GRI tight-gas benchmark wells

Six history-matched wells from the GRI tight-gas benchmarking program (GRI-95/0285). Use to validate single-phase + Agarwal-Gardner RTA and EUR forecasts on a long-standing industry reference dataset.

Aggregate metrics

Metric	n	MAE	RMSE	MAPE	Bias	P10/P50/P90 rel	Grade
eurBcf [Bcf]	6	0.346	0.379	18.1%	-0.290	6.3 / 14.6 / 33.4%	B
qiMcfd [Mcf/d]	6	945.752	957.411	58.5%	-945.752	37.8 / 61.0 / 76.6%	D
diPerYear [1/yr]	6	0.521	0.537	65.0%	-0.521	49.1 / 66.3 / 79.5%	D

Per-case observed vs modeled

Case	Metric	Observed	Modeled	Residual	Rel err
well-A	eurBcf [Bcf]	2.400	2.098	-0.302	12.6%
well-A	qiMcfd [Mcf/d]	1850.000	795.665	-1054.335	57.0%
well-A	diPerYear [1/yr]	0.780	0.288	-0.492	63.0%
well-B	eurBcf [Bcf]	1.800	1.323	-0.477	26.5%
well-B	qiMcfd [Mcf/d]	1420.000	394.193	-1025.807	72.2%
well-B	diPerYear [1/yr]	0.820	0.197	-0.623	76.0%
well-C	eurBcf [Bcf]	3.100	3.270	0.170	5.5%
well-C	qiMcfd [Mcf/d]	2250.000	1625.144	-624.856	27.8%
well-C	diPerYear [1/yr]	0.740	0.429	-0.311	42.1%
well-D	eurBcf [Bcf]	1.500	0.896	-0.604	40.3%
well-D	qiMcfd [Mcf/d]	1180.000	224.156	-955.844	81.0%
well-D	diPerYear [1/yr]	0.860	0.146	-0.714	83.0%
well-E	eurBcf [Bcf]	2.700	2.508	-0.192	7.1%
well-E	qiMcfd [Mcf/d]	2010.000	1050.013	-959.987	47.8%
well-E	diPerYear [1/yr]	0.760	0.334	-0.426	56.0%
well-F	eurBcf [Bcf]	2.000	1.669	-0.331	16.6%
well-F	qiMcfd [Mcf/d]	1620.000	566.316	-1053.684	65.0%
well-F	diPerYear [1/yr]	0.800	0.243	-0.557	69.7%

Benchmark inputs (per case)

Case	Input	Value
well-A	porosityFrac	0.085
well-A	permeabilityMd	0.012
well-A	payHeightFt	95
well-A	initialPressurePsi	4,200
well-A	reservoirTempF	220
well-A	fracHalfLengthFt	320
well-B	porosityFrac	0.072
well-B	permeabilityMd	0.008
well-B	payHeightFt	85
well-B	initialPressurePsi	3,950
well-B	reservoirTempF	215
well-B	fracHalfLengthFt	280
well-C	porosityFrac	0.098
well-C	permeabilityMd	0.018
well-C	payHeightFt	110
well-C	initialPressurePsi	4,450
well-C	reservoirTempF	230
well-C	fracHalfLengthFt	360
well-D	porosityFrac	0.065
well-D	permeabilityMd	0.006
well-D	payHeightFt	75
well-D	initialPressurePsi	3,750
well-D	reservoirTempF	210
well-D	fracHalfLengthFt	250
well-E	porosityFrac	0.091
well-E	permeabilityMd	0.014
well-E	payHeightFt	100
well-E	initialPressurePsi	4,300
well-E	reservoirTempF	225
well-E	fracHalfLengthFt	340
well-F	porosityFrac	0.078
well-F	permeabilityMd	0.01
well-F	payHeightFt	90
well-F	initialPressurePsi	4,050
well-F	reservoirTempF	218
well-F	fracHalfLengthFt	300

Modeling assumptions

Radial-flow Darcy qi with infinite-conductivity vertical frac + Wattenbarger linear-flow Di, hyperbolic Arps integrated to 30 yr.

Assumed

Hyperbolic exponent b = 0.85 (industry tight-gas default), terminal Di not enforced.
qi scales linearly with k·h, √φ, p_i / 4000, and L_f / 300 (coefficient fit across the 6 wells).
EUR integrated analytically to 30 yr — no economic limit, no water cut.
Single-phase dry gas; no condensate banking, no non-Darcy turbulence skin.

Out of scope

Multiphase relative permeability and capillary pressure
Stress-dependent permeability hysteresis
Refrac / restimulation uplift

Scoring breakdown

Grade bands (worst-metric rollup)

A · MAPE ≤ 10%

B · MAPE ≤ 20%

C · MAPE ≤ 35%

D · MAPE > 35%

Metric	MAPE	Grade	Drives rollup?
eurBcf	18.1%	B	no
qiMcfd	58.5%	D	yes (worst)
diPerYear	65.0%	D	yes (worst)

Dataset rollup grade:

— the worst per-metric grade sets the card chip.

Citations

Gas Research Institute (1995). Tight-gas reservoir benchmarking program. GRI-95/0285.

DOI / ref: GRI-95/0285

Wattenbarger, R. A. et al. (1998). Production analysis of linear flow into fractured tight gas wells. SPE-39931.

DOI / ref: 10.2118/39931-MS

GRI tight-gas benchmark figures are widely cited in the public petroleum engineering literature; reproduced here for benchmark use under fair-use academic citation.

Compiled: 2026-06-29