Discovering hidden geothermal signatures using non-negative matrix factorization with customized k-means clustering

Velimir V. Vesselinov; Bulbul Ahmmed; Maruti K. Mudunuru; Jeffrey D. Pepin; Erick R. Burns; Drew L. Siler; Satish Karra; Richard S. Middleton

doi:10.1016/j.geothermics.2022.102576

USGS · 70237376

Discovering hidden geothermal signatures using non-negative matrix factorization with customized k-means clustering

Velimir V. Vesselinov·Bulbul Ahmmed·Maruti K. Mudunuru·Jeffrey D. Pepin·Erick R. Burns·Drew L. Siler·Satish Karra·Richard S. Middleton

View original source ↗Export metadata ↓

Abstract

Discovery of hidden geothermal resources is challenging. It requires the mining of large datasets with diverse data attributes representing subsurface hydrogeological and geothermal conditions. The commonly used play fairway analysis approach typically incorporates subject-matter expertise to analyze regional data to estimate geothermal characteristics and favorability. We demonstrate an alternative approach based on machine learning (ML) to process a geothermal dataset from southwest New Mexico (SWNM). The study region includes low- and medium-temperature hydrothermal systems. Several of these systems are not well characterized because of insufficient existing data and limited past explorative work. This study discovers hidden patterns and relations in the SWNM geothermal dataset to improve our understanding of the regional hydrothermal conditions and energy-production favorability. This understanding is obtained by applying an unsupervised ML algorithm based on non-negative matrix factorization coupled with customized k-means clustering (NMFk). NMFk can automatically identify (1) hidden signatures characterizing analyzed datasets, (2) the optimal number of these signatures, (3) the dominant data attributes associated with each signature, and (4) the spatial distribution of the extracted signatures. Here, NMFk is applied to analyze 18 geological, geophysical, hydrogeological, and geothermal attributes at 44 locations in SWNM. Using NMFk, we find data patterns and identify the spatial associations of hydrothermal signatures within two physiographic provinces (Colorado Plateau and Basin and Range) and two sub-regions of these provinces (the Mogollon-Datil volcanic field and the Rio Grande rift) in SWNM. The ML algorithm extracted five hydrothermal signatures in the SWNM datasets that differentiate between low (<90) and medium (90-150)-temperature hydrothermal systems. The algorithm also suggests that the Rio Grande rift and northern Mogollon-Datil volcanic field are the most favorable regions for future geothermal resource discovery. NMFk also identified critical attributes to identify medium-temperature hydrothermal systems in the study area. The resulting NMFk model can be applied to predict geothermal conditions and their uncertainties at new SWNM locations based on limited data from unexplored regions. The code to execute the performed analyses as well as the corresponding data can be found at https://github.com/SmartTensors/GeoThermalCloud.jl.

Explore related subjects

Geothermics Article

Source-reported bounding extent: 32.008075959291055° to 35.69299463209881° latitude; -109.05029296875° to -106.094970703125° longitude. This indicates report coverage, not an exact sampling location. View area on OpenStreetMap.

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Velimir V. Vesselinov, Bulbul Ahmmed, Maruti K. Mudunuru, Jeffrey D. Pepin, Erick R. Burns, Drew L. Siler, Satish Karra, Richard S. Middleton. 2022. Discovering hidden geothermal signatures using non-negative matrix factorization with customized k-means clustering. https://doi.org/10.1016/j.geothermics.2022.102576

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related USGS reports

USGS2026

Prioritizing resource protection and understanding potential susceptibility of springs to surficial changes in a low-temperature geothermal system

Geothermal systems are vulnerable to changes in water budget and composition, requiring science-based management. This study uses a dataset of spring water temperatures, time series of groundwater residence time tracers (tritium and carbon-14), and stable isotopes of water to understand geothermal flow in a low-temperature geothermal system in north west Colorado, United States (Steamboat Springs). The geothermal system is bisected by the Yampa River, necessitating a stream mass balance approach to quantify total discharge. Time series analysis of water temperature data provides a ranked list of features more susceptible to surficial changes, which is corroborated using time series of tritium which indicate spatially distinct patterns of mixing between modern and pre-modern groundwater. All springs contain a portion of pre-modern groundwater that is thousands to tens of thousands of years old, a period coinciding with melting of extensive Pleistocene glaciers that was likely one of the recharge sources to the geothermal system. Stream mass balance indicates that greater than 80% of the total geothermal discharge is derived from diffuse or small springs, highlighting the extensive nature of the geothermal outflow zone and the association with local geologic structures. This study provides baseline data to support management of the Steamboat Springs geothermal system and indicates the utility of these approaches in developing science-based geothermal management.

Colorado↗

USGS2025

Reservoir thermal energy storage pre-assessment for the United States

Storing thermal energy underground for later use in electricity production or direct-use heating/cooling is a promising, viable, and economical green energy option. Reservoir thermal energy storage (RTES) is one such option, which stores energy in underutilized permeable strata with low ambient groundwater flow rates and more geochemically evolved (e.g. brackish/saline) waters relative to overlying principal aquifer systems. The U.S. Geological Survey has begun assessing RTES potential nationally by focusing on five generalized geologic regions (Basin and Range, Coastal Plain, Illinois Basin, Michigan Basin, Pacific Northwest) across the United States. Hydrogeologic reservoir models are developed for the following eight metropolitan area cities within those regions to evaluate RTES performance across different climates and subsurface conditions: Albuquerque, New Mexico; Charleston, South Carolina; Chicago and Decatur, Illinois; Lansing, Michigan; Memphis, Tennessee; Phoenix, Arizona; and Portland, Oregon. Evaluated metrics include estimated required well spacing, thermal storage capacity, and thermal recovery efficiency through time. Also considered for each reservoir are potential complicating factors, including reservoir depth, thermally driven free convection, and groundwater salinity. This work focuses on direct-use cooling because the need for cooling modern office buildings greatly exceeds that for heating in most parts of the country (Falta and others, 2016); however, the evaluated metrics are also relevant to heating and electricity applications. Results indicate that favorable RTES conditions exist in each region, with the Coastal Plain and Basin and Range being especially favorable for thermal storage capacity, while the Pacific Northwest and Michigan Basin excel at energy recovery for the evaluated cooling application. The results underscore the utility of developing maps of thermal storage capacity, subsurface temperature models, and volumetric estimates of thermal storage capacity to serve as key RTES resource classification standards. Overall, this pre-assessment provides a basic understanding of RTES potential in several cities and geologic regions throughout the country and will aid ongoing thermal energy storage assessment efforts.

Arizona, Illinois, Michigan, New Mexico, Oregon, S↗

USGS2025

Favorability mapping for hydrothermal power resource assessments of the Great Basin, USA

The U.S. Geological Survey (USGS) is updating the 2008 assessment of conventional hydrothermal resources for the Great Basin in the western United States. As part of this work, the workflow for hydrothermal resource favorability maps is being modified to integrate modern data-driven machine learning (ML) methods. Improvements include: [1] using new and refined evidence layers (features); [2] using an order of magnitude more training sites (labeled examples); [3] utilizing simple but non-linear supervised ML algorithms; [4] representing positive training sites (wells with measured heat flow) with their ordinal value proportional to the magnitude of convective upflow (i.e., low, high, or very high convective signals instead of past strategies using positive-negative labels); [5] supplementing training sites with additional sites with low convective signals to represent diverse under-sampled areas where hydrothermal systems are unlikely to exist; [6] comparing with competing approaches; and [7] utilizing Monte Carlo cross-validation to estimate and evaluate prediction uncertainty. For the new favorability map, over half of the power-producing systems (i.e., 15 of 28) are predicted in the 99th percentile of most favorable locations (i.e., the highest 1 % of favorability, corresponding to 1 % of the map area), exceeding the performance of past models that have explicitly used power plants as training sites. Previous favorability maps predicted approximately half of the power-producing hydrothermal systems above the 80th percentile (i.e., 20 % of the map area). For the new favorability map, 93 % of power-producing systems (i.e., 26 of 28) are above the 80th percentile. The power-producing systems for which the new model does not perform well are either comparatively small, low-temperature systems or systems also not predicted well by prior modeling approaches, suggesting that these few systems are unusual when compared with most power-producing systems. Focusing research on these known, seemingly different systems may yield new insights and subsequent discovery of new prospects.

California, Idaho, Nevada, Oregon, Utah↗