Skip to content

BOOK_CONCEPT_2026.md

Latest commit

 

History

History
120 lines (95 loc) · 8.26 KB

BOOK_CONCEPT_2026.md

File metadata and controls

120 lines (95 loc) · 8.26 KB

The State of Hidroinformatics in 2026

Vision

A comprehensive, application-centric online knowledge base (~600 pages) for scientists, engineers, water managers, students, and decision-makers. Not an AI textbook. Not a GIS manual. A practical guide showing how modern tools solve real water problems, with AI as one tool among many.

The reader should finish each chapter understanding: "Here's a water problem. Here's what data exists. Here's how we solve it today."

Format: Interactive Online Asset

  • RAG-enabled — readers can ask questions about any chapter, get answers with citations
  • Searchable TOC with cross-linked chapters and concepts
  • Interactive navigation — topic index, concept map, chapter dependencies
  • Embedded resources — links to live data portals, downloadable datasets, video lectures
  • Layered depth — summary boxes for quick readers, expandable deep-dive sections for specialists
  • ~600 pages equivalent, ~25 pages per chapter average
  • Living document — updatable as data sources and tools evolve

Guiding Principles

  • Problem-first, tool-second — every chapter starts with a real water challenge
  • Layered content — executive summary → practical walkthrough → technical deep-dive
  • International — Hungary as primary example, US as reference, global alternatives
  • 2026 state of practice — what's operational today, not theoretical
  • AI where it helps — agentic AI for parameterization as a practical capability, not as the main topic
  • Self-contained — every chapter works standalone for RAG retrieval

Proposed Structure (25 chapters)

PART I: The Water Information Landscape

  1. Why Hidroinformatics Matters — water crises (floods, drought, groundwater depletion), why spatial data + computing changed everything, the Maidment/Tarboton legacy
  2. Mapping Water: From Paper to Pixels — coordinate systems, projections, how maps work (accessible version of geodesy)
  3. Where the Data Lives — global tour of water data: US (USGS, NHDPlus), EU (Copernicus, EFAS), Hungary (OVF, OMSZ), Asia/Africa (HydroSHEDS, CHIRPS)
  4. GIS as a Water Tool — what GIS does for hydrology, from desktop to cloud, the three views (geodatabase, visualization, processing)

PART II: Reading the Landscape

  1. The Grid: How Computers See Terrain — raster fundamentals, DEMs, resolution trade-offs
  2. Calculating with Maps — map algebra, combining layers, the precipitation-minus-infiltration paradigm
  3. Slope, Aspect, and the Shape of the Land — how terrain derivatives drive water movement
  4. Measuring Rain Where It Falls — interpolation, rain gauges vs. radar vs. satellite, watershed averaging

PART III: Following the Water Downhill

  1. Preparing the Digital Landscape — DEM conditioning (pit filling, stream burning) — why raw DEMs don't work for hydrology
  2. Which Way Does Water Flow? — flow direction, flow accumulation, the digital river network
  3. Drawing Watersheds from Data — automated watershed delineation, stream networks, comparison with mapped rivers
  4. Automating the Workflow — scripting with Python, processing thousands of watersheds

PART IV: When Water Becomes Dangerous

  1. Seeing the Ground in 3D: LiDAR — how laser scanning revolutionized terrain mapping
  2. How High Above the River? — the HAND concept, flood-prone terrain identification
  3. Mapping Where Floods Go — inundation mapping, rating curves, damage estimation
  4. Forecasting Floods in Real Time — NWM, EFAS, GloFAS, Hungarian systems — operational flood forecasting today
  5. Who Lives in the Flood Zone? — population exposure, building databases, emergency response

PART V: The Hidden Water

  1. Seeing Underground: 3D Subsurface GIS — representing geology and aquifers in 3D
  2. Wells, Boreholes, and Aquifer Maps — groundwater data collection and management
  3. Building a Picture of the Underground — hydrogeologic framework modeling
  4. Simulating Groundwater Flow — MODFLOW and finite-difference modeling in practice

PART VI: The Intelligent Water System

  1. When Models Meet Data — the calibration problem, why hydrological models need tuning, traditional approaches (manual, SCE-UA, DREAM)
  2. AI as a Hydrologist's Assistant — practical ML applications: flood prediction with LSTM, groundwater level forecasting, satellite-based ET estimation — application examples, not theory
  3. Agentic AI: The Autonomous Modeler — how AI agents run models, evaluate results, and optimize parameters autonomously. Practical demonstration: an agent calibrating a rainfall-runoff model for a Hungarian catchment. What this means for water management in 2026.
  4. The Future of Water Intelligence — digital twins, real-time data assimilation, climate adaptation, open science, the convergence of sensors + models + AI

Appendices

  • A. Data Source Quick Reference — table of all datasets mentioned, with URLs
  • B. Software Guide — ArcGIS, QGIS, TauDEM, HEC-HMS, MODFLOW, Python
  • C. Hungarian Water Data Portal Guide — step-by-step for data.vizugy.hu, OMSZ, KSH
  • D. Glossary

Chapter Design Template

Each chapter follows this pattern:

  1. The Problem (1 page) — a real-world water challenge, with a photo/map
  2. The Science (3-5 pages) — accessible explanation of the underlying concepts
  3. The Practice (3-5 pages) — how it's done today, with real examples and results
  4. Hungarian Example (1-2 pages) — applied to a Hungarian watershed or dataset
  5. Global Perspective (1 page) — how this applies in Asia, Africa, Middle East
  6. Key Takeaways (half page) — bullet summary

What Changes from Current State

Current (19 chapters) New (25 chapters) Change
Ch 1-4: Foundations Part I (Ch 1-4): Water Information Landscape Rewritten for general public, less technical
Ch 5-8: Raster Analysis Part II (Ch 5-8): Reading the Landscape Same content, accessible titles/language
Ch 9-12: Terrain Hydrology Part III (Ch 9-12): Following the Water Same content, accessible framing
Ch 13-17: Flood Analysis Part IV (Ch 13-17): When Water Becomes Dangerous Same content + EFAS/Hungarian systems
Ch 18-19: Water Info Absorbed into other chapters Ch 18 → Part I; Ch 19 → Part II/VI
— Part V (Ch 18-21): The Hidden Water NEW from Arc Hydro GW
— Part VI (Ch 22-25): The Intelligent Water System NEW: AI/ML applications + Agentic AI

Content Depth — No Limits

Size is unlimited. Each chapter grows to whatever depth the topic needs. Some chapters may be 10 pages, others 50. The online format has no printing constraints.

This allows:

  • Rich worked examples with real data
  • Hungarian + international case studies in every chapter
  • Detailed figures, maps, and result visualizations
  • "Try it yourself" interactive boxes with data download links
  • Further reading sections linking to the Handbook of HydroInformatics for ML deep-dives
  • Full derivations where they matter, skippable for casual readers via layered depth

RAG Architecture Notes

  • Each chapter should be self-contained with its own context (definitions, data source references)
  • Use consistent metadata tags per chapter: topic, methods, datasets, regions, difficulty level
  • Include a concept index that maps questions to chapters (e.g., "How do I delineate a watershed?" → Ch 11)
  • Every data source mention should include a structured reference block (name, URL, coverage, format, access) for RAG extraction

Key Design Decisions

  1. Ch 24 (Agentic AI) — Practical demonstration, not theory. Show an agent calibrating a model step by step. Use a Hungarian catchment (e.g., Zala River). Include results and workflow diagrams, not raw code.

  2. Tone — Accessible to a water manager or MSc student, not just GIS specialists. Minimize jargon. Define every acronym. Use analogies. Layer complexity (summary → detail → deep-dive).

  3. Interactive features — Embedded links to live data portals, downloadable sample datasets, video lecture clips from Maidment's course where relevant.

  4. Updatability — Design chapters as modular units. When a data portal URL changes or a new tool emerges, only one chapter needs updating.