded ground or ice, nor to replace a failed visit with another place.
- Frame parameters. Record the place and time of the visit, the focal length, the camera format, the frame aspect ratio, the tripod height, the azimuth of the axis and the position of the horizon in the viewfinder; recording place, time, notes and photographs is described in source [1]. The azimuth is a note for recreating the setup, not a result calculated by this page.
- Return. Set the tripod at the reference mark, overlay a preview of the first photograph and record the visible alignment error as a % of frame width. The foreground reveals a shifted tripod, the distant forest a rotated axis.
- Fact separately. Record what you can see: the extent of the water, the colour of the plants, the weather. Record any supposition about the cause separately. Visible water is not a measurement of the river level, and a different colour in a photograph is not automatically a biological change (a methodological note by the page's author, not a conclusion from a source).
- A gap stays a gap. When there is no photograph, record the month as missing, with the reason: access, weather, time, equipment or an entry restriction. Do not replace it with a frame from a neighbouring visit.
Details of the calculations used above
W sensor width (nominal size), f - focal length FOV = 2 · atan( W / (2f) ) p = 0.5 + (f/W) · tan(θ) position in the frame (0 = left edge, 1 = right edge) tripod s to the right, object at distance d: tan θ' = tan θ − s/d axis rotated by δ to the right: θ'' = θ' − δ shift = (p'' − p) · 100 [% of frame width] tripod margin for a pole on the axis ≈ d · (W/f) · tolerance rotation margin at the frame centre = atan( tolerance · W/f )
The distances of the willow (80 m) and the spruce (1500 m) and the positions of the layers in frame 1 (22%, 50%, 78%) are assumptions of the schematic.
What the frame is meant to show, and what must not be read into it
The rhythm of the valley as a question
The park describes high spring water levels on the Biebrza linked to snowmelt, and summer-to-autumn low flows [3]. This is a general rhythm that does not describe every year identically. We therefore do not attribute flooding to a chosen month, and we do not turn a single photograph into a measurement of the river level. Fixed frame boundaries are a method proposed for this project, not a requirement of the park (a methodological note by the page's author, not a conclusion from a source).
Limits of this page
- No photo series: there is no visit date here, so there is no timeline with dates.
- The model is a schematic: the distances of the layers are assumed, and the lens has no distortion.
- Cropping a photograph or a different frame aspect ratio changes the effective angle of view, so two frames with different crops are not comparable without recalculation. This page does not model that.
- Agreement between frames proves nothing about the water, the plants or the weather.
- The sources from V2 have been carried over, not re-checked for this version.
What was removed from version V2, and why
- A 12-month notebook with saving in the browser and a .txt export: without photographs it stored declarations that looked like documentation. Replaced by the protocol above.
- A two-visit comparison tool with text fields: replaced by a simultaneous comparison of frames 1 and 2 on one scale.
- The 2024 and 2025 cycles: with no material for either year they added no knowledge, so they were dropped.
- Frame aspect ratio: in V2 it changed the effective sensor width, and with it the angle of view and the permitted shift. Without photographs it added no knowledge here, so the slider was removed; the effect of framing is described only in text, in the page's limits and in the protocol. The azimuth compass was decoration: the azimuth remains a note in the protocol.
- A photograph embedded from the web: replaced with an off-axis link so that the page works offline.
- Kept: the angle of view, the tolerance and the margin for tripod shift, the separation of fact from hypothesis, the explicit absence of a photograph, the sources.
Sources
Links carried over from version V2. They were not opened when this version was built, so their content and currency need to be checked.
- Obserwacja i badania terenowe — ZPE [Observation and fieldwork — ZPE]: recording place, time, notes and photographs. to be checked
- Nasza Biebrza nr 54, lato 2024 — BbPN [Our Biebrza, no. 54, summer 2024 — Biebrza National Park]: a year of landscape in the park. to be checked
- Wody — archiwalny serwis BbPN [Waters — archived Biebrza National Park website]: seasonality of runoff. to be checked
- Rozlewiska Biebrzy — Wikimedia Commons [Biebrza floodplain wetlands — Wikimedia Commons]: photo by Adikpl, CC BY 3.0, contextual material with no month. licence and description to be checked