Pulsed-Neutron Comparison of Open- and Casedhole Wells: An Alaskan Case Study
You have access to this full article to experience the outstanding content available to SPE members and JPT subscribers.
This paper compares the results of gas identification and lithology identification using pulsed-neutron spectroscopy in openhole and casedhole environments. Most pulsed-neutron tools are run after casing; this study provides a unique opportunity to examine the effect of casing on spectroscopy by comparing casedhole measurements to measurements taken in the open hole before the casing was run.
Pulsed-neutron logging has evolved over the last 50 years, but the intrinsic physical measurements have remained unchanged, which means that operators cannot obtain a complete picture of the rock and fluids behind casing with conventional tools. However, advances in tool design and a new fast-neutron cross-section (FNXS) measurement provide for an alternative gas-identification technique. Gas in open holes is typically identified from neutron porosity and gamma-gamma density crossover. In casedhole environments, gamma-gamma density measurements are challenging because of the large casing and cement corrections needed. Previous gas identification in casedhole environments has relied on the formation hydrogen index (HI) or neutron porosity (TPHI) log and sigma.
In openhole environments, density and neutron porosity crossover is a typical gas identifier, but, in many instances, shale can mask the identification of gas. This is a common problem in some gas reservoirs in Alaska, and it leads to ambiguous interpretations about the gas saturation and potential producibility of different zones. Gas identification in casedhole environments is even more complicated because the density measurement is not commonly available.
The FNXS measurement responds primarily to formation atom density, for which most rocks, clays, and liquids have similar values. Comparatively, gas has a low atom density, and its presence will make the FNXS measurement read low. Thus, a gas pay zone can be differentiated from tight zones by the shift toward lower FNXS values. Also, the difference in FNXS between clean lithologies and clay is less than for sigma and TPHI, so FNXS, which is less affected by variable clay content, can be a more-robust gas indicator when variable clay is present.
A vertical well was drilled in an area known to have varying amounts of sands, shales, and coals. The lithology variation may not always be obvious on the logs, creating problems for gas identification. The case-study well was drilled with the following openhole logging-while-drilling (LWD) formation-evaluation measurements: gamma ray, induction resistivity, neutron porosity, and gamma-gamma density. In this well, an advanced pulsed-neutron log was also run to investigate whether it could add information that could reduce gas-identification ambiguity. To identify a baseline for gas identification in this area, the advanced pulsed-neutron log was run in two different environments: in open hole and after cementing and casing the well. The well was cased with 4.5-in., 12.6-lbm/ft casing in a 6.75-in. hole. The pulsed-neutron log was logged at 1,000 ft/hr in a mode in which both the time domain and the energy domain are recorded simultaneously, and logs of FNXS, TPHI, sigma, and spectroscopy were available. The measurements in open hole and in cased hole were compared for gas identification. Additionally, the spectroscopy from the open hole and cased hole were compared to examine the accuracy of the spectroscopy logged behind casing, where the environmental corrections for the casing and cement are a potential concern. Over 5,000 ft of data were logged, and two potential productive gas zones were identified for this study, Zone 1 and Zone 2.
Basic openhole logging analysis within the two zones was performed with gamma ray, resistivity, neutron porosity, and density from LWD measurements.
FNXS, TPHI, and sigma were measured on wireline in an openhole and a casedhole environment to better identify gas zones in this well (Fig. 1). Sigma, TPHI, and FNXS all follow a linear volumetric mixing law; therefore, volumetric components that have a significant difference can be easily seen in the logs. This is evident in the sigma measurements (Track 4), which show a decrease in the sigma values caused by sand in the reservoirs of Zones 1 and 2. Identification of the reservoirs in Zone 1 is much easier from the change in sigma compared with the gamma ray measurements, which are affected by the lithics in the area.
Gas is indicated by a decrease in the FNXS values. Tracks 6 and 7 show the FNXS values from the openhole and casedhole measurements. A gas-indicator cutoff line is drawn at a value of 7.5, the approximate value of FNXS in a water-filled shaly sand, to suggest pay zones where FNXS is less than the cutoff, assuming this is caused by the presence of gas. This suggests that the entire reservoir in Zone 1 and Zone 2 contains gas. Such an analysis contrasts with that of the neutron-density crossover, indicating that the shale effect is preventing gas detection from the neutron-density crossover.
The most-robust use of FNXS as a gas indicator is to plot it together with a matrix FNXS measurement that accounts for varying lithology. To do this, a lithology must first be determined, commonly from the spectroscopy data of the advanced pulsed-neutron tool.
This study also examines what kind of information can be obtained from different kinds of log data, including openhole LWD and advanced pulsed-neutron run on wireline in open hole and cased hole. Crossplots were made showing comparable measurements from each run. The crossplots were compiled over the entire 5,000-ft interval within the sands and shales. These plots show that a neutron porosity measurement can be made reliably during drilling, in open hole or in cased hole in this field.
Spectroscopy Log Analysis
Spectroscopy analysis provides a measure of the relative amounts of different elements in the borehole and formation. In spectroscopic analysis from casedhole logs, the elemental contributions from the casing and cement must be separated from the contributions from the formation. This was achieved, in this example, by subtracting the calcium, silicon, and iron contributions from the cement and casing in normalized yield space before performing the oxides closure. Several of the key measurements computed from the pulsed-neutron spectroscopy showed good repeatability between the open hole and cased hole.
The matrix FNXS measurement can then be compared with the FNXS, rather than using a simple fixed cutoff, as was shown in Fig. 1. In this way, any crossover where FNXS is less than matrix FNXS should be caused by gas, because the effect of variable lithology is accounted for. The reservoir in Zone 1 has a large difference between the matrix FNXS and the FNXS logs, which indicates gas. The lithology from the spectroscopy indicates that the reservoir has a similar clay volume to that of the shales above and below, but the gas can still be identified because FNXS is less than matrix FNXS. The FNXS shows a much clearer gas indication in this zone than any of the other measurements, including the openhole neutron density. Production results confirmed the FNXS interpretation. This zone may not have been identified as a pay zone using only the conventional openhole logs and is an example of how FNXS can add information and reduce uncertainty to the interpretation of complex shaly sand gas reservoirs.
The reservoir in Zone 2 is easier to identify from the shales above and below because of the shift in the gamma ray and the decrease in clay volume determined from the spectroscopy. It is interesting to note that a comparison between the FNXS and the FNXS matrix indicates that only the upper portion of the reservoir contains gas. In the lower portion of the reservoir, the FNXS and the matrix FNXS overlap, indicating that the shift to lower FNXS values is not the result of gas but of the lithology in the formation. This is consistent with the crossover of the neutron-density logs, where only the upper portion of the reservoir has crossover. In this zone, the reservoir is likely cleaner than in Zone 1, so the interpretation of FNXS and conventional neutron-density logs is more consistent than in Zone 1. This example shows how the FNXS can be used to identify gas behind casing as well as in the openhole environment where casing and cement are not factors in the analysis.
This study compared gas identification from an advanced pulsed-neutron tool that was run in a well before and after casing with traditional gas identification from standard openhole logs. The FNXS measurement from the pulsed-neutron tool was used to identify gas in this reservoir primarily because it is less affected by varying clay and lithology than other gas indicators, including the traditional openhole neutron density logs. Pulsed-neutron spectroscopy is also a helpful addition to this case to quantify the lithology and allow for the computation of matrix FNXS, which improves the robustness of gas indication of FNXS by compensating for the small but variable lithology effect on the FNXS log. Analysis of the pulsed neutron sigma, TPHI, FNXS, and spectroscopy showed that, even at relatively fast logging speeds of 1,000 ft/hr, the results in open hole and cased hole are similar, which gives the operator the flexibility to run pulsed-neutron logs in either condition in the future. Gas was also identified on the basis of the FNXS; traditional logs might not have identified gas. A more-confident identification of reservoir pay zones will lead to higher production and lower operational costs.
Pulsed-Neutron Comparison of Open- and Casedhole Wells: An Alaskan Case Study
01 August 2018
Old CaTS Technology Learns a New Trick
Acquiring data from an abandoned subsea well has been done before, but never quite like this.
As I read through the abstracts and papers that have been presented in the past year, I notice several key themes: verification of cement placement, development of new materials as a barrier, development of new additives to improve the cement barrier, and enhancement of existing techniques.
Technology Trends in Evaluating Cement Jobs Using Logging Tools
This paper reviews the technology trends in cement-job evaluation using logging tools and considers the main advantages and concerns associated with each technology.
Don't miss out on the latest technology delivered to your email weekly. Sign up for the JPT newsletter. If you are not logged in, you will receive a confirmation email that you will need to click on to confirm you want to receive the newsletter.
07 August 2018
09 August 2018