Data Center Reference — The Spatial Data That Feeds IGW-NET

23.1 DEM / Elevation Datasets

The top elevation of the aquifer — and the surface drainage via the DEM-as-drain mechanism (Ch. 14 §14.2.1) — is sourced from digital elevation models. IGW-NET supports multiple resolutions and sources with automatic selection by default.

23.1.1 Supported resolutions and sources

The parser maintains this mapping as DEM_RESOLUTION_MAP. The auto-by-grid default is almost always the right choice — IGW-NET matches DEM resolution to cell size, which avoids both under-resolving (losing topographic detail) and over-resolving (wasting computation on sub-cell topography).

23.1.2 Why DEM resolution matters

Even though the DEM affects only the top elevation and the DEM-as-drain boundary, its resolution has practical consequences:

• The DEM-as-drain in Level-1 SW representation (Ch. 14) uses the DEM directly as the drain elevation. Coarse DEMs miss small streams and wetlands; fine DEMs capture them but may over-represent small-scale topography that doesn't matter for regional flow.
• T-PROGS 3D geology visualization (Ch. 17) uses the DEM as the top surface for cross-sections. Coarse DEMs produce less-readable cross-sections for fine-scale work.
• Coupled lake modeling (Ch. 15) uses the DEM for lake-adjacent topography and runoff routing. Fine DEMs support better lake water balance representation.

For Michigan and other well-mapped USGS NED areas, 10m is typically the sweet spot for detailed work. For global or remote-area work, 30m from ASTER+SRTM is the practical default.

23.2 Hydraulic Conductivity Datasets

K is the most-tuned parameter in groundwater modeling. The Data Center provides regional and global K datasets that give reasonable starting points before calibration (Ch. 18) refines them.

23.2.1 The K dataset catalog

Parser catalog: K_DATASETS. For Michigan models, IDs 1, 2, and 4 give three distinct K choices depending on whether you're modeling glacial drift, a specific aquifer unit, or bedrock. For models outside Michigan in the USA, ID 5 (USA Glacial K) is typical. For global models without a regional dataset, ID 3 (GLHYMPS 2.0) provides the baseline.

23.2.2 Data Center K vs. T-PROGS K

The Data Center K datasets are smoothed regional compilations — spatial patterns of mean K across regions, suitable as a starting point for regional-scale work. T-PROGS (Ch. 17) is a different path that produces realistic heterogeneous K from borehole lithology. The two aren't alternatives; they're different tools for different questions:

• Data Center K: for regional flow patterns, water-supply studies, and first-pass calibration. Captures the broad-scale spatial variation but not the stratigraphic heterogeneity.
• T-PROGS K: for problems where stratigraphic control matters — contaminant transport with preferential flow paths, wellhead protection in complex geology, coupled lake modeling where lake-bed lithology matters. Captures the discrete layered structure.

Both draw on spatial data; Data Center K is delivered as a raster via WMTS; T-PROGS is delivered as a .tp file from the server-side T-PROGS workflow using borehole databases (§23.5).

23.3 Aquifer Thickness / Bottom Elevation Datasets

The aquifer bottom elevation (and equivalently, aquifer thickness given the DEM top) is a structural parameter. The Data Center provides several datasets covering different regions and resolutions.

23.3.1 The bottom elevation catalog

Parser catalog: BOTE_DATASETS. These datasets give a bottom elevation (or thickness subtracted from DEM) as spatially variable input. Bottom elevation is especially important for:

• Unconfined aquifer models — the saturated thickness depends on the bottom elevation; wrong bottom → wrong flow-through capacity
• Multi-layer models (Ch. 10) — where layer boundaries anchor to aquifer geometry
• Bedrock-aquifer systems — where the bottom represents the bedrock interface

23.4 Recharge Datasets

Recharge — water entering the aquifer from the surface — is another heavily-calibrated parameter. The Data Center provides climate-and-hydrology-based estimates at multiple scales.

23.4.1 The recharge catalog

Parser catalog: RECH_DATASETS. The USA Long-term Mean (ID 2) is the most common choice for US-based work; it's a USGS-derived dataset tuned for steady-state and long-term simulation. For global work or finer temporal resolution, IMERG (ID 3) provides precipitation-based proxies.

23.4.2 Watch the units (Ch. 14 §14.7)

23.5 Borehole Lithology and Static Water Levels

Regional well databases provide two critical inputs: borehole lithology for T-PROGS heterogeneous K (Ch. 17) and Static Water Levels for calibration observations (Ch. 18). Both come from the same regional database integrations.

23.5.1 Regional coverage

Approximately 10 U.S. states, plus all Canadian provinces have water well databases integrated with IGW-NET's Data Center for automated access. Michigan's Wellogic is the prototype; equivalent integrations exist for additional US states with state-run water well record databases, and for every Canadian province (each province maintains a provincial well record database).

Within the covered regions, the Data Center provides:

• Borehole lithology logs — depth-interval records of the materials encountered at each well; used by T-PROGS (Ch. 17) to generate realistic 3D heterogeneous K fields
• Static Water Level (SWL) records — the water-table elevation measured at each well when not pumping; used as calibration observations (Ch. 18)
• Well construction metadata — depth, screen interval, construction date — for filtering and interpretation

Outside the covered regions, users supply their own well data in the appropriate format. The methodology (T-PROGS, calibration) is identical; only the data-acquisition path differs.

23.5.2 Where these datasets are used

• T-PROGS (Ch. 17) — borehole lithology feeds the transition-probability geostatistical simulation; server-side T-PROGS produces a .tp file that's delivered to IGW-NET as heterogeneous K
• Calibration (Ch. 18) — SWLs are the primary calibration target; accessed via IGWServer connection with filtering through the Well Data Processing Tool

23.6 Streams and Lakes

When you define a model domain, IGW-NET auto-loads streams and lakes from the Global Database. This is the Level-2 surface water representation covered in Ch. 14.

23.6.1 Global Database streams and lakes

• Coverage: Global (with denser coverage in well-mapped regions like the USA and Europe)
• Source: Compiled from hydrography databases — NHD (National Hydrography Dataset) in the USA; equivalent national datasets globally
• Auto-loading: Streams and lakes within the model domain appear automatically when the domain is drawn; Ch. 14 covers the Level-2 representation
• Default coupling: One-way default (SW as BC); user can switch to two-way coupling for specific features (Ch. 14 §14.6.4; Ch. 15)

Parser detection: .server_streams_lakes lists the auto-loaded features; each feature has isGlobal=0 for Global Database and isGlobal=1 for user-data/DataNET-transferred features.

23.6.2 User-added or DataNET-transferred features

For features not in the Global Database — a small local stream, a specialized site-survey lake boundary, a regional hydrography not yet integrated — users can draw or import features manually. Features from DataNET transfers are identifiable in the model file by their filename prefixes (dataNetWCS or dataNetmWCS) and are tracked separately by the parser.

23.7 Global Database vs DataNET — The Distinction

IGW-NET users encounter two data-source labels: Global Database and DataNET. Understanding the distinction helps when inspecting model provenance or debugging data-import questions.

23.7.1 Global Database

The Global Database is the bundled set of curated datasets that ships with IGW-NET. Characteristics:

• Always available — accessible from any model, any time, no separate transfer needed
• Curated — datasets are selected, processed, and maintained by the IGW-NET team
• Default sources — the datasets IGW-NET defaults to for DEM, K, recharge, and streams/lakes unless the user overrides
• WMTS delivery — raster data is served via Web Map Tile Service; vector features (streams, lakes) come from the Global Database infrastructure
• Identified in the parser — Global Database features have isGlobal=0; the dataset IDs resolve through the parser's catalogs (K_DATASETS, BOTE_DATASETS, RECH_DATASETS, DEM_RESOLUTION_MAP)

23.7.2 DataNET

DataNET is the separate user-driven data transfer service for specialized datasets. Characteristics:

• On-demand transfer — user initiates a transfer; specific datasets move into the model's context
• Specialized data — for datasets not in the curated Global Database (e.g., a specific regional transmissivity dataset, a site-specific drilling compilation)
• Identifiable by filename prefix — imported raster files starting with dataNetWCS or dataNetmWCS are DataNET imports; the parser detects and reports these
• Reported separately — model reports distinguish DataNET-sourced imports from Global Database defaults and from user-uploaded custom rasters

23.7.3 The metadata gap