Contaminants from cretaceous black shale: II. Effect of geology, weathering, climate, and land use on salinity and selenium cycling, Mancos Shale landscapes, southwestern United States

doi:10.1016/j.apgeochem.2013.12.011

Applied Geochemistry

Volume 46, July 2014, Pages 72-84

https://doi.org/10.1016/j.apgeochem.2013.12.011 Get rights and content

Highlights

•
Mancos Shale landscapes accumulate and store salt and Se.
•
Salt and Se reservoirs: dependent on geology, weathering, climate and land use.
•
Saturation paste-extract data accurately predict salt and Se fluxes from soil.
•
Irrigated soil is 16% of watershed area; produces 38% salt and 77% Se river loads.
•
Irrigation of Mancos Shale releases seven times more Se than all of pedogenesis.

Abstract

The Cretaceous Mancos Shale (MS) is a known nonpoint source for a significant portion of the salinity and selenium (Se) loads in the Colorado River in the southwestern United States and northwestern corner of Mexico. These two contaminants pose a serious threat to rivers in these arid regions where water supplies are especially critical. Tuttle et al. (companion paper) investigates the cycling of contaminants in a Colorado River tributary watershed (Uncompahgre River, southwestern Colorado) where the MS weathers under natural conditions. This paper builds on those results and uses regional soil data in the same watershed to investigate the impact of MS geology, weathering intensity, land use, and climate on salt and Se storage in and flux from soils on the natural landscape, irrigated agriculture fields, areas undergoing urban development, and wetlands. The size of salinity and Se reservoirs in the MS soils is quantified. Flux calculations show that during modern weathering, natural landscapes cycle salt and Se; however, little of it is released for transport to the Uncompahgre River (10% of the annual salinity and 6% of the annual Se river loads). When irrigated, salinity and Se loads from the MS soil increase (26% and 57% of the river load, respectively), causing the river to be out of compliance with Federal and State Se standards. During 100 years of irrigation, seven times more Se has been removed from agricultural soil than what was lost from natural landscapes during the entire period of pedogenesis. Under more arid conditions, even less salt and Se are expected to be transported from the natural landscape. However, if wetter climates prevail, transport could increase dramatically due to storage of soluble phases in the non-irrigated soil. These results are critical input for water-resource and land-use managers who must decide whether or not the salinity and Se in a watershed can be managed, what sustainable mitigation strategies are possible, and what landscapes should be targeted. The broader implications include providing a reliable approach for quantifying nonpoint-source contamination from MS and other rock units elsewhere that weather under similar conditions and, together with results from our companion paper, address the complex interplay of geology, weathering, climate, and land use on contaminant cycling in the arid Southwest.

Introduction

Nearly 36 million people and 4.5 million acres of farmland in the United States and Mexico depend on the Colorado River (U.S. Department of Interior, 2005). Serageldin (2000) considers this river one of the most stressed in the world because of increasing salinization, with damages estimated at $380 million per year in the United States alone (U.S. Department of Interior, 2005). In addition to salinization, selenium (Se) loading is of concern because Se concentrations exceed the U.S. Environmental Protection Agency’s freshwater aquatic life chronic criteria (5 μg L⁻¹) in some reaches of the river inhabited by threatened and endangered species of fish (Hamilton, 1998), and Se impacts the habitat of other wildlife in the area such as waterfowl (Presser et al., 1994, Butler et al., 1996). Much of the salinity and most of the Se loads in the Colorado River are sourced from rocks in the upper Colorado River Basin (UCRB). Tuttle and Grauch (2009) estimated that a third of the 2005 solute load in the river at Cisco, Utah (equivalent to dissolution of 1,400,000 t a⁻¹ gypsum), and nearly all the Se (22 t a⁻¹) were derived from the Mancos Shale (MS) in western Colorado.

In 1974, the U.S. Congress passed the Colorado River Basin Salinity Control Act. The purpose of the act is to control the salinity of water delivered to users in the United States and Mexico by constructing, operating, and maintaining projects in the Colorado River Basin. As of 2011, these efforts have reduced salinity loading in the river by 1,200,000 tons per year; however, it is estimated that an additional 660,000 tons per year of salinity are needed to meet the 2030 goal of 1,900,000 tons per year (U.S. Bureau of Reclamation, 2011). To target this additional salinity reduction, a more in-depth understanding of salinity sources and transport in the basin is required. This includes determining the role of natural weathering, climate, and land use on salinity. Because salinity and Se weathering cycles overlap (Tuttle et al., companion paper), reducing salinity will help reduce Se.

A number of studies have described salinity and Se derived from the MS in the UCRB and its tributary watersheds (e.g., Butler et al., 1991, Butler et al., 1996, Clark, 1995, Engberg, 1999, Brummer et al., 2002, Butler and Leib, 2002, Fisher, 2005, Kenney et al., 2009). None of these studies have quantified the geochemical cycles on different MS landscapes that control the distribution of salinity and Se between soil and mobile aqueous phases in the watershed, the goal of this study. This paper builds on results from Tuttle et al. (companion paper), which describe the geochemical processes that control salinity and Se during pedogenesis and erosion of the MS weathering naturally in the Uncompahgre River watershed (Fig. 1). Geochemical surveys provide regional data for salinity and Se on two MS landscapes in the watershed. The first landscape is located upland from the floodplain on the Gunnison Gorge National Conservation Area (GGNCA) and has not been disturbed by development or irrigation (Fig. 1, Fig. 2; referred to as natural landscape). The effects of geology and weathering intensity on reservoirs and fluxes for this landscape are quantified. Comparison with results from the regional survey on the second landscape (floodplain) shows the effect of climate and land use on reservoir size and contaminant fluxes. Salinity and Se cycling on the Uncompahgre River floodplain are further constrained with data from detailed studies of soil irrigated for variable lengths of time and laboratory leaching experiments. These results provide understanding of processes that control water quality in the watershed now and in the future. Together with the companion paper, this study provides resource managers with a quantitative approach for managing landscapes in semiarid to arid climates elsewhere where black shale weathers under similar conditions.

Section snippets

Study area

The Uncompahgre River in southwestern Colorado is a tributary to the Gunnison River that flows into the Upper Colorado River (Fig. 1). The climate in our study area is semiarid with less than 25 cm of precipitation per year. Beginning in 1908, the U.S. Bureau of Reclamation created the Uncompahgre (irrigation) Project (UP), which provides an average of 412 km³ a⁻¹ (334,400 acre-feet a⁻¹; U.S. Bureau of Reclamation, 1982) of diverted Gunnison River water for irrigating 334 km² of farmland in the lower

Sample collection

The samples collected for this study (Fig. 2) include soils across the GGNCA and UP; sediment from Sweitzer Lake; and water from drain tiles, shallow groundwater wells, Sweitzer Lake, and the Uncompahgre River. In 2005, a regional soil survey collected three depths of soil (0–5, 5–25, and 25–45 cm) from sites located on a 2.6-km² (1-mi²) grid covering the GGNCA. Soil samples were collected with an auger, dried, and ground to less than 75 μm for bulk chemistry, and sieved to less than 2 mm for

Natural Mancos Shale landscapes

Tuttle et al. (companion paper) described the processes that occur during natural weathering and pedogenesis of the MS. Results showed that salt and Se reservoirs vary in size across the natural MS landscape and appear to be sensitive to a variety of factors. To capture the effect of two important variables (geology and extent of weathering), the regional GGNCA soil data (0–45 cm depth interval) are divided with respect to subsurface geology (Juana Lopez and Niobrara Members) and relative extent

Cycling of salinity and Se in the Uncompahgre River watershed

Cycling of salinity and Se for the lower portion of the Uncompahgre River watershed is reflected in three budgets calculated from: (1) data for surface water in the watershed (water budget); (2) the GGNCA regional data scaled to include all natural MS landscapes in the watershed outside of the UP (natural landscape budget); and (3) regional data for irrigated, non-irrigated, and wetland soil (UP landscape budget). Details on the construction of these budgets are in the online supplement.

Implication of results

Fig. 8 summarizes the contribution of each Mancos Shale (MS) landscape to the salinity and Se loads in the Uncompahgre River as calculated in the scaled GGNCA and UP budgets. The MS landscapes account for 25% of the area in the Uncompahgre River watershed and supplies 38% of the Uncompahgre River’s salinity load. Of the 38%, irrigated soil supplies 66% (26% in tailwater and 40% in tile drainage), 10% is from non-irrigated land in the UP, and 24% from the natural landscape. We assume that none

Acknowledgements

We thank Karen Tucker, Jim Ferguson, Amanda Clements, and Dennis Murphy of the U.S. Bureau of Land Management for managing logistics and providing access to the Gunnison Gorge National Conservation Area. We also thank Jean Morrison, Joshua Linard, and Janet Slate of the U.S. Geological Survey and two anonymous reviewers for their thoughtful comments that have greatly improved this paper. Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by

References (29)

C.A. Rice et al.
The analysis of forms of sulfur in ancient sediments and sedimentary rock: comments and cautions
Chem. Geol.
(1993)
M.L. Tuttle et al.
An analytical scheme for determining forms of sulphur in oil shales and associated rocks
Talanta
(1986)
D.W. Anning et al.
Dissolved Solids in Basin-fill Aquifers and Major River Systems in the Southwestern United States
U.S. Geol. Surv. Sci. Invest. Rep. 2006-5313
(2007)
Barnhart, R.A., 1957. Chemical Factors Affecting the Survival of Game Fish in a Western Colorado Reservoir. Master of...
Brummer, J., Fahy, J., Schaefer, S., 2002. Relationship of selenium in drainage water to the selenium content of...
Butler, D.L., Leib, K.J., 2002. Characterization of Selenium in the Lower Gunnison River Basin, Colorado, 1988–2000....
Butler, D.L., Krueger, R.P., Osmundson, B.C., Thompson, A.L., McCall, S.K., 1991. Reconnaissance Investigation of Water...
Butler, D.L., Wright, W.G., Steward, K.G., Campbell Osmundson, B., Krueger, R.P., Crabtree, D.W., 1996. Detailed Study...
Clark, S.L., 1995. The Distribution of Selenium in the Upper Colorado River. Dissertation, Arizona State University,...
Elliott, J.E., Herring, J.R., Ingersoll, G.P., Kosovick, J.J., Fahy, J., 2007. Rainfall–Runoff and Erosion Data from...

R.A. Engberg

Selenium budgets for Lake Powell and the Upper Colorado River

J. Am. Water Res. Assoc.

(1999)

Fisher, F.S., 2005. Uncompahgre River Basin Selenium Phytoremediation: Final Report for EPA Grant #WQC 01-00057....

S.J. Hamilton

Selenium effects on endangered fish in the Colorado River Basin

Hansen, W.R., 1971. Geologic Map of the Black Canyon of the Gunnison River and Vicinity, Western Colorado. U.S. Geol....

Cited by (35)

Irrigation leads to new Se-toxicity paddy fields in and around typical Se-toxicity area
2023, Science of the Total Environment
Although the issue has been of much concern and has subsequently been controlled for years, the environmental risk of excess selenium (Se) in farmlands still has not been eliminated in Se-toxicity areas. Different types of farmland utilization can change Se behavior in soil. Thus, located field monitoring and surveys of various farmland soils in and around typical Se-toxicity areas spanning eight years were conducted in the tillage layer and deeper soils. The source of new Se contamination in farmlands was traced along the irrigation and natural waterway. This research indicated that 22 % of paddy fields increased to Se-toxicity in surface soil led by irrigation with high-Se river water. Selenate is the dominant Se species in rivers (90 %) originating from geological background areas with high Se. Both soil organic matter (SOM) and amorphous iron content played important roles in the fixation of input Se. Thus, available Se was increased by more than twofold in paddy fields. The release of residual Se and eventual bounding by organic matter is commonly observed, thus suggesting that stable soil Se availability seems sustainable for a long time. This study is the first report in China that shows how new soil Se-toxicity farmland is caused by high-Se water irrigation. This research warns that external attention should be paid to the selection of irrigation water in high-Se geological background areas to avoid new Se contamination.
Selenium mobility in a major Chalk aquifer (Lille metropolis, northern France): Contaminants cycles driven by geology, redox processes and pumping
2021, Chemical Geology
Chalk groundwater in northern France presents selenium (Se) concentrations from <10 to 70 μg L⁻¹, partly exceeding the latest European Framework Directive's drinking-water limit value of 20 μg L⁻¹. Better to understand the heterogeneous dynamics of Se, we used a combination of geochemical, isotopic and geophysical tools on a study site belonging to the Metropolis of Lille, France. This approach provided a fine understanding of the Se fate in a dynamic redox system constrained by geology, related redox processes and pumping, illustrating how local versus global controls affect the Se cycling. A remarkable redox sequence controls element transfers in groundwater with a progressive creation of reductive conditions. Highlighted by geophysical tools, a wide fracture corridor in the Chalk formation disrupts the geological setting of this redox sequence. Under pumping, this corridor allows the mixing of oxygenated groundwater with groundwater under reducing conditions. The evolution of isotopic compositions of sulphate molecules confirms the global reduction trend of sulphates, while pyrite oxidation occurs very locally, together with high Se concentrations. Pyrite is expected to play a predominant role in Se mobility in the Chalk aquifer. Se and other redox sensitive elements (Fe, Mn, N and S) undergo multiple redox cycles, resulting in a Se-rich redox front that migrates downward over time within the water-level fluctuation zone of the porous Chalk. With the decreasing trend of water levels caused by global changes, a Se stock could be immobilized in the unsaturated zone, but nitrate content and redox conditions in the saturated zone will be major drivers for Se mobility.
Impact of sedimentary environment on pore parameters of thermally mature Permian shale: A study from Kommugudem Formation of Krishna Godavari Basin, India
2021, Marine and Petroleum Geology
The pores are significant factors to assess the fracture system of a shale reservoir. The present study evaluates organic matter-associated parameters, redox condition, paleoclimate and paleosalinity and their role on the pore structure of thermally mature Kommugudem shale of Krishna Godavari Basin. The XRD analysis supported by FE-SEM mainly indicates silica-dominated lithotype and clay-rich siliceous mudstone in the Kommugudem Formation. Intercrystalline and intracrystalline pores were observed in association with mineral and organic matter that provide storage sites for hydrocarbon. Excellent TOC content and genesis of gas prone, type III kerogen dominantly observed in admixture with type II or type IV kerogen. The clay, TOC and Tmax have a positive correlation with micropore volume indicating a good contribution of clay micropore and associated organic matter. Fractal dimensions were estimated using the fractal FHH method with an average value of 2.65 suggesting that the pore surfaces are rough and exhibit complex and heterogeneous pore structure. The Kommugudem Formation constitutes detrital sediments sourced from felsic volcanic rocks in humid and warm climate along the passive tectonic continental margin. The redox and paleosaline investigations reveal the fluvio-marine transitional environment preserved under suboxic to oxic conditions. The REEs and trace elements are positively correlated with pore volume indicating their possible role in the pore genesis of silicate minerals and corroborating the positive influence of the environment. The redox proxies are strongly correlated with TOC and pore volume supports the probable development of organopores. The investigation reveals the significant role of multiproxy paleoenvironmental indices to understand the complex mechanism of pore characteristics.
Key driving factors of selenium-enriched soil in the low-Se geological belt: A case study in Red Beds of Sichuan Basin, China
2021, Catena
Selenium (Se) is an essential micronutrient for humans given its varying health benefits. It is generally recognized that China has a wide belt of low-Se soil stretching from the northeast to southwest. Nevertheless, there are Se-enriched areas distributed in the low-Se belt of China. However, the quantificational relationships among soil properties, topographic characteristics, parent materials, land use and soil Se content in those Se-enriched soils remain to be elucidated. Similarly, the key driving factors of the Se-enriched soil in the low-Se geological belt need to be documented. These aims could be an useful basis for evaluating the health of the soil ecosystem (in terms of Se toxicity or deficiency) and the potential intake of Se by humans from soils to food crops and animal products. To solve the above questions, Jiangjin district, Chongqing City, as an area in low Se red beds, was selected, and 156 topsoil samples were collected to explore the relationships between Se content in topsoil under various land use type and soil properties (pH, organic matter (OM), major elements content in topsoil) and topographic characteristics such as elevation (H), slope (SL), and topographic wetness index (TWI) and geological condition (stratum). Geostatistics, principal component analysis (PCA) and Geodetector were used to analyze the controlling factors of Se distribution in topsoil. The results show that in the analyzed topsoil (1) Se contents vary from 0.039 to 1.110 mg/kg, with a mean of 0.315 mg/kg, and higher than the background value of Chinese soil (0.290 mg/kg). 82.3% are classified as having moderate Se levels (Se > 0.175 mg/kg). (2) The Se contents in northeast of Jiangjin district are higher (Se > 0.315 mg/kg) than the midwest (Se < 0.315 mg/kg). (3) The key controlling factor of the soil Se distribution is OM. In addition, the soil weathering and leaching process and pedogenic rock are the secondary factors controlling Se distribution. Together, these findings reveal that Se-rich soil of low Se belt tends to be distributed in regions with higher OM, stronger leaching, and carbonate parent materials. These observations are beneficial to explore the Se-rich soil resources in low Se region in China.
Parent material modulates land use effects on soil selenium bioavailability in a selenium-enriched region of southwest China
2020, Geoderma
Citation Excerpt :
Land use has been well recognized to be a major factor causing far-reaching consequences for element biogeochemical cycles (Xiao et al., 2009). To date, however, most studies examining the influence of land use focus on macroelements such as carbon, nitrogen, and phosphorus (Herpin et al., 2002; Potthast et al., 2012; Cherubin et al., 2016), not the trace elements (Anguelov and Anguelova, 2009; Tu et al., 2013; Islam et al., 2015), especially Se (Tuttle et al., 2014; Yanai et al., 2015; Plak and Bartminski, 2017). The direction and magnitude of land use effect on soil Se dynamics is not well known at present.
Land use effects on the biogeochemical cycling of the essential trace element selenium (Se) and underlying mechanisms are not well understood yet. Here, total Se (Se_total) and its potential bioavailability, as represented by phosphate extractability (Se_phosphate), were evaluated in top soils (0–15 cm) of cropland and woodland over two contrasting geological parent materials (i.e. limestone versus clasolite) in a Se-enriched region of southwest China. After about twenty years of conversion of land use from cropland to woodland, Se_total in soils significantly (P < 0.05) and marginally significantly (P < 0.1) increased over limestone and clasolite, respectively. In terms of Se_phosphate, it was not significantly changed after land use change from cropland to woodland in soil over limestone, whereas remarkably increased in soil over clasolite (P < 0.01). Furthermore, a significant positive relationship between Se_phosphate and Se_total was found in soil over clasolite, but not in soil over limestone. Land use conversion from cropland to woodland caused a decrease in the proportion of Se_phosphate in Se_total over limestone, while the opposite was found over clasolite. In comparison, the proportion of Se_phosphate in Se_total was greater in cropland over limestone than that in cropland over clasolite, indicating that limestone derived soils may have an advantage of producing Se-rich food over clasolite derived soils. SOC content and quality (i.e. C:N ratio) crucially determined Se accumulation and bioavailability in soil over limestone, while pH played a vital role in soil over clasolite. Taken together, our results provide strong evidence that land use effects on Se biogeochemistry can be substantially modulated by the underlying geology, and have important practical implications for effective utilization and management of Se-enriched soil resources in other parts of the world.
Impact of land use/land cover change on the topsoil selenium concentration and its potential bioavailability in a karst area of southwest China
2020, Science of the Total Environment
Selenium (Se) is an essential micronutrient for human health, and its abundance and potential bioavailability in the soil are of increasing concern worldwide. To date, how total soil Se and its bioavailability would respond to human disturbance or future environmental change is not yet clear, and associated controlling factors remain incompletely understood. Here, we collected soil samples (0–15 cm) from different land use/land cover types, including active cropland, grassland, shrubland, and secondary forest, in a Se-enriched area of Guangxi, southwest China. Total Se concentration and its potential bioavailability, as estimated by phosphate extractability, were investigated. Total soil Se concentration (Se_total) for all samples ranged from 220 to 1820 μg kg⁻¹, with an arithmetic average value of 676 ± 24 μg kg⁻¹ (Mean ± SE, the same below). The concentration of phosphate extractable Se (Se_phosphate) varied between 1 and 257 μg kg⁻¹, with an arithmetic mean value of 79 ± 5 μg kg⁻¹, accounting for on average 13 ± 1% of the Se_total. Among the four land use/land cover types, Se_total and Se_phosphate were generally more enriched in the secondary forest than those in the grassland and cropland. The content of soil organic carbon (SOC) was the overriding edaphic factor controlling the abundance and potential bioavailability of Se in topsoils. In addition, climatic variables such as mean annual precipitation and mean annual temperature were also key factors affecting the abundance and potential bioavailability of soil Se. Our results suggest that changes in land use/land cover types may deeply influence Se biogeochemistry likely via alterations in soil properties, particularly SOC content.

View all citing articles on Scopus

View full text

Published by Elsevier Ltd.