SEARCH · USGS Science

Results for “Earth”

Search indexed USGS publications on groundwater, aquifers, geologic maps, mineral resources and earthquakes. Explore source records by subject and place.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 325 records · Page 18Linked to original sources

USGS2012

Acidification of Earth: An assessment across mechanisms and scales

In this review article, anthropogenic activities that cause acidification of Earth’s air, waters, and soils are examined. Although there are many mechanisms of acidification, the focus is on the major ones, including emissions from combustion of fossil fuels and smelting of ores, mining of coal and metal ores, and application of nitrogen fertilizer to soils, by elucidating the underlying biogeochemical reactions as well as assessing the magnitude of the effects. These widespread activities have resulted in (1) increased CO 2 concentration in the atmosphere that acidifies the oceans; (2) acidic atmospheric deposition that acidifies soils and bodies of freshwater; (3) acid mine drainage that acidifies bodies of freshwater and groundwaters; and (4) nitrification that acidifies soils. Although natural geochemical reactions of mineral weathering and ion exchange work to buffer acidification, the slow reaction rates or the limited abundance of reactant phases are overwhelmed by the onslaught of anthropogenic acid loading. Relatively recent modifications of resource extraction and usage in some regions of the world have begun to ameliorate local acidification, but expanding use of resources in other regions is causing environmental acidification in previously unnoticed places. World maps of coal consumption, Cu mining and smelting, and N fertilizer application are presented to demonstrate the complex spatial heterogeneity of resource consumption as well as the overlap in acidifying potential derived from distinctly different phenomena. Projected population increase by country over the next four decades indicates areas with the highest potential for acidification, so enabling anticipation and planning to offset or mitigate the deleterious environmental effects associated with these global shifts in the consumption of energy, mineral, and food resources.

Applied Geochemistry↗

USGS2015

EverVIEW: a visualization platform for hydrologic and Earth science gridded data

The EverVIEW Data Viewer is a cross-platform desktop application that combines and builds upon multiple open source libraries to help users to explore spatially-explicit gridded data stored in Network Common Data Form (NetCDF). Datasets are displayed across multiple side-by-side geographic or tabular displays, showing colorized overlays on an Earth globe or grid cell values, respectively. Time-series datasets can be animated to see how water surface elevation changes through time or how habitat suitability for a particular species might change over time under a given scenario. Initially targeted toward Florida's Everglades restoration planning, EverVIEW has been flexible enough to address the varied needs of large-scale planning beyond Florida, and is currently being used in biological planning efforts nationally and internationally.

Computers & Geosciences↗

USGS2026

Utilization of multiple geochronology techniques to constrain the age of laterization and mineralization of the world-class Mount Weld rare earth element deposit, Western Australia

Pervasive chemical weathering on stable cratons may form thick regoliths and elemental enrichment, but constraining the age of regolith formation is challenging. In this study we utilize multiple geochronological techniques on different minerals from the world-class Mount Weld rare earth element (REE) deposit, formed by lateritic weathering of a carbonatite, to constrain the age of formation and provide insight into landscape evolution. The oldest dates, ca. 100 to 38 Ma, are from Lu-Hf dating of churchite [HREE(PO 4 )·2(H 2 O)], a heavy REE phosphate mineral. Growth bands on individual minerals show a younging outwards. 40 Ar/ 39 Ar geochronology of cryptomelane [K(Mn 4+ ,Mn 2+ )₈O₁₆] yielded dates from ca. 40 to 27 Ma. Similarly, (U-Th)/He geochronology of goethite [FeO(OH)] yielded dates ranging from ca. 45 to 19 Ma. Integrating results into regional constraints, suggests 1) churchite formed by mineral saturation in a karst-like setting below the water table from ca. 100 to 40 Ma, 2) with minor uplift and erosion, cryptomelane and goethite formed at or near the water table between ca. 45 and 19 Ma, 3) after ca. 15 to 10 Ma chemical weathering within the profile had ended. Other studies document that the region experienced minimal uplift and a wet, warm climate from ca. 100 Ma to 15 Ma. These conditions and the high carbonate content of the carbonatite promote extensive chemical weathering, a deep weathering profile, and the preservation of the weathered section. This study highlights the use of multiple geochronological techniques utilizing different minerals to provide insight into how laterites form and to constrain the timing and history of the formation of this important mineral deposit.

Western Australia↗

USGS2024

Determination and prediction of micro scale rare earth element geochemical associations in mine drainage treatment wastes

Acid mine drainage (AMD) has been proposed as a novel source of rare earth elements (REE), a group of elements that includes critical metals for clean energy and modern technologies. REE are sequestered in the Fe–Al–Mn-rich precipitates produced during the treatment of AMD. These AMD solids are typically managed as waste but could be a REE source. Here, results from AMD solids characterization and geochemical modeling are presented to determine the minerals/solid phases that are enriched in REE and identify the mechanism(s) of REE attenuation. AMD solids collected from limestone-based AMD treatment systems were subjected to sequential extraction and synchrotron microprobe analyses to characterize the binding nature of the REE. The results of these analyses indicated REEs were mainly associated with Al or Mn phases. Only selected REE (Gd, Dy) were associated with Fe phases, which were less abundant than Al and Mn phases in analyzed samples. The sequential extractions demonstrated that acidic and/or reducing extractions effectively mobilize REE from the AMD solids evaluated. The observed element associations in solids are consistent with geochemical model results that indicate dissolved REE can be effectively attenuated by adsorption on freshly precipitated Fe, Al, and Mn oxides/hydroxides. The model, which simulates dissolution of CaCO 3 and the precipitation of Fe, Al, and Mn oxides with increased pH, accurately predicts the pH dependent accumulation of dissolved REE with Al, Mn, and Fe oxides/hydroxides in the studied AMD treatment systems. The methods and results presented here can be used to identify conditions favorable for accumulation of REE-enriched AMD solids and possible passive or active treatment(s) to extract REE from AMD. This information can be used to design AMD treatment systems for the recovery of REE and is an opportunity to transform the challenges of addressing polluted mine drainage into an environmental and economic asset.

Pennsylvania↗

USGS2014

Surface disposal of produced waters in western and southwestern Pennsylvania: potential for accumulation of alkali-earth elements in sediments

Waters co-produced with hydrocarbons in the Appalachian Basin are of notably poor quality (concentrations of total dissolved solids (TDS) and total radium up to and exceeding 300,000 mg/L and 10,000 pCi/L, respectively). Since 2008, a rapid increase in Marcellus Shale gas production has led to a commensurate rise in associated wastewater while generation of produced water from conventional oil and gas activities has continued. In this study, we assess whether disposal practices from treatment of produced waters from both shale gas and conventional operations in Pennsylvania could result in the accumulation of associated alkali earth elements. The results from our 5 study sites indicate that there was no increase in concentrations of total Ra (Ra-226) and extractable Ba, Ca, Na, or Sr in fluvial sediments downstream of the discharge outfalls (p > 0.05) of publicly owned treatment works (POTWs) and centralized waste treatment facilities (CWTs). However, the use of road spreading of brines from conventional oil and gas wells for deicing resulted in accumulation of Ra-226 (1.2 ×), and extractable Sr (3.0 ×), Ca (5.3 ×), and Na (6.2 ×) in soil and sediment proximal to roads (p < 0.05). Although this study is an important initial assessment of the impacts of these disposal practices, more work is needed to consider the environmental consequences of produced waters management.

Pennsylvania↗

USGS2022

Machine learning and data augmentation approach for identification of rare earth element potential in Indiana Coals, USA

Rare earth elements and yttrium (REYs) are critical elements and valuable commodities due to their limited availability and high demand in a wide range of applications and especially in high-technology products. The increased demand and geopolitical pressures motivate the search for alternative sources of REYs, and coal, coal waste, and coal ash are considered as new sources for these critical elements. This research evaluates the REY potential of coals from Indiana (USA). However, although coal data revealed REY potential, it suffered from sparse samples with complete REY measurements. Therefore, we explore the applicability of machine learning (ML) models and data augmentation techniques to demonstrate their applicability to evaluate REY potential in Indiana, and other areas in coal basins, using selected coal parameters (Al2O3, Fe2O3, C, Ash, S, P, Mo, Zn, and As contents) as covariates (indicators). Due to the relatively small sample size with complete REY data in the Indiana Coal Database, two data augmentation techniques (Random Over-Sampling Examples and Synthetic Minority Over-Sampling Technique) were used. Four machine learning algorithms (linear discriminate analysis, support vector machine, random forest, and artificial neural networks) were applied for modeling REY potential as a classification problem. The results show that application of Synthetic Minority Over-Sampling Technique prior to development of the support vector machine (SVM) models generated the best REY classification with an accuracy of 95%. The encouraging results based on Indiana coal data may suggest that a similar approach can be used for other coal basins for screening the locations with REY potential. Those locations then can be targeted for more detailed geochemical surveys to identify most promising areas and evaluate overall REY resources.

Indiana↗

USGS2023

Broadening the perspectives of sedimentary organic matter analysis to understand Earth system response to change

This paper broadens the description of sedimentary organic matter from the conventional use of coal petrography to include palynological and geochemical sedimentary organic matter. Palynological sedimentary organic matter includes all chemically resistant organic microfossils, such as pollen and spores, dinocysts, microforaminifera (chitinoid-like linings of foraminifera), microscopic algae, charcoal, palynodebris, acritarchs, chitinozoans, and scolecodonts. Geochemical sedimentary organic matter includes organic biomarkers, lipids, and photosynthetic pigments. We provide examples of the use of palynological and geochemical analysis of sedimentary organic matter to understand patterns and impacts of changing climate, fire regimes, hydrologic extremes, water quality, and land change on terrestrial and marine systems through geologic time to provide a long-term perspective to evaluate anthropogenic impacts on the Earth as well as to support forensics investigations.

International Journal of Coal Geology↗

USGS2024

Exploring the uncertainty of machine learning models and geostatistical mapping of rare earth element potential in Indiana coals, USA

Rare earth elements and yttrium (REEs) have a wide range of applications in high- and low-carbon technologies. The strategic significance of REEs has grown due to their expanding applications in manufacturing industries and the constrained availability of these essential resources. This research explores the applicability of machine learning models and their uncertainty for assessing the REE potential in coal beds using various coal parameters as inputs. The work focuses on developing a predictive model based on geological variables, excluding considerations related to potential shifts in the commodities market. The Indiana Coal Quality Database was used as the data source. The promising and unpromising indicators derived from the outlook coefficient of samples from the database were used as the REE potential indicator for machine learning classification models. The filter-based approach with bootstrap was used to evaluate the importance of the coal parameters and their prediction uncertainties. Four machine learning methods (linear discriminant analysis (LDA), random forest (RF), support vector machine (SVM), and artificial neural networks (ANN), a data balancing and augmentation approach (Synthetic Minority Over-sampling Technique), and bootstrap resampling techniques were used for building the models and evaluating their prediction capabilities under uncertainty. It was determined that the SVM bootstrap model with ten-times balanced and augmented data provided superior results compared with other models. Finally, stochastic spatial maps of the REE potential within the coal basin were generated using sequential indicator simulation. The spatial maps of the REE potential showed that a 29% area of the Indiana section of the Illinois coal basin has economic potential of REEs, with 90% confidence.

Indiana↗

USGS2024

Evaluation of coal mine drainage and associated precipitates for radium and rare earth element concentrations

Coal mine drainage (CMD) and associated metal-rich precipitates have recently been proposed as unconventional sources of rare earth elements (REEs). However, the potential occurrence of radium (Ra), a known carcinogen, with the REE-bearing phases has not been investigated. We hypothesized that Ra may occur in solids that are precipitated from CMD as a “radiobarite” solid solution ((Ba,Sr,Ra)SO 4 ) and/or adsorbed with hydrous metal oxides. REEs have been documented to sorb or co-precipitate with iron (Fe), manganese (Mn), and aluminum (Al) oxyhydroxide in CMD solids. Likewise, Ra has been documented to sorb to hydrous Fe and Mn oxides especially where sulfate (SO 4 ) and/or barium (Ba) concentrations are insufficient to precipitate radiobarite. Thus, we conducted the first-ever survey of Ra concentrations in corresponding CMD water and solid samples in the United States. Samples were analyzed from 4 untreated and 9 treated CMD sites in both the bituminous and anthracite coal regions of Pennsylvania across a range of pH and SO 4 concentrations. The dissolved Ra in CMD was relatively low (<0.5 Bq/L), consistent with radiobarite solubility; however, CMD solids were largely composed of amorphous Fe, Al, and Mn oxyhydroxide and silicate minerals. Ra was associated with Mn-enriched CMD solids, upwards of 875 Bq/kg. Total REE + yttrium (Y) content in the CMD solids was enriched upwards of 3600 mg/kg and was significantly correlated with Al content. These preliminary results suggest that REE extraction may target Al-rich solids to avoid Ra in Mn-rich solids.

Pennsylvania↗

USGS2017

Rare earth element behavior during groundwater – seawater mixing along the Kona Coast of Hawaii

Groundwater and seawater samples were collected from nearshore wells and offshore along the Kona Coast of the Big Island of Hawaii to investigate rare earth element (REE) behavior in local subterranean estuaries. Previous investigations showed that submarine groundwater discharge (SGD) is the predominant flux of terrestrial waters to the coastal ocean along the arid Kona Coast of Hawaii. Groundwater and seawater samples were filtered through 0.45 μm and 0.02 μm pore-size filters to evaluate the importance of colloidal and soluble (i.e., truly dissolved ionic species and/or low molecular weight [LMW] colloids) fractions of the REEs in the local subterranean estuaries. Mixing experiments using groundwater collected immediately down gradient from a wastewater treatment facility (WWTF) proximal to the Kaloko-Hanokohau National Historic Park, and more “pristine” groundwater from a well constructed in a lava tube at Kiholo Bay, were mixed with local seawater to study the effect of solution composition (i.e., pH, salinity) on the concentrations and fractionation behavior of the REEs as groundwater mixes with seawater in Kona Coast subterranean estuaries. The mixed waters were also filtered through 0.45 or 0.02 μm filters to ascertain the behavior of colloidal and soluble fractions of the REEs across the salinity gradient in each mixing experiment. Concentrations of the REEs were statistically identical (two-tailed Student t -test, 95% confidence) between the sequentially filtered sample aliquots, indicating that the REEs occur as dissolved ionic species and/or LMW colloids in Kona Coast groundwaters. The mixing experiments revealed that the REEs are released to solution from suspended particles or colloids when Kona Coast groundwater waters mix with local seawater. The order of release that accompanies increasing pH and salinity follows light REE (LREE) > middle REE (MREE) > heavy REE (HREE). Release of REEs in the mixing experiments is driven by decreases in the free metal ion activity in solution and the concomitant increase in the amount of each REE that occurs in solution as dicarbonato complexes [i.e., Ln(CO 3 ) 2 - ] as pH increases across the salinity gradient. Input-normalized REE patterns of Kona Coast groundwater and coastal seawater are nearly identical and relatively flat compared to North Pacific seawater, indicating that SGD is the chief source of these trace elements to the ocean along the Kona Coast. Additionally, REE concentrations of the coastal seawater are between 10 and 50 times higher than previously reported open-ocean seawater values from the North Pacific, further demonstrating the importance of SGD fluxes of REEs to these coastal waters. Taken together, these observations indicate that large-scale removal of REEs, which characterizes the behavior of REEs in the low salinity reaches of many surface estuaries, is not a feature of the subterranean estuary along the Kona Coast. A large positive gadolinium (Gd) anomaly characterizes groundwater from the vicinity of the WWTF. The positive Gd anomaly can be traced to the coastal ocean, providing further evidence of the impact of SGD on the coastal waters. Estimates of the SGD fluxes of the REEs to the coastal ocean along the Kona Coast (i.e., 1.3 – 2.6 mmol Nd day -1 ) are similar to recent estimates of SGD fluxes of REEs along Florida’s east coast and to Rhode Island Sound, all of which points to the importance of SGD as significant flux of REEs to the coastal ocean.

Hawaii↗

USGS2015

The Landscape Evolution Observatory: a large-scale controllable infrastructure to study coupled Earth-surface processes

Zero-order drainage basins, and their constituent hillslopes, are the fundamental geomorphic unit comprising much of Earth's uplands. The convergent topography of these landscapes generates spatially variable substrate and moisture content, facilitating biological diversity and influencing how the landscape filters precipitation and sequesters atmospheric carbon dioxide. In light of these significant ecosystem services, refining our understanding of how these functions are affected by landscape evolution, weather variability, and long-term climate change is imperative. In this paper we introduce the Landscape Evolution Observatory (LEO): a large-scale controllable infrastructure consisting of three replicated artificial landscapes (each 330 m 2 surface area) within the climate-controlled Biosphere 2 facility in Arizona, USA. At LEO, experimental manipulation of rainfall, air temperature, relative humidity, and wind speed are possible at unprecedented scale. The Landscape Evolution Observatory was designed as a community resource to advance understanding of how topography, physical and chemical properties of soil, and biological communities coevolve, and how this coevolution affects water, carbon, and energy cycles at multiple spatial scales. With well-defined boundary conditions and an extensive network of sensors and samplers, LEO enables an iterative scientific approach that includes numerical model development and virtual experimentation, physical experimentation, data analysis, and model refinement. We plan to engage the broader scientific community through public dissemination of data from LEO, collaborative experimental design, and community-based model development.

Arizona↗

USGS2024

The addition of 144Nd atomic mass to routine ICP-MS analysis as a Quick Screening Tool for Approximating Rare Earth Elements (Q-STAR) in natural waters

Rare earth elements (REEs) are a class of critical minerals, all of which can have supply chain vulnerability that impacts economic security. These elements are widely measured in environmental matrices via inductively coupled plasma mass spectrometry (ICP-MS); however, successful quantification can require time-consuming, sample-specific optimization. While a sample-by-sample approach is appropriate for targeted quantification studies, this approach is not suitable for mineral exploration efforts where rapidly screening thousands of samples for the presence of REEs is desired. Here, we demonstrated the use of a Quick Screening Tool for Approximating REEs (Q-STAR) to detect REEs in surface water and groundwater matrices, collected as part of existing environmental studies. A mass-to-charge ratio of 144 ( m / z = 144) was added to an ICP-MS method to screen for REEs in filtered water samples submitted for metals analyses to the U.S. Geological Survey (USGS) National Water Quality Laboratory. We detected the presence of REEs above a reference threshold of 1200 counts per second in 18 % of pre-selected 6626 samples. Using this screened dataset, we mapped estimated dissolved REE concentrations across the United States in relation to ecoregions and underlying geology. Data are constrained to where sample collection took place but nevertheless show estimated aqueous dissolved REE concentrations on a geographic scale that has not yet been studied. To validate Q-STAR, REEs were measured in a USGS standard reference sample, a subset of 88 archived filtered water samples, and in fresh filtered surface water samples. Our targeted analyses demonstrated a strong linear relationship between Q-STAR predicted and measured values in all archived samples for Nd (r 2 = 0.94), and light REEs (LREEs) such as lanthanum (La) (r 2 = 0.93), praseodymium (Pr) (r 2 = 0.94) and samarium (Sm) (r 2 = 0.94). Using Q-STAR screen values, nine field sites were identified and surface water samples recollected to confirm the continued presence of Nd and LREEs. Q-STAR can be used to screen an unlimited number of water samples for the presence of REEs prior to time-intensive and costly quantitative analyses and to generate large REE datasets for further investigation.

Journal of Geochemical Exploration↗

USGS2022

Maize yield forecasts for Sub-Saharan Africa using Earth Observation data and machine learning

Food insecurity continues to grow in Sub-Saharan Africa (SSA). In 2019, chronically malnourished people numbered nearly 240 million, or 20% of the population in SSA. Globally, numerous efforts have been made to anticipate potential droughts, crop conditions, and food shortages in order to improve early warning and risk management for food insecurity. To support this goal, we develop an Earth Observation (EO) and machine-learning-based operational, subnational maize yield forecast system and evaluate its out-of-sample forecast skills during the growing seasons for Kenya, Somalia, Malawi, and Burkina Faso. In general, forecast skills improve substantially during the vegetative growth period (VP) and gradually during the reproductive development period (RP). Thus, mid-season assessment can provide effective early warning months before harvest. Skillful forecasts (Nash Sutcliffe Efficiency (NSE) > 0.6 and Mean Absolute Percentage Error (MAPE) < 20%) appear approximately two dekads after the VP; for example, skillful forecasts appear in May in Kenya and Somalia, January in Malawi, and July in Burkina Faso. During model development, effective EO features are also identified, such as precipitation and available water during VP, and dry days and extreme temperatures in early VP. Compared to monthly standard EO features, sub-monthly (dekadal), non-standard, and serial EO features significantly improve forecast skills by ＋ 0.3 NSE and -10% of MAPE, demonstrating the ability to precisely and effectively capture favorable or detrimental crop development conditions. Finally, skillful forecasts and practical utility are demonstrated in the recent normal and dry years in each region. Overall, the developed yield forecasting system can provide skillful predictions during the growing season, supporting regional and international agricultural decision-making processes, including informing food-security planning and management, thereby helping to mitigate food shortages caused by unfavorable climate conditions.

Global Food Security↗

USGS2015

Rare earth elements in sedimentary phosphate deposits: Solution to the global REE crisis?

The critical role of rare earth elements (REEs), particularly heavy REEs (HREEs), in high-tech industries has created a surge in demand that is quickly outstripping known global supply and has triggered a worldwide scramble to discover new sources. The chemical analysis of 23 sedimentary phosphate deposits (phosphorites) in the United States demonstrates that they are significantly enriched in REEs. Leaching experiments using dilute H 2 SO 4 and HCl, extracted nearly 100% of their total REE content and show that the extraction of REEs from phosphorites is not subject to the many technological and environmental challenges that vex the exploitation of many identified REE deposits. Our data suggest that phosphate rock currently mined in the United States has the potential to produce a significant proportion of the world's REE demand as a byproduct. Importantly, the size and concentration of HREEs in some unmined phosphorites dwarf the world's richest REE deposits. Secular variation in phosphate REE contents identifies geologic time periods favorable for the formation of currently unrecognized high-REE phosphates. The extraordinary endowment, combined with the ease of REE extraction, indicates that such phosphorites might be considered as a primary source of REEs with the potential to resolve the global REE (particularly for HREE) supply shortage.

Gondwana Research↗

USGS2015

Global land cover mapping using Earth observation satellite data: Recent progresses and challenges

Land cover is an important variable for many studies involving the Earth surface, such as climate, food security, hydrology, soil erosion, atmospheric quality, conservation biology, and plant functioning. Land cover not only changes with human caused land use changes, but also changes with nature. Therefore, the state of land cover is highly dynamic. In winter snow shields underneath various other land cover types in higher latitudes. Floods may persist for a long period in a year over low land areas in the tropical and subtropical regions. Forest maybe burnt or clear cut in a few days and changes to bare land. Within several months, the coverage of crops may vary from bare land to nearly 100% crops and then back to bare land following harvest. The highly dynamic nature of land cover creates a challenge in mapping and monitoring which remains to be adequately addressed. As economic globalization continues to intensify, there is an increasing trend of land cover/land use change, environmental pollution, land degradation, biodiversity loss at the global scale, timely and reliable information on global land cover and its changes is urgently needed to mitigate the negative impact of global environment change.

ISPRS Journal of Photogrammetry and Remote Sensing↗

USGS2017

Automated cropland mapping of continental Africa using Google Earth Engine cloud computing

The automation of agricultural mapping using satellite-derived remotely sensed data remains a challenge in Africa because of the heterogeneous and fragmental landscape, complex crop cycles, and limited access to local knowledge. Currently, consistent, continent-wide routine cropland mapping of Africa does not exist, with most studies focused either on certain portions of the continent or at most a one-time effort at mapping the continent at coarse resolution remote sensing. In this research, we addressed these limitations by applying an automated cropland mapping algorithm (ACMA) that captures extensive knowledge on the croplands of Africa available through: (a) ground-based training samples, (b) very high (sub-meter to five-meter) resolution imagery (VHRI), and (c) local knowledge captured during field visits and/or sourced from country reports and literature. The study used 16-day time-series of Moderate Resolution Imaging Spectroradiometer (MODIS) normalized difference vegetation index (NDVI) composited data at 250-m resolution for the entire African continent. Based on these data, the study first produced accurate reference cropland layers or RCLs (cropland extent/areas, irrigation versus rainfed, cropping intensities, crop dominance, and croplands versus cropland fallows) for the year 2014 that provided an overall accuracy of around 90% for crop extent in different agro-ecological zones (AEZs). The RCLs for the year 2014 (RCL2014) were then used in the development of the ACMA algorithm to create ACMA-derived cropland layers for 2014 (ACL2014). ACL2014 when compared pixel-by-pixel with the RCL2014 had an overall similarity greater than 95%. Based on the ACL2014, the African continent had 296 Mha of net cropland areas (260 Mha cultivated plus 36 Mha fallows) and 330 Mha of gross cropland areas. Of the 260 Mha of net cropland areas cultivated during 2014, 90.6% (236 Mha) was rainfed and just 9.4% (24 Mha) was irrigated. Africa has about 15% of the world’s population, but only about 6% of world’s irrigation. Net cropland area distribution was 95 Mha during season 1, 117 Mha during season 2, and 84 Mha continuous. About 58% of the rainfed and 39% of the irrigated were single crops (net cropland area without cropland fallows) cropped during either season 1 (January-May) or season 2 (June-September). The ACMA algorithm was deployed on Google Earth Engine (GEE) cloud computing platform and applied on MODIS time-series data from 2003 through 2014 to obtain ACMA-derived cropland layers for these years (ACL2003 to ACL2014). The results indicated that over these twelve years, on average: (a) croplands increased by 1 Mha/yr, and (b) cropland fallows decreased by 1 Mha/year. Cropland areas computed from ACL2014 for the 55 African countries were largely underestimated when compared with an independent source of census-based cropland data, with a root-mean-square error (RMSE) of 3.5 Mha. ACMA demonstrated the ability to hind-cast (past years), now-cast (present year), and forecast (future years) cropland products using MODIS 250-m time-series data rapidly, but currently, insufficient reference data exist to rigorously report trends from these results.

ISPRS Journal of Photogrammetry and Remote Sensing↗

USGS2018

A 30-m landsat-derived cropland extent product of Australia and China using random forest machine learning algorithm on Google Earth Engine cloud computing platform

Mapping high resolution (30-m or better) cropland extent over very large areas such as continents or large countries or regions accurately, precisely, repeatedly, and rapidly is of great importance for addressing the global food and water security challenges. Such cropland extent products capture individual farm fields, small or large, and are crucial for developing accurate higher-level cropland products such as cropping intensities, crop types, crop watering methods (irrigated or rainfed), crop productivity, and crop water productivity. It also brings many challenges that include handling massively large data volumes, computing power, and collecting resource intensive reference training and validation data over complex geographic and political boundaries. Thereby, this study developed a precise and accurate Landsat 30-m derived cropland extent product for two very important, distinct, diverse, and large countries: Australia and China. The study used of eight bands (blue, green, red, NIR, SWIR1, SWIR2, TIR1, and NDVI) of Landsat-8 every 16-day Operational Land Imager (OLI) data for the years 2013–2015. The classification was performed by using a pixel-based supervised random forest (RF) machine learning algorithm (MLA) executed on the Google Earth Engine (GEE) cloud computing platform. Each band was time-composited over 4–6 time-periods over a year using median value for various agro-ecological zones (AEZs) of Australia and China. This resulted in a 32–48-layer mega-file data-cube (MFDC) for each of the AEZs. Reference training and validation data were gathered from: (a) field visits, (b) sub-meter to 5-m very high spatial resolution imagery (VHRI) data, and (c) ancillary sources such as from the National agriculture bureaus. Croplands versus non-croplands knowledge base for training the RF algorithm were derived from MFDC using 958 reference-training samples for Australia and 2130 reference-training samples for China. The resulting 30-m cropland extent product was assessed for accuracies using independent validation samples: 900 for Australia and 1972 for China. The 30-m cropland extent product of Australia showed an overall accuracy of 97.6% with a producer’s accuracy of 98.8% (errors of omissions = 1.2%), and user’s accuracy of 79% (errors of commissions = 21%) for the cropland class. For China, overall accuracies were 94% with a producer’s accuracy of 80% (errors of omissions = 20%), and user’s accuracy of 84.2% (errors of commissions = 15.8%) for cropland class. Total cropland areas of Australia were estimated as 35.1 million hectares and 165.2 million hectares for China. These estimates were higher by 8.6% for Australia and 3.9% for China when compared with the traditionally derived national statistics. The cropland extent product further demonstrated the ability to estimate sub-national cropland areas accurately by providing an R 2 value of 0.85 when compared with province-wise cropland areas of China. The study provides a paradigm-shift on how cropland maps are produced using multi-date remote sensing. These products can be browsed at www.croplands.org and made available for download at NASA’s Land Processes Distributed Active Archive Center (LP DAAC) https://www.lpdaac.usgs.gov/node/1282 .

ISPRS Journal of Photogrammetry and Remote Sensing↗

USGS2019

Mapping irrigated cropland extent across the conterminous United States at 30 m resolution using a semi-automatic training approach on Google Earth Engine

Accurate and timely information on the distribution of irrigated croplands is crucial to research on agriculture, water availability, land use, and climate change. While agricultural land use has been well characterized, less attention has been paid specifically to croplands that are irrigated, in part due to the difficulty in mapping and distinguishing irrigation in satellite imagery. In this study, we developed a semi-automatic training approach to rapidly map irrigated croplands across the conterminous United States (CONUS) at 30 m resolution using Google Earth Engine. To resolve the issue of lacking nationwide training data, we generated two intermediate irrigation maps by segmenting Landsat-derived annual maximum greenness and enhanced vegetation index using county-level thresholds calibrated from an existing coarse resolution irrigation map. The resulting intermediate maps were then spatially filtered to provide a training data pool for most areas except for the upper midwestern states where we visually collected samples. We then used random samples extracted from the training pool along with remote sensing-derived features and climate variables to train ecoregion-stratified random forest classifiers for pixel-level classification. For ecoregions with a large training pool, the procedure of sample extraction, classifier training, and classification was conducted 10 times to obtain stable classification results. The resulting 2012 Landsat-based irrigation dataset (LANID) identified 23.3 million hectares of irrigated croplands in CONUS. A quantitative assessment of LANID showed superior accuracy to currently available maps, with a mean Kappa value of 0.88 (0.75–0.99), overall accuracy of 94% (87.5–99%), and producer’s and user’s accuracy of the irrigation class of 97.3% and 90.5%, respectively, at the aquifer level. Evaluation of feature importance indicated that Landsat-derived features played the primary role in classification in relatively arid regions while climate variables were important in the more humid eastern states. This methodology has the potential to produce annual irrigation maps for CONUS and provide insights into the field-level spatial and temporal aspects of irrigation.

conterminous United States↗