mike-results

Read MIKE+ .res1d results via the mike-plus MCP server — list contents, summarize peaks, extract a single time series, or check which nodes flood (peak water level vs ground level). Use when an agent

SKILL.md

GitHub

MIKE+ Results

Reads MIKE+ .res1d result files (via mikeio1d). No license / no MIKE+ install needed — fully portable.

When to use

To find out what is in a result file, get peak values, extract a series for analysis/plotting, or answer "which manholes flood?". Use after mike_run, or on any existing .res1d. (Two result files side by side: mike-compare.)

Tools

  • mike_results_list — {res1d} → quantities, element counts + sample ids (nodes/reaches/structures/catchments), time range. Start here to discover valid quantity/element values.
  • mike_results_summary — {res1d, quantities?, skip_hours?} → per-quantity peak (value, element, chainage, time), in the canonical schema.
  • mike_results_read — {res1d, quantity, element, max_points?} → a single downsampled time series (times, values, unit, peak_value, peak_time).
  • mike_results_flooding — {res1d, skip_hours?, top_n?, include_outlets?} → per node: peak water level vs ground_level (and critical_level) → n_flooded, flooded[] ranked by exceedance (freeboard_m < 0, with peak_time, node type), n_above_critical, closest_to_flooding[]. Outlets are skipped by default (their level is the boundary, not flooding).

Conventions

  • Column identity is Quantity:ElementId[:chainage] (e.g. Discharge:Link_29:33.5333). element matches the id prefix.
  • Output follows the canonical, engine-agnostic schema ({engine, quantity, element_id, chainage, unit, ...}) so the same downstream skills work for SWMM/LSTM later. (mike_results_summary reports the peak's element as peak_element — same idea, named for the peak.)
  • Global-max caveat: mike_results_summary returns the max over the WHOLE run, which can land in the warm-up / initial-condition period (e.g. a base-flow value near t=0), not the storm peak. Always check peak_time; when you want the event peak, pass skip_hours (e.g. skip_hours: 6) to drop the initial window. The same applies to mike_results_flooding.
  • Flooding is judged against the ground level stored in the result file. Nodes with an obviously artificial critical_level (e.g. 100 m) or river/basin nodes may carry a nominal ground level: report the type and level, and let the user judge; do not filter them silently (only outlets are excluded by default, and that is stated in the output).
  • Report only what the file contains; if a quantity/element has no series, say so (the tool errors with a clear message) rather than guessing.

Orchestration

mike_results_list      -> learn quantities + element ids
mike_results_summary   -> peaks for a quick read (skip_hours)
mike_results_flooding  -> flooded nodes ranked (skip_hours)
mike_results_read      -> a series for a specific element  -> mike-plot