River Pulse
Canal del Dique – Lower Magdalena: more rain downstream in October 2026, and Plato with an incomplete series in 2026
Written by Eduardo Garcia · Co-founder / CEO / CTOPublished 19 November 2026
#River-Pulse#Colombia#Canal-del-Dique#Lower-Magdalena
From 1 to 31 October 2025, peak stage at Plato was 11.15 m, on the 28th. In 2026 IDEAM’s instantaneous level feed only carries 26–31 October; the 1–25 gap is in that feed, so the maximum read stays at 9.39 m on the 30th. It is not a scored contraction pulse.
That does not match the rain in the same October at first glance. In 2026 more fell on the canal corridor and the plain: 208.3 mm and 214.0 mm, against 159.8 mm and 146.4 mm in 2025. Upstream, in the headwaters, less fell: 298.7 mm against 384.0 mm.
This River Pulse reading on Canal del Dique publishes the available data: satellite rain and soil on headwaters, valley and plain, radar water on the floodplain, and stage at four IDEAM gauges. That is not enough for a closed answer on the flood pulse, because that count needs an official flood mark in metres that is not published here.
Without that mark, 1 of 4 is neither a confirmed nor a denied flood pulse. It is not a dry pulse either. It is a half count that leaves the channel question open.
Anyone who lives in the valley needs to know. The rain they felt in the street may be the rain that moved the plain, and still the river at the gauges read can sit below the year before.
Why compare two years
“It rained a lot” and “the river rose” repeat every October on the Caribbean coast. One month alone does not say whether the whole basin moved together toward rise. You need the same period, the same basin and the same rules in two years running.
That is what River Pulse does. It reads four rise signs on the same dates in two years and asks whether they showed up together.
The four signs follow the path of the water toward rise: headwaters, soil, stage and plain. When all four close in the same period, there was a flood pulse.
On the Taquari Valley, with flood marks published by ANA, late August to late September 2026 closed 4 of 4 on flood pulse. Here, with the IDEAM mark missing at Plato, both readings stay at 1 of 4, and the river signal stays open.
Headwaters, stage and discharge
Three terms run through this reading.
- Headwaters
- The high part of the basin, upstream of Calamar. Here they are the headwaters of the Bajo Cauca and Middle Magdalena: about 38,612 km².
- Stage
- The height of the river at the gauge, in metres, read against a fixed staff. It is not depth. IDEAM publishes the stage series; on this network it does not publish the alert and flood stages other countries use to score the river signal.
- Discharge
- How much water passes each second, in cubic metres. Stage says whether water left the bed. Discharge says how much was passing. On these gauges discharge is not in IDEAM’s public data.
Where the reading was made
The headwaters add up to about 38,612 km² upstream of Calamar. The Canal del Dique corridor and south of Atlántico valley covers about 3,086 km². The canal plain, about 574 km². The union of the three areas, used for rain and soil by stretch, is about 41,425 km².
Plato is the reference gauge for this run. San Pedrito, El Banco and Gamarra complete the network read. None of the four sits on Canal del Dique: they are on the Magdalena. Calamar and Gambote appear in the IDEAM catalog, but the instantaneous level feed has no rows for them in these Octobers; they are left out of the reference. Gamarra only has a series in 2026.
Rain: less upstream in 2026, more on the corridor and the plain
In the headwaters 384.0 mm fell in 2025 and 298.7 mm in 2026. Both years pass the 180 mm mark. Neither had a day over 50 mm; the wettest day was 39.3 mm in 2025 and 31.2 mm in 2026.
Downstream the map changes pattern. In 2026 the strongest colour shifts to the canal corridor and the plain. In 2025 the weight of rain sat higher up. That explains part of what people nearer the river felt, even though it does not close the channel signal.
2025, 1 to 31 October. The scale is the same in both years. Orange marks mainly the headwaters and the north of the corridor.
2026, the same period and the same scale. Red concentrates on the canal corridor and the plain, while the headwaters stay lighter than in 2025.
The river: peaks read, and Plato incomplete in 2026
At Plato, full October 2025 reached 11.15 m on the 28th. Plato’s 1–25 October 2026 gap is in IDEAM’s published instantaneous level feed, so the maximum read stays at 9.39 m on the 30th. El Banco fell from 8.90 m on 31 October 2025 to 5.92 m on 24 October 2026 (the same peak repeats on the 26th), with nearly the full month in both years; the 25th is missing. San Pedrito went from 6.26 m on 31 October 2025 to 6.12 m on 1 October 2026. Gamarra only has a series in 2026: 4.59 m on 24 October.
The chart shows stage at Plato on the days in IDEAM’s October 2026 instantaneous level feed (26–31). It is not the whole month. It does not draw alert or flood lines, because those marks are not published for this gauge.
Plato, days in IDEAM’s instantaneous level in October 2026 (26–31). The blue line is stage in metres. The maximum read was 9.39 m on the 30th. Source: IDEAM.
Maximum stage read at each gauge. Plato 2026: 26–31 October only in IDEAM’s instantaneous level feed. Discharge is omitted: IDEAM’s public data do not carry it. Gamarra: no series in 2025.
| Station | 2025 (m) | 2026 (m) |
|---|---|---|
| Plato | 11.15 | 9.39 |
| San Pedrito | 6.26 | 6.12 |
| El Banco | 8.90 | 5.92 |
| Gamarra | — | 4.59 |
The plain: 226.26 km² with water in 2025 and 220.15 km² in 2026
On the canal plain, the area is the greatest extent radar saw on the passes in the period: 226.26 km² in 2025 (five passes) and 220.15 km² in 2026 (six passes). It is not a mean. The plain signal counts when it beats the prior year: in 2026 it did not; in 2025 there is no 2024 in this run, so the row stays No data.
Water outside the usual channel measured 38.30 km² in 2025 and 30.84 km² in 2026. That is the part the satellite record does not mark as permanent water. It also fell versus 2025.
Soil on the plain was slightly drier in 2026: 0.253 against 0.292 on the 0 to 1 scale. Water on the map came through the system, not from soil already saturated in the same stretch.
2025. Blue is water seen in the period. The patch covers the channel and wide areas of the plain.
2026, the same zone. Extent is somewhat smaller than in 2025, in line with the area table.
The count: 1 of 4, and the flood pulse still open
Each rise sign counts when it reaches its mark. Headwaters rain: 180 mm or more in the period, or more than the prior year with at least one day over 50 mm. Headwaters soil: mean 0.40 or more, from 0 to 1. River: published flood stage for the reference gauge. Plain: area with water greater than the prior year.
In 2025 and 2026 headwaters rain reached its mark. Soil did not reach 0.40 in either year. In 2026 the plain did not beat 2025; in 2025 the comparison with 2024 is not in this run (No data). The river row says No public mark: without an official mark, it is not scored.
1 of 4 in both years is isolated wetness in what can be scored. It does not confirm a flood pulse. It is not a contraction pulse either: that count would need other marks that are not published yet.
The count does not measure the height of a flood. Peak stages read at Plato and on the rest of the network are already in the table. What is missing to close the flood pulse is the mark with which the country calls “flood” on that staff.
The four rise signs, 1 to 31 October. Plain 2025: No data (2024 missing). The count only scores signs with a public mark; without an IDEAM flood stage the river signal stays open and the flood pulse is inconclusive. Source: Neometrico reading and IDEAM stages.
| Measure | 2025 | 2026 |
|---|---|---|
| Headwaters rain | Yes | Yes |
| Headwaters soil | No | No |
| IDEAM flood stage | No public mark | No public mark |
| Plain water > prior year | No data | No |
| Pulse signals | 1 of 4 | 1 of 4 |
All measures
In the headwaters soil stayed below 0.40 in both years: 0.366 in 2025 and 0.324 in 2026. The valley and plain concentrate the 2026 rain increase.
The measures, 1 to 31 October. Source: Neometrico reading and IDEAM stages.
| Measure | 2025 | 2026 |
|---|---|---|
| Headwaters rain (mm) | 384.0 | 298.7 |
| Wettest headwaters day (mm) | 39.3 | 31.2 |
| Headwaters days over 50 mm | 0 | 0 |
| Headwaters soil (0 to 1) | 0.366 | 0.324 |
| Valley rain (mm) | 159.8 | 208.3 |
| Floodplain rain (mm) | 146.4 | 214.0 |
| Floodplain soil (0 to 1) | 0.292 | 0.253 |
| Area of the plain with water (km²) | 226.26 | 220.15 |
| Water outside the usual channel (km²) | 38.30 | 30.84 |
| Peak stage at Plato (m) | 11.15 | 9.39 |
How the reading is made
The reading uses the same calendar day and month in both years, from 1 to 31 October, both ends included. Published marks score the flood pulse, the rise side. A contraction pulse would need other marks that are not published yet.
Rain is measured in millimetres, soil on a 0 to 1 scale, water on the plain in square kilometres, and the river in metres of stage. Rain and soil come from the whole area of each stretch, not a point. Each rain value is the daily mean in cells of about 11 km summed over the period; soil is the mean for the period in cells of about 9 km.
Plain water comes from satellite radar, with a pixel of about 20 m. On this plain radar passed about every six days. In 2025 the passes were 5, 11, 17, 23 and 29 October. In 2026, 1, 7, 13, 19, 25 and 31. The map shows the greatest extent seen on those passes, not the mean.
Stages come from IDEAM. The reference gauge is Plato, code 25027450. Plato’s 1–25 October 2026 gap is in IDEAM’s published instantaneous level feed, so the maximum read stays at 9.39 m on the 30th. Alert and flood stages are not invented when the agency does not publish them.
Conclusion
October 2026 brought more rain downstream in the basin. At Plato, the maximum read in the feed stays at 9.39 m on the 30th, below the peak of full October 2025; at El Banco, with nearly the full month, the drop is clearer. That does not close a contraction pulse in this count.
The practical lesson is the data gap. Without public flood stage and discharge, and with gaps in IDEAM’s instantaneous level feed, the flood pulse cannot close here. Whoever runs a municipality or lives on the bank loses the same comparator ANA provides in Brazil.
In the Taquari Valley the marks exist, and the reading could count days above flood stage. Here we publish metres read and state what is missing. Publishing official marks is not a technical detail: it is what turns stage into warning.
Questions
What is a river pulse?
In ecology, the movement of the basin as a system: the rise and the drawdown. In this reading, a flood pulse is when headwaters rain, soil, river stage and water on the plain close together toward rise in the same period.
Why is this reading inconclusive?
Because we seek a flood pulse, and the river signal is scored against the published flood stage for the reference gauge. IDEAM does not publish that mark for Plato. Without it, the count stays at 1 of 4 even when we know it rained in the headwaters.
Is the stage at Plato in 2026 the peak for all of October?
No. Plato’s 1–25 October 2026 gap is in IDEAM’s published instantaneous level feed, so the maximum read stays at 9.39 m on the 30th. In 2025 the full month reached 11.15 m on the 28th.
Why not use Calamar or Gambote?
Because this run took Plato as reference, with San Pedrito, El Banco and Gamarra. Calamar and Gambote appear in the IDEAM catalog, but the instantaneous level feed has no rows for them in October 2025 or 2026. They are left out of the reference.
Why compare with the prior year?
Because the prior year is the reference the reader lived. Saying 214 mm on the plain says little without knowing that in 2025 it was 146.4 mm on the same dates.
What is the difference between stage and discharge?
Stage is the height of the river in metres at the gauge. Discharge is how much water passes per second. Here we publish peak stages for the period. Discharge does not come in IDEAM’s public data for these gauges.
Why do the water map and stage not count the same thing?
Stage at Plato in 2026 is the maximum in IDEAM’s published instantaneous level feed (9.39 m on the 30th), not a full October: days 1–25 are missing from that feed. The map shows the greatest water extent radar saw on the passes in the period, five dates in 2025 and six in 2026, not the mean. One is height at a point; the other is area on the plain.
Where do rain, soil and water come from?
From satellite. Rain is a daily mean in cells of about 11 km. Soil is a mean for the period in cells of about 9 km. Water is radar, with a pixel of about 20 m, inside the plain. Stages come from IDEAM.
Why do public IDEAM data matter?
Because without a flood stage there is no comparable river signal across years or basins. Rain and radar tell part of the story; channel warning depends on marks that are not published here.
Let’s read your valley
If you want the same period in another valley, write with the place and the dates.
Talk to us