Search USGS⌕ Search

SEARCH · Search USGS

Results for “SOIL”

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

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

At least 847 records · Page 47Linked to original sources
USGS2012

History and progress of the North American Soil Geochemical Landscapes Project, 2001-2010

In 2007, the U.S. Geological Survey, the Geological Survey of Canada, and the Mexican Geological Survey initiated a low-density (1 site per 1600 km 2 , 13323 sites) geochemical and mineralogical survey of North American soils (North American Soil Geochemical Landscapes Project). Sampling and analytical protocols were developed at a series of workshops in 20032004 and pilot studies were conducted from 20042007. The ideal sampling protocol at each site includes a sample from 05 cm depth, a composite of the soil A horizon, and a sample from the soil C horizon. The <2-mm fraction of each sample is analyzed for Al, Ca, Fe, K, Mg, Na, S, Ti, Ag, Ba, Be, Bi, Cd, Ce, Co, Cr, Cs, Cu, Ga, In, La, Li, Mn, Mo, Nb, Ni, P, Pb, Rb, Sb, Sc, Sn, Sr, Te, Th, Tl, U, V, W, Y, and Zn by inductively coupled plasma-mass spectrometry and inductively coupled plasma-atomic emission spectrometry following a near-total digestion in a mixture of HCl, HNO 3 , HClO 4 , and HF. Separate methods are used for As, Hg, Se, and total C on this same size fraction. The major mineralogical components are determined by a quantitative X-ray diffraction method. Sampling in the conterminous U.S. was completed in 2010 (c. 4800 sites) with chemical and mineralogical analysis currently underway. In Mexico, approximately 66% of the sampling (871 sites) had been done by the end of 2010 with completion expected in 2012. After completing sampling in the Maritime provinces and portions of other provinces (472 sites, 7.6% of the total), Canada withdrew from the project in 2010. Preliminary results for a swath from the central U.S. to Florida clearly show the effects of soil parent material and climate on the chemical and mineralogical composition of soils. A sample archive will be established and made available for future investigations.

Earth Science Frontiers↗
USGS2017

Iron isotope systematics of shale-derived soils as potentially influenced by small mineral particle loss

Loss of small mineral particles from soil has been suggested as a process that can produce net isotopic fractionation in the remaining soil. We extracted water dispersible colloids (WDCs) from bulk soil collected at the Susquehanna/Shale Hills Critical Zone Observatory (SSHO) and measured their Fe isotopic composition for comparison to published data from the site. The goal was to explain soil δ56Fe values that become lighter as Fe is lost from soil. The range of δ56Fe values for WDCs was 0.22 to 0.59 ‰, barely intersecting the value of ~0.8 ± 0.3‰ predicted by mass balance for particulate Fe loss by a previous study. The WDCs extracted likely represent a mixture of unfractionated Fe inherited from shale minerals and secondary Fe fractionated by weathering zone processes. Thus, although the WDC compositions do not confirm small mineral particle losses as causing overall Fe isotope fractionation in SSHO soils, they are compatible with that interpretation.

Conference Paper↗
USGS1995

Soil-solution chemistry in a low-elevation spruce-fir ecosystem, Howland, Maine

Soil solutions were collected monthly by tension and zero-tension lysimeters in a low-elevation red spruce stand in east-central Maine from May 1987 through December 1992. Soil solutions collected by Oa tension lysimeters had higher concentrations of most constituents than the Oa zero-tension lysimeters. In Oa horizon soil solutions growing season concentrations for SO 4 , Ca, and Mg averaged 57, 43, and 30 μmol L −1 in tension lysimeters, and 43, 28, and 19 μmol L −1 in zero-tension lysimeters, respectively. Because tension lysimeters remove water held by the soil at tensions up to 10 kPa, solutions are assumed to have more time to react with the soil compared to freely draining solutions collected by zero-tension lysimeters. Solutions collected in the Bs horizon by both types of collectors were similar which was attributed to the frequency of time periods when the water table was above the Bs lysimeters. Concentrations of SO 4 and NO 3 at this site were lower than concentrations reported for most other eastern U.S. spruce-fir sites, but base cation concentrations fell in the same range. Aluminum concentrations in this study were also lower than reported for other sites in the eastern U.S. and Ca/Al ratios did not suggest inhibition of Ca uptake by roots. Concentrations of SO 4 , Ca, K, and Cl decreased significantly in both the Oa and Bs horizons over the 56-month sampling period, which could reflect decreasing deposition rates for sulfur and base cations, climatic influences, or natural variation. A longer record of measured fluxes will be needed to adequately define temporal trends in solution chemistry and their causes.

Maine↗
USGS2005

Microbial responses and nitrous oxide emissions during wetting and drying of organically and conventionally managed soil under tomatoes

The types and amounts of carbon (C) and nitrogen (N) inputs, as well as irrigation management are likely to influence gaseous emissions and microbial ecology of agricultural soil. Carbon dioxide (CO2) and nitrous oxide (N2O) efflux, with and without acetylene inhibition, inorganic N, and microbial biomass C were measured after irrigation or simulated rainfall in two agricultural fields under tomatoes (Lycopersicon esculentum). The two fields, located in the California Central Valley, had either a history of high organic matter (OM) inputs ("organic" management) or one of low OM and inorganic fertilizer inputs ("conventional" management). In microcosms, where short-term microbial responses to wetting and drying were studied, the highest CO2 efflux took place at about 60% water-filled pore space (WFPS). At this moisture level, phospholipid fatty acids (PLFA) indicative of microbial nutrient availability were elevated and a PLFA stress indicator was depressed, suggesting peak microbial activity. The highest N 2O efflux in the organically managed soil (0.94 mg N2O-N m-2 h-1) occurred after manure and legume cover crop incorporation, and in the conventionally managed soil (2.12 mg N2O-N m-2 h-1) after inorganic N fertilizer inputs. Elevated N2O emissions occurred at a WFPS >60% and lasted <2 days after wetting, probably because the top layer (0-150 mm) of this silt loam soil dried quickly. Therefore, in these cropping systems, irrigation management might control the duration of elevated N2O efflux, even when C and inorganic N availability are high, whereas inorganic N concentrations should be kept low during times when soil moisture cannot be controlled.

Biology and Fertility of Soils↗
USGS1993

Influence of Pb on microbial activity in Pb-contaminated soils

Investigations of the influence of Pb on soil microbial communities have focused on Pb concentrations of 1 g kg-’ or less (Barkay et al., 1985; Capone et al., 1983; Chang and Broadbent, 1981; Doelman and Haanstra, 1979; Trevors et al., 1985). However, a number of environments exist in which Pb concentrations exceed 1 g kg-’ dry soil (Davenport and Peryea, 1991; Davis et al., 1992; Bisessar, 1982). Bisessar (1982) reported an inverse correlation between Pb concentration and the bacterial population size in soil near a secondary lead smelter. However, similar trends in the concentrations of Pb, As, Cd, and Cu at the site make it difficult to attribute the reductions in population size to Pb alone. Although the effects on microbial carbon mineralization of Pb concentrations as high as 20,000 g kg-’ dry soil were investigated by Debosz et a/. (1985), differences in pH between Pb treatments and the lack of controls for abiotic CO, evolution make the results of the study equivocal. Our purpose was to examine the effects of g kg-’ Pb concentrations on the growth and productivity of soil microbial communities.

Soil Biology and Biochemistry↗
USGS2007

Assessing exotic plant species invasions and associated soil characteristics: A case study in eastern Rocky Mountain National Park, Colorado, USA, using the pixel nested plot design

Rocky Mountain National Park (RMNP), Colorado, USA, contains a diversity of plant species. However, many exotic plant species have become established, potentially impacting the structure and function of native plant communities. Our goal was to quantify patterns of exotic plant species in relation to native plant species, soil characteristics, and other abiotic factors that may indicate or predict their establishment and success. Our research approach for field data collection was based on a field plot design called the pixel nested plot. The pixel nested plot provides a link to multi-phase and multi-scale spatial modeling-mapping techniques that can be used to estimate total species richness and patterns of plant diversity at finer landscape scales. Within the eastern region of RMNP, in an area of approximately 35,000 ha, we established a total of 60 pixel nested plots in 9 vegetation types. We used canonical correspondence analysis (CCA) and multiple linear regressions to quantify relationships between soil characteristics and native and exotic plant species richness and cover. We also used linear correlation, spatial autocorrelation and cross correlation statistics to test for the spatial patterns of variables of interest. CCA showed that exotic species were significantly (P < 0.05) associated with photosynthetically active radiation (r = 0.55), soil nitrogen (r = 0.58) and bare ground (r = -0.66). Pearson's correlation statistic showed significant linear relationships between exotic species, organic carbon, soil nitrogen, and bare ground. While spatial autocorrelations indicated that our 60 pixel nested plots were spatially independent, the cross correlation statistics indicated that exotic plant species were spatially associated with bare ground, in general, exotic plant species were most abundant in areas of high native species richness. This indicates that resource managers should focus on the protection of relatively rare native rich sites with little canopy cover, and fertile soils. Using the pixel nested plot approach for data collection can facilitate the ecological monitoring of these vulnerable areas at the landscape scale in a time- and cost-effective manner. ?? 2006 Elsevier B.V. All rights reserved.

Applied Soil Ecology↗
USGS2006

Distribution and abundance of fungi in the soils of Taylor Valley, Antarctica

The occurrence and distribution of culturable fungi in Taylor Valley, Antarctica was assessed in terms of soil habitat. Soil transects throughout the valley revealed differential habitat utilization between filamentous and non-filamentous (yeast and yeast-like) fungi. In addition, there were significant differences in species distribution patterns with respect to soil pH, moisture, distance from marine coastline, carbon, chlorophyll a, salinity, elevation and solar inputs. Filamentous fungal abundance is most closely associated with habitats having higher pH, and soil moistures. These close associations were not found with yeast and yeast-like fungi demonstrating that yeast and yeast-like fungi utilize a broader range of habitat. An intensive survey of the Victoria Land is necessary to gain a better understanding of their role in soil functioning and nutrient cycling processes. ?? 2006 Elsevier Ltd. All rights reserved.

Soil Biology and Biochemistry↗
USGS2014

Soil fluxes of methane, nitrous oxide, and nitric oxide from aggrading forests in coastal Oregon

Soil exchanges of greenhouse and other gases are poorly known for Pacific Northwest forests where gradients in nutrient availability and soil moisture may contribute to large variations in fluxes. Here we report fluxes of methane (CH 4 ), nitrous oxide (N 2 O), and nitric oxide (NO) over multiple seasons from three naturally N-rich, aggrading forests of coastal Oregon, USA. Mean methane uptake rates (3.2 mg CH 4 m −2 d −1 ) were high compared with forests globally, negatively related to water-filled pore space (WFPS), but unrelated to N availability or temperature. Emissions of NO (6.0 μg NO–N m −2 h −1 ) exceeded N 2 O (1.4 μg N 2 O–N m −2 h −1 ), except when WFPS surpassed 55%. Spatial variation in NO fluxes correlated positively with soil nitrate concentrations (which generally exceeded ammonium concentrations, indicating the overall high N status for the sites) and negatively with soil pH, and at one site increased with basal area of N 2 -fixing red alder. Combined NO and N 2 O emissions were greatest from the site with highest annual net N mineralization and lowest needle litterfall C/N. Our findings of high CH 4 uptake and NO/N 2 O ratios generally >1 most likely reflect the high porosity of the andic soils underlying the widespread regenerating forests in this seasonally wet region.

Oregon↗
USGS2022

Mechanisms for retention of low molecular weight organic carbon varies with soil depth at a coastal prairie ecosystem

Though primary sources of carbon (C) to soil are plant inputs (e.g., rhizodeposits), the role of microorganisms as mediators of soil organic carbon (SOC) retention is increasingly recognized. Yet, insufficient knowledge of sub-soil processes complicates attempts to describe microbial-driven C cycling at depth as most studies of microbial-mineral-C interactions focus on surface horizons. We leveraged a well-studied paleo-marine terrace (90 ka) located near Santa Cruz, CA, to characterize the short-term (days to weeks) and intermediate-term (months to years) fate of two low molecular weight organic carbon. compounds at three depths in the soil profile (∼25 cm, A horizon; ∼75 cm A/B transition; and ∼125 cm, B horizon). We employed isotopically-labeled glucose (GLU) and oxalic acid (OXA) to represent two common classes of rhizodeposits: carbohydrates and organic acids. Using a combination of laboratory (9 d) and field (490 d) incubations, we traced the fate of GLU-C and OXA-C through dissolved-, metal-associated-, and microbially-respired CO2 and bulk SOC pools. Our results suggest new SOC retention (i.e., defined as 13C label identified in solid or aqueous fractions) over intermediate time frames (490 d) is correlated with patterns in short-term (9 d) cycling dynamics, which in turn is related to the theoretical efficiency by which microorganisms process each substrate. For all horizons (A, A/B, and B) GLU-C was converted to CO2 more quickly than OXA-C with modeled decomposition rates ∼2–4 times faster for GLU depending on microbial density (higher in A than B horizon). The faster decomposition rates of GLU-C increased fractional recovery (0.399 ± 0.026 to 0.504 ± 0.030 for GLU-C) compared to OXA-C (0.035 ± 0.003 to 0.127 ± 0.010) among all horizons in our field experiment (490 d). Though the overall proportion of GLU-C recovered in solid fractions did not vary significantly with horizon, based on 13C recovered in aqueous fractions the apparent mechanism for retention did. After the 9-d laboratory incubation, fractional recovery for GLU-C among C pools associated with microbial biomass was almost 20× higher than OXA-C (0.192 versus 0.010, respectively across all horizons). More than a year later, 43–46% of GLU-C retained in the field incubation was extractable with a neutral salt (representing a pool of soil C residing within or available to microbial biomass) among A and A/B horizons, while only 6% of retained GLU-C was similarly extractable in the B horizon. Thus, it appears among depths with higher microbial density (A, A/B horizons), anabolic recycling is the most likely process contributing to the persistence of glucose C, whereas abiotic sinks contributed more to intermediate-term stability for GLU-C in the B horizon. By contrast, most OXA-C was lost, presumably as CO2, over the short-term from the A and A/B horizons (fractional recovery: 0.136 ± 0.011 and 0.091 ± 0.002, respectively). However, though substantially lower than GLU-C recovered at the conclusion of our field experiment, the fraction of oxalic acid C retained in the B horizon over both short- (0.72 ± 0.037) and intermediate-time (0.127 ± 0.010) frames was several-fold higher than for overlying horizons. The specific process(es) (e.g., more efficient microbial utilization, metal-organic complexation, direct adsorption to the mineral matrix, etc.) contributing to higher retention for OXA-C at depth are discussed but remain unresolved.

Soil Biology and Biochemistry↗
USGS2003

100 years of Pb deposition and transport in soils in Champaign, Illinois, U.S.A

In Illinois, atmospheric deposition is one major source of heavy metal inputs to agricultural land. The atmospheric Pb deposition and transport record in agricultural soils in Champaign, Illinois, was established by studying surface and subsurface soil samples collected during the past 100 years from the Morrow Plots on the campus of the University of Illinois at Urbana-Champaign. The Pb content in the soil samples was measured and the Ph deposition fluxes were calculated. The Pb content in surface soils increased sharply in the first half of the 20th century, and stayed invariant since. The maximum Pb flux from the atmosphere was estimated to be 27 (??14) ??g cm-2 yr-1 around 1940. The major pollution source for this increase probably was residential coal burning. It was estimated that in 50 yr, more than 50% of the Pb input had been lost from the surface soils.

Water, Air, & Soil Pollution↗
USGS2021

Linking decomposition rates of soil organic amendments to their chemical composition

The stock of organic carbon contained within a soil represents the balance between inputs and losses. Inputs are defined by the ability of vegetation to capture and retain carbon dioxide, effects that management practices have on the proportion of captured carbon that is added to soil and the application organic amendments. The proportion of organic amendment carbon retained is defined by its rate of mineralisation. In this study, the rate of carbon mineralisation from 85 different potential soil organic amendments (composts, manures, plant residues and biosolids) was quantified under controlled environmental conditions over a 547 day incubation period. The composition of each organic amendment was quantified using nuclear magnetic resonance and mid- and near-infrared spectroscopies. Cumulative mineralisation of organic carbon from the amendments was fitted to a two-pool exponential model. Multivariate chemometric algorithms were derived to allow the size of the fast and slow cycling pools of carbon to be predicted from the acquired spectroscopic data. However, the fast and slow decomposition rate constants could not be predicted suggesting that prediction of the residence time of organic amendment carbon in soil would likely require additional information related to soil type, environmental conditions, and management practices in use at the site of application.

Soil Research↗
USGS2013

Arsenic and mercury in the soils of an industrial city in the Donets Basin, Ukraine

Soil and house dust collected in and around Hg mines and a processing facility in Horlivka, a mid-sized city in the Donets Basin of southeastern Ukraine, have elevated As and Hg levels. Surface soils collected at a former Hg-processing facility had up to 1300 mg kg −1 As and 8800 mg kg −1 Hg; 1M HCl extractions showed 74–93% of the total As, and 1–13% of the total Hg to be solubilized, suggesting differential environmental mobility between these elements. In general, lower extractability of As and Hg was seen in soil samples up to 12 km from the Hg-processing facility, and the extractable (1M HCl, synthetic precipitation, deionized water) fractions of As are greater than those for Hg, indicating that Hg is present in a more resistant form than As. The means (standard deviation) of total As and Hg in grab samples collected from playgrounds and public spaces within 12 km of the industrial facility were 64 (±38) mg kg −1 As and 12 (±9.4) mg kg −1 Hg; all concentrations are elevated compared to regional soils. The mean concentrations of As and Hg in dust from homes in Horlivka were 5–15 times higher than dust from homes in a control city. Estimates of possible exposure to As and Hg through inadvertent soil ingestion are provided.

Donets Basin↗
USGS1986

Transport of chloride ion in a water-unsaturated soil exhibiting anion exclusion

Miscible displacement techniques were used to create Cl - concentration profiles in unsaturated laboratory columns of Delhi sand (Typic Xeropsamments), each having a nearly uniform water content. The three steady flow rates used resulted in three different, average water contents. Chloride concentrations near the top of the column were smaller and penetration of Cl - in the column was deeper than expected assuming that Cl - is a noninteracting solute. Such observations indicate the presence of anion exclusion. This interpretation is further substantiated by chloride and tritium breakthrough curves obtained from a saturated column of the same soil. The saturated experiments show that tritium occupies the entire measured pore volume of the column, but that Cl - is restricted to a smaller pore volume. The formulation of the conventional convection-dispersion theory for solute transport in soil which includes anion exclusion resulted in model calculations that fitted the unsaturated Cl - concentration profiles quite well. The dispersion coefficients obtained for the unsaturated profiles increase with water velocity and are lower than those previously reported for comparable water velocities in the same but saturated soil. The dispersivity of the unsaturated soil is also smaller than that reported for the saturated soil. For the experimental conditions used, the effective Cl - exclusion volume was found to be independent of water content and velocity and occupied about ten percent of the unsaturated water content.

Soil Science Society of America Journal↗
USGS1986

Recovery of compacted soils in Mojave Desert ghost towns

Residual compaction of soils was measured at seven sites in five Mojave Desert ghost towns. Soils in these Death Valley National Monument townsites were compacted by vehicles, animals, and human trampling, and the townsites had been completely abandoned and the buildings removed for 64 to 75 yr. The soils studied (generally sandy, mixed, Typic Calciorthids) were derived from granitic or volcanic alluvium at elevations from 1310 to 1730 m. Compaction measurements in the townsites, including penetration depth, penetration resistance, bulk density, and peak shear stress, indicated that only one site had completely recovered to ambient soil conditions after 75 yr. Recovery times extrapolated using a linear recovery model ranged from 80 to 140 yr and averaged 100 yr. The recovery times were related to elevation, suggesting freeze-thaw loosening as an important factor in ameliorating soil compaction in the Mojave Desert.

California↗
USGS1991

Collimated neutron probe for soil water content measurements

The use of uncollimated (undirected) neutron moisture meters is common in the biological and geophysical sciences. A collimated neutron probe was designed to enable measurements in specific directions from the access tube. To determine the size and shape of soil volume affecting the neutron counts, experiments were conducted to evaluate: (i) the vertical distance of soil above and below the probe that influences neutron counts, (ii) the horizontal distance away from the probe into the soil that influences neutron counts, (iii) the angle of soil viewed by the probe from the collimator, and (iv) the three-dimensional thermal-neutron density field. The distance and the angular dimensions of the volume of influence were defined as the horizontal distance of neutron penetration from the edge of the probe, the vertical distance above and below the center of the effective measurement point of the probe, and the angle from the center of the probe, which would allow the determination of relative water content to within 95%. The vertical distance was ≈0.5 m, the horizontal distance was ≈0.2 m, and the angle of soil viewed by the probe from the collimator was ≈120°. Thermal neutrons detected from distances or angles larger than these values influence the determination of relative water content by 5% or less.

Soil Science Society of America Journal↗
USGS1991

Selenium mobility and distribution in irrigated and nonirrigated alluvial soils

Dissolution and leaching of soil salts by irrigation water is a primary source of Se to shallow groundwater in the western San Joaquin Valley, California. In this study, the mobility and distribution of selenite and selenate in soils with different irrigation and drainage histories was evaluated using sorption experiments and an advection-dispersion model. The sorption studies showed that selenate (15–12400 µg Se L −1 ) is not adsorbed to soil, whereas selenite (10–5000 µg Se L −1 ) is rapidly adsorbed. The time lag between adsorption and desorption of selenite is considerable, indicating a dependence of reaction rate on reaction direction (hysteresis). Selenite adsorption and desorption isotherms were different, and both were described with the Freundlich equation. Model results and chemical analyses of extracts from the soil samples showed that selenite is resistant to leaching and therefore can represent a potential long-term source of Se to groundwater. In contrast, selenate behaves as a conservative constituent under alkaline and oxidized conditions and is easily leached from soil.

Soil Science Society of America Journal↗
USGS2021

Winter flooding to conserve agricultural peat soils in a temperate climate: Effect on greenhouse gas emissions and global warming potential

This study investigated the CO 2 and CH 4 emission rates from agricultural operations on peat soils in the Sacramento-San Joaquin Delta, comparing two common soil management treatments: leaving the field fallow in the winter and flooding the field in winter. Winter flooding is intended to reduce the oxidative loss of soil organic matter to CO 2 , as well as other putative benefits. The goal of the study was to assess if winter flooding lowers overall net carbon loss from the field and if it increases the net CH 4 emissions to a degree that results in a net increase in Global Warming Potential (GWP). The two treatments had similar annual carbon emissions (1.7–1.8 g C/m 2 /d) with the measured annual CO 2 flux rates for both among the highest previously reported, indicating that the flooding treatment did not effectively mitigate subsidence and loss of soil carbon. The flooded treatment had among the highest annual CH 4 emissions previously reported (64.1 mg C/m 2 /d), an order of magnitude greater than that measured on the fallow treatment (5.9 mg C/m 2 /d). Despite the much larger flux of CH 4 from the flooded treatment, when the net carbon export associated with grain harvest and hydrologic transport is included, the differences in the carbon balance and GWP (equivalent to CO 2 emissions of ~775 g C/m 2 /yr) between treatments is insignificant. Total greenhouse gas emissions and GWP of both sites are among the largest previously documented from cultivated peat systems, putting their climatic effect on par with freshwater wetlands, but without the concomitant soil conservation and carbon sequestration benefits of continuous flooding.

Book chapter↗
USGS2018

Vegetation influences on infiltration in Hawaiian soils

Changes in vegetation communities caused by removing trees, introducing grazing ungulates, and replacing native plants with invasive species have substantially altered soil infiltration processes and rates in Hawaii. These changes directly impact run-off, erosion, plant-available water, and aquifer recharge. We hypothesize that broad vegetation communities can be characterized by distributions of field-saturated hydraulic conductivity (Kfs). We used 290 measurements of Kfs calculated from infiltration tests from 5 of the Hawaiian Islands to show this effect. We classified the data using 3 broad ecosystem categories: grasses, trees and shrubs, and bare soil. The soils of each site have coevolved with past and present ecological communities without significant mechanical disturbance by agriculture or urban development. Geometric mean values Kfs are 203 mm/hr for soils hosting trees and shrubs, 50 mm/hr for grasses, and 13 mm/hr for bare soil. Differences are statistically significant at the 95% confidence level. These examples show that it is feasible to make maps of relative Kfs based on field and ecosystem data. These ecosystem trends can be used to estimate ongoing changes to run-off and recharge from climate and land use change. Greater Kfs for ecosystems with primarily trees and shrubs suggests that management decisions concerning reforestation or other changes of vegetation can have substantial hydrologic impacts.

Hawaii↗