Relative Optimized Reservoir Access

Relative Optimized Reservoir Access: A Flexible Statistical Method of Tying Geosteering Efficiency and Target Selection to Well Production - by Kent C. Stewart, Blue Dragon Geoscience, LLC, March 2015

Theoretically, there can be better prediction of well production if geosteering efficiency is known. Geosteering efficiency is simply percentage of well in target. If target zone is ideally selected, then knowing % in target probably correlates well to production. Another criteria may be net hot shales greater than a certain percent hot. Usually, as in net sand mapping, one likes to see a coarse and a fine isopach trend: in this case, one is not mapping, so isopach trend is misleading, but one may obtain useful reservoir access comparisons. This method is basically a recipe for comparing reservoir access by statistically combining different variables and relationships. Back to comparing hot shale percentages –maybe one as net ft> 50% hot shale (as defined by a calibrated zero-line (min) and a calibrated peak or 100% line (max)), and another similarly as net ft > 80% hot shale. In terms of reservoir access these net counts show how much footage was drilled in the hottest rock both in and out of target. It is a different way of measuring than simple % in target. Another way of comparing wells through gamma ray counts is to take a mean % hot shale. Mean hot shale, as I define it, is equivalent to normalized mean gamma ray of the lateral. The main reason the mean gamma ray is normalized is to filter out anything below zero hot shale (clean limestone in the Marcellus and Upper Devonian Plays). Once it is normalized a better avg. or mean gamma ray can be obtained as mean % hot shale. The method of normalizing is to make histograms of the lateral section of MWD gamma ray log of measured depth. Different wells should be treated similarly as net ft. values are sensitive to histogram intervals. Values of all variables can be preserved on a spread sheet and compared separately and in various combinations. The combination I have selected involves several different ways of comparing and weighting each to be mostly equivalent. The final criteria is mean % hot shale deviated from type section. This involves making and evaluating a TVD log type section of the target interval and counting net hot shales in ft. and % of section. Then one may compare these values to % of lateral that is beyond that hot shale threshold. Here one gets a value, positive or negative. In this case by adding 10 to each value one can convert all values to positive and still keep the comparative relationships between them. Each of the five values in this formula is averaged out, or divided by five. The formula is as follows:

Relative Optimized Reservoir Access (RORA) = (% in target + %>50% hot shale + %>80% hot shale + mean % hot shale + (mean % hot shale deviated from type section+10))/5. (Avg. of all five values) (first percentage refers to percentage of lateral in net hot shales; Ex:%>50% hs)

It should be noted that reservoir access is just one component of productivity and is probably best measured in local areas of similar characteristics rather than over large areas.  Other factors such as porosity, pressure, and gas-in-place are not considered, except perhaps whatever can be crudely estimated through gamma ray. The key to RORA is not the exact formula but that it is a thorough way of relativizing in order to predict productivity. Target selection can also be a factor in optimizing reservoir access, particularly where there is thin-bed geology of quality reservoir interbedded with non-reservoir rock.

This method may be most applicable where fracture treatments are predominantly horizontal with little growth, which is generally most often the case so that more drilling of hot gamma zones is equivalent to more production. This is the case in the Marcellus but in other plays like the Utica/Point Pleasant, or where silica or carbonate content leads to better recovery efficiencies through breaking more brittle rock, the method is probably less useful in those cases, where highest TOC shale is not the immediate reservoir.

Output of these data can be in the form of histogram charts and/or bubble maps. One can determine a percentage of reservoir access where 100% would be defined as drilling in a high TOC very hot gamma section for the entire lateral. This is rare where the hottest zones are just a few feet thick. With a decent dataset one can predict with 80% reservoir access relative to the 100% standard then production should be 80% or more of the 100% well. Of course, there may be other factors and small changes in hydraulic fracture growth can cause changes in production.

Another comparative criteria is avg. total gas for the lateral. This can be easily calculated from a mud log las file after filtering out any anomalous spikes. In one recent paper at the 2014 Eastern Section SPE meeting (using data mostly from Northeastern Pennsylvania Marcellus), a study was presented comparing various fracking, drilling, gamma, and gas show parameters to well productivity. In this study it was found that avg. total mud gas was among the best indicators, and was a better indicator than mean gamma ray. Perhaps if the mean gamma ray was normalized there would have been a better correlation to production. The best fracking indicators were stage spacing and proppant placed per stage. (will try to add author and title of study later)

Extreme dips, frequently changing dips, and faulting are geological constraints that affect geosteering efficiency. If being in target strongly affects well production, then these geological constraints can be seen as geological hazards in terms of reservoir access. Relative optimized reservoir access and footage in-target will likely both decrease under those conditions.

Blue Dragon Geoscience is well-equipped and experienced to do relative optimized reservoir access studies in conjunction with production analysis. In one sense, RORA is to geosteering as SRV (stimulated reservoir volume) is to fracking. In-target reservoir access is access via drilling and SRV is reservoir access via fracking. SRV can be estimated with the help of frack records and especially microseismic. Knowing reservoir access and stimulated volume one can theoretically predict production based on a standard correlation for the area. This is simply another geology-based tool that can be used to help predict and optimize well production and improve well economics.