The waterway: where the river loses draft, and which stretch to dredge first

≈ 3 min read · updated Jul 2026

The river does not drop evenly: it loses draft at specific passes, and that is where each convoy's load gets decided. Knowing which ones close, how far ahead, and which stretch to dredge first changes shipping and infrastructure decisions.

Corridor map + 7-30 d forecast · LSTM
The Paraguay river on a single map: stretches, bottlenecks and forecast
How to read it: the river runs north to south and each stretch takes the color of its worst pass (green open, amber restricted, red critical); each point shows its measured level and the small bar compares it against the design threshold (dotted mark) · below, solid line = what already happened, dashed = what the model would project, band = the probable range
Paraguay river · north → south design threshold · 3.0 m dredging plan = MILP under a finite budget cargo direction downstream, to the sea dredge 1st dredge 2nd bottleneck: the lowest pass on the route it sets how much cargo the whole convoy carries Bahía Negra 4.21 m · open Fuerte Olimpo 4.27 m · open Vallemí 3.17 m · open Concepción 2.55 m · watch Antequera 2.27 m · critical Asunción 2.53 m · watch Villeta 3.03 m · open Pilar 4.33 m · open confluence with the Paraná B · Forecast for the critical stretch: 90 d observed, 30 d ahead draft restriction threshold high operational risk −90 d −50 d −10 d +15 d +30 d high low TODAY ← observed (public hydrometry) forecast → likely threshold crossing ≈ 2 weeks of lead time
Open · ≥ 3.0 m Restricted · 2.3 to 3.0 m Critical · < 2.3 m Dredging priority Observed level Model forecast Probable range
  • DHN levels · Navy · Jul 1, 2026
  • LSTM · time series
  • ECMWF / GFS
  • AIS · real traffic
  • MILP · what to dredge on a budget
Put simply: the map turns the level at each point into a decision. The levels come from the Navy public bulletin (Jul 1, 2026); the thresholds, the priority and the forecast are illustrative of the method over public historical data: the real study calibrates each pass with the design draft and the traffic, and it does not evaluate contracts or contractors.
Cargo by product + the cost of low water
What the river moves, in millions of tons, and what missing draft costs
How to read it: on the left, the 2020 cargo of the Brazil-Paraguay-Bolivia stretch by product (BCR on CPTCP data) · on the right, what low water cost according to each source, each bar with its period and scope
A · Cargo on the Brazil-Paraguay-Bolivia stretch · 2020 downstream, to the sea · exports ↓ soybeans 5.2 Mt soybean meal 2.6 Mt cargo from Brazil · mostly iron ore 2.3 Mt soybean oil 0.6 Mt upstream, from the sea · imports ↑ fuels 2.0 Mt fertilizers 0.75 Mt stretch total in 2020: 19.1 Mt · whole waterway: 94.4 Mt CAFyM (Aug 2026): 28 Mt projected · minerals from 6 to 8 Mt B · The cost of low water, in USD revenue the fleet missed · full 2020 low water USD 100 M shipping sector · La Nación PY · Jul 2021 fleet at 70% · July estimate for all of 2021 USD 120-150 M shipping sector · La Nación PY · Jul 2021 Argentine soybean meal and oil · Jan-Aug 2021 USD 620 M BCR · Weekly Report · Sep 24, 2021 USD per cm of low water no public source ties USD to centimeters: that curve is what the proposed model would estimate from your own shipments
Downstream · exports Upstream · imports Low water cost · Paraguay Low water cost · Argentine side
  • BCR · Weekly Report N° 2039 · Feb 2022
  • CPTCP · La Plata Basin
  • CAFyM · Aug 2026
  • La Nación PY · Jul 2021
Put simply: soy rules downstream and fuel rules upstream. 2020 tonnage from BCR report N° 2039 (Feb 11, 2022) on CPTCP data: the Brazil-Paraguay-Bolivia stretch moved 19.1 Mt and the whole waterway 94.4 Mt that year; the cargo from Brazil is mostly iron ore. The costs quote sources with different periods and scopes (the full 2020 low water, the 2021 full-year estimate made in July, eight months on the Argentine side) and cannot be added together; the USD-per-centimeter relation would be estimated from each operator’s shipments.

Where the river loses draft

The whole convoy loads for the worst pass on the route, and the decisions that matter get made weeks before loading, when the forecast only says how much it will rain.

The river restricts navigation at short stretches where the riverbed rises and draft drops, and the model we propose maps them by crossing two signals:

  • Public hydrometric data (ANNP): how much water each stretch has.
  • AIS: the positions that vessels themselves transmit (picked up by satellite or coastal stations), showing where convoys slow down, wait, or transfer cargo.

The second chart shows what is at stake: the Brazil-Paraguay-Bolivia stretch moved 19.1 million tons in 2020, with soy and its derivatives in front (BCR, Feb 2022), and low water has a price: in July, the Paraguayan fleet estimated it would miss between USD 120 and 150 million across all of 2021 (La Nación PY, Jul 2021). Every centimeter of draft lost is cargo left sitting at the port.

The model we propose, and the decision it enables

This is how we design it: hydrometric data, AIS and the global weather models ECMWF/GFS feed an LSTM, a neural network built for time series, with two outputs:

  1. Expected level per stretch, with an uncertainty band, 7 to 30 days ahead.
  2. Projected state of each critical pass: when it crosses the threshold and for how long.

The data: DHN and ANNP level series and AIS traces, all of them public. The freight and demurrage costs and the dredging cost per stretch are data we do not have today: they get collected with each client at the start, with source and frequency agreed on.

The technique: the LSTM forecasts the level; on top of it, mixed integer linear programming (MILP) builds the dredging plan under a finite budget.

The decision: which stretches to dredge, in what order and with which dredge; and for you as operator, when to sail and how much to load.

Which stretch to dredge first. Dredging the fourth stretch helps little if the third stays restricted; that is why the MILP optimizes the plan over the whole network. The analysis is strictly technical: it does not evaluate contracts, bids or contractors.

This week’s decision (illustrative example)

What the forecast saysDecision
The critical pass crosses the threshold in ~2 weeksSail earlier, full load
Stretch already restricted, level risingWait for the window
Chronic restriction with heavy trafficDredge that stretch first

Example scenarios, not a real recommendation: the rule gets validated with the operator’s data.

Other options on the same window: scheduled lightering · draft adjustment · rescheduled departure.

The validation

The validation will put the model up against real corridor data, and for that we are looking for operators: the paid pilot is one operator, one stretch, one season and metrics agreed on before starting. The result gets documented, whether it works or not.

The business case. Two numbers almost nobody calculates, and they come straight from your own shipments:

  • What the last low-water period cost you and how much was avoidable with two to four weeks of notice.
  • Which pass costs you draft and which dredged stretch gives you back the most cargo per trip.

How to measure it: forecast error against the observed level per stretch, and days of notice gained over the reference bulletin. It rests on a digital twin of the critical stretch: your own water-level and sedimentation sensors on top of the Navy and ANNP bulletins. A sail-and-load rule learned from your operation comes later, on top of that twin.

This map of critical passes stays open for debate across the corridor’s whole chain: cargo owners, shipowners, ports, exporters. If your operation reads the draft differently, or your records tell a different story, write to us and we will check the map against the data.

If you operate on the waterway, let's talk about your case →

Initial diagnosis at no cost · we reply within 24 business hours