Chan Diagnostic Plot in Waterflooding

Chan Diagnostic Plot - The Waterflood Engineer's Early Warning System

Most waterflood problems are invisible until they become expensive. By the time you see a sharp water cut increase or a pressure anomaly, the damage - channeling, coning, or poor sweep - has already been developing for months. The Chan Diagnostic Plot is one of the few tools that lets you catch these problems early, using only data you already have: production rates and water cut history. This guide shows you how to build it, read it, and act on it.

1. What is the Chan Diagnostic Plot?

Developed by K.S. Chan in 1995, the Chan Diagnostic Plot is a log-log plot of Water-Oil Ratio (WOR) and its derivative (dWOR/dt) versus time. It was specifically designed to differentiate between water production mechanisms - coning, channeling, and normal displacement - based on the shape and slope of the curves.

Unlike simple water cut plots, the Chan plot uses the derivative of WOR as a diagnostic fingerprint. Each production problem has a characteristic signature on this plot, making it a powerful diagnostic tool even when production history is limited.

Key insight: The Chan plot does not just show you that water production is increasing - it tells you WHY it is increasing and which corrective action to take.

2. Key Parameters and Definitions

2.1 Water-Oil Ratio (WOR)

WOR is the ratio of water produced to oil produced at surface conditions:

WOR = Qw / Qo
Where Qw = water production rate (STB/day), Qo = oil production rate (STB/day)

Note the difference with Water Cut (WC):

• WOR = Qw / Qo - ranges from 0 to infinity
• WC = Qw / (Qw + Qo) - ranges from 0 to 1 (0% to 100%)

At WC = 50%, WOR = 1.0. At WC = 90%, WOR = 9.0. At WC = 99%, WOR = 99. The WOR scale is much more sensitive at high water cuts, which is why the Chan plot uses it instead of WC.

2.2 WOR Derivative (dWOR/dt)

The derivative of WOR with respect to time is the key diagnostic element. It measures how fast the water-oil ratio is changing. On a log-log scale, different production mechanisms produce distinct slopes:

Mechanism	WOR Slope	dWOR/dt Slope	Visual Pattern
Bottom water coning	Steep positive	Parallel to WOR	Both lines rise steeply together
Channeling / fractures	Moderate positive	Below WOR, converging	Lines converge at late time
Normal displacement	Gentle positive	Below WOR, stable gap	Stable separation between lines
Cusping (gas or water)	Sharp spike then decline	Crosses WOR line	dWOR/dt crosses above WOR

3. How to Build the Chan Diagnostic Plot - Step by Step

Step 1 - Data Collection

You need monthly (or daily) production data for each well:

• Oil production rate (Qo) in STB/day or STB/month
• Water production rate (Qw) in STB/day or STB/month
• Corresponding time (months or days since water breakthrough)

Important: Start the time axis from water breakthrough, not from first production. Pre-breakthrough data adds noise without diagnostic value.

Step 2 - Calculate WOR for Each Time Step

Month	Qo (STB/d)	Qw (STB/d)	WOR	WC (%)
1 (BT)	1,200	50	0.042	4%
6	980	420	0.43	30%
12	650	1,850	2.85	74%
18	380	3,420	9.0	90%
24	210	5,890	28.0	97%

Step 3 - Calculate the WOR Derivative

Use the finite difference approximation:

dWOR/dt = (WOR(n+1) - WOR(n-1)) / (t(n+1) - t(n-1))
Use central difference for interior points, forward/backward difference for endpoints

Step 4 - Plot on Log-Log Scale

Plot both WOR and dWOR/dt on the Y-axis (log scale) against time on the X-axis (log scale, starting from breakthrough). The resulting pattern gives you the diagnosis.

4. Real Field Case - Chan Plot Diagnosis and Intervention

On a mature sandstone waterflood in North Africa, Well P-12 showed a rapidly increasing water cut from 45% to 92% over 8 months - faster than any offset well in the same pattern.

The Chan plot for P-12 showed the classic channeling signature: WOR rising steeply on a log-log scale with dWOR/dt converging toward the WOR line. This immediately ruled out coning (which would show both lines parallel and steep) and confirmed preferential flow through a high-permeability streak between injector I-8 and producer P-12.

Actions taken:

1. Confirmed diagnosis with tracer test - dye appeared in P-12 within 18 days from I-8 (expected: 90+ days)
2. Applied polymer gel treatment in I-8 to plug the high-perm streak
3. Reduced I-8 injection rate by 30%, redistributed to I-9
4. Result: P-12 water cut dropped from 92% to 71% within 4 months, oil rate recovered from 210 to 480 STB/day

Without the Chan plot, this would have been diagnosed as a reservoir depletion problem and the well might have been shut in prematurely - leaving significant recoverable oil behind.

5. Chan Plot vs Other Diagnostic Tools

Tool	What it diagnoses	Data needed	Best used for
Chan Plot	Mechanism of water production	Qo, Qw vs time	Early diagnosis, well selection
VRR Analysis	Injection-production balance	Wi, Np, Wp, FVF	Pattern management
X-plot (Dake)	Recovery efficiency	Np, Wi cumulative	Field-level sweep assessment
Hall Plot	Injector performance	Pwf, Wi cumulative	Injection well diagnosis

Use the Chan plot first for producer diagnosis, then combine with VRR analysis at the pattern level for a complete waterflood assessment.

6. Practical Limitations

• Minimum data requirement: You need at least 6-8 data points after breakthrough for a reliable diagnosis - do not interpret with less
• Rate normalization: If production rates change significantly (workovers, choke changes), normalize WOR to eliminate operational noise
• Multi-layer reservoirs: In commingled production from multiple layers, the Chan plot shows a composite signature that may be misleading - consider using it per layer if PLT data is available
• Heterogeneous reservoirs: The classic Chan signatures were developed for relatively homogeneous sandstones - carbonates and naturally fractured reservoirs may show hybrid signatures requiring additional interpretation

Conclusion

The Chan Diagnostic Plot is one of the most cost-effective tools in the reservoir engineer's toolkit. It requires no special software - Excel is sufficient - and uses only routine production data. Yet it can differentiate between mechanisms that have completely different solutions: a coning problem requires rate reduction or recompletion, while a channeling problem requires conformance control.

Apply it systematically to every producer in your waterflood portfolio on a quarterly basis. The wells that show abnormal Chan signatures are your priority candidates for intervention - and the ones where your workover budget will deliver the highest return.

Want to see how to build a Chan Plot in Excel with your own field data? Join our Telegram group where we share templates and discuss real cases. Also check our YouTube channel for step-by-step tutorials on waterflood diagnostic tools.