Net autotrophy in a fluvial lake: the relative role of phytoplankton and floating-leaved macrophytes

This study combined water- and sediment flux measurements with mass balances of dissolved gas and inorganic matter to determine the importance of pelagic and benthic processes for whole-system metabolism in a eutrophic fluvial lake. Mass balances of dissolved O₂, inorganic carbon (DIC), nitrogen (DIN), phosphorous (SRP), particulate N (PN) and P (PP) and Chl a were calculated at a nearly monthly frequency by means of repeated sampling at the lake inlet and outlet. Simultaneously, benthic fluxes of gas and nutrients, including denitrification rates, and the biomass of the dominant pleustophyte (Trapa natans) were measured, and fluxes of O₂ and CO₂ across the water–atmosphere interface were estimated from diel changes in outlet concentrations. On an annual scale, Middle Lake exhibited CO₂ supersaturation, averaging 313% (range 86–562%), but was autotrophic with a net O₂ production (6.35 ± 2.05 mol m⁻² y⁻¹), DIC consumption (−31.18 ± 18.77 mol m⁻² y⁻¹) and net export of Chl a downstream (8.38 ± 0.95 mol C m⁻² y⁻¹). Phytoplankton was the main driver of Middle Lake metabolism, with a net primary production estimated at 33.24 mol O₂ m⁻² y⁻¹, corresponding to a sequestration of 4.18 and 0.26 mol m⁻² y⁻¹ of N and P, respectively. At peak biomass, T. natans covered about 18% of Middle Lake’s surface and fixed 2.46, 0.17 and 0.02 mol m⁻² of C, N and P, respectively. Surficial sediments were a sink for O₂ (−14.47 ± 0.65 mol O₂ m⁻² y⁻¹) and a source of DIC and NH₄ ⁺ (18.84 ± 2.80 mol DIC m⁻² y⁻¹ and 0.83 ± 0.16 mol NH₄ ⁺ m⁻² y⁻¹), and dissipated nitrate via denitrification (1.44 ± 0.11 mol NO₃ ⁻ m⁻² y⁻¹). Overall, nutrient uptake by primary producers and regeneration from sediments were a minor fraction of external loads. This work suggests that the creation of fluvial lakes can produce net autotrophic systems, with elevated rates of phytoplanktonic primary production, largely sustained by allochtonous nutrient inputs. These hypereutrophic aquatic bodies are net C sinks, although they simultaneously release CO₂ to the atmosphere.     

Nitrogen retention in a floodplain backwater of the upper Mississippi River (USA)

Backwaters connected to large rivers retain nitrate and may play an important role in reducing downstream loading to coastal marine environments. A summer nitrogen (N) inflow-outflow budget was examined for a flow-regulated backwater of the upper Mississippi River in conjunction with laboratory estimates of sediment ammonium and nitrate fluxes, organic N mineralization, nitrification, and denitrification to provide further insight into N retention processes. External N loading was overwhelmingly dominated by nitrate and 54% of the input was retained (137 mg m⁻² day⁻¹). Ammonium and dissolved organic N were exported from the backwater (14 and 9 mg m⁻² day⁻¹, respectively). Nitrate influx to sediment increased as a function of increasing initial nitrate concentration in the overlying water. Rates were greater under anoxic versus oxic conditions. Ammonium effluxes from sediment were 26.7 and 50.6 mg m⁻² day⁻¹ under oxic and anoxic conditions, respectively. Since anoxia inhibited nitrification, the difference between ammonium anoxic–oxic fluxes approximated a nitrification rate of 29.1 mg m⁻² day⁻¹. Organic N mineralization was 64 mg m⁻² day⁻¹. Denitrification, estimated from regression relationships between oxic nitrate influx versus initial nitrate concentration and a summer lakewide mean nitrate concentration of 1.27 mg l⁻¹, was 94 mg m⁻² day⁻¹. Denitrification was equivalent to only 57% of the retained nitrate, suggesting that another portion was assimilated by biota. The high sediment organic N mineralization and ammonium efflux rate coupled with the occurrence of ammonium export from the system suggested a possible link between biotic assimilation of nitrate, mineralization, and export.     

Simulating bubble number density of rhyolitic pumices from Plinian eruptions: constraints from fast decompression experiments

Decompression experiments of a crystal-free rhyolitic liquid with ≈ 6.6 wt. % H₂O were carried out at a pressure range from 250 MPa to 30–75 MPa in order to characterize effects of magma ascent rate and temperature on bubble nucleation kinetics, especially on the bubble number density (BND, the number of bubbles produced per unit volume of liquid). A first series of experiments at 800°C and fast decompression rates (10–90 MPa/s) produced huge BNDs (≈ 2 × 10¹⁴ m⁻³ at 10 MPa/s ; ≈ 2 × 10¹⁵ m⁻³ at 90 MPa/s), comparable to those in natural silicic pumices from Plinian eruptions (10¹⁵–10¹⁶ m⁻³). A second series of experiments at 700°C and 1 MPa/s produced BNDs (≈ 9×10¹² m⁻³) close to those observed at 800°C and 1 MPa/s (≈ 6 × 10¹² m⁻³), showing that temperature has an insignificant effect on BNDs at a given decompression rate. Our study strengthens the theory that the BNDs are good markers of the decompression rate of magmas in volcanic conduits, irrespective of temperature. Huge number densities of small bubbles in natural silicic pumices from Plinian eruptions imply that a major nucleation event occurs just below the fragmentation level, at which the decompression rate of ascending magmas is a maximum (≥ 1 MPa/s).     

Vertical hydraulic exchange and the contribution of hyporheic community respiration to whole ecosystem respiration in the River Lahn (Germany)

To quantify the contribution of hyporheic community respiration to whole running-water ecosystem respiration in a cultural landscape setting, we studied the vertical hydraulic exchange in riffle–pool sequences of the River Lahn (Germany). We used flow through curves from four tracer experiments to estimate flow velocities in the surface and subsurface water. Generally, vertical exchange velocities were higher in riffle sections and a high temporal variability was observed (range of values 0.11–1.08 m day⁻¹). We then used (1) the exchange velocities and (2) time series of dissolved oxygen concentration in surface and subsurface water to calculate hyporheic respiration. Hyporheic respiration was estimated in a range of 10–50 mg O₂ m⁻³ day⁻¹ for the upper sediment layer (first 20 cm). It was much lower in the deeper sediment layer (20–40 cm), ranging from 0 to 10 mg O₂ m⁻³ day⁻¹ (volumes are volumes of interstitial water; the average porosity was 20%). We determined primary production and respiration of the biofilm growing on the sediment by modelling dissolved oxygen concentration time series for a 2,450 m long stream reach (dissolved oxygen concentrations with diurnal variations from 8 to 16 mg L⁻¹). Modelled respiration rates ranged from 2 to 21 g O₂ m² day⁻¹. All information was integrated in a system analysis with numerical simulations of respiration with and without sediments. Results indicated that hyporheic respiration accounted for 6 to 14% of whole ecosystem respiration. These values are much lower than in other whole system respiration studies on more oligotrophic river systems.     

Thermal structure and heat loss at the summit crater of an active lava dome

Forward-Looking Infrared (FLIR) nighttime thermal images were used to extract the thermal and morphological properties for the surface of a blocky-to-rubbley lava mass active within the summit crater of the Caliente vent at Santiaguito lava dome (Guatemala). Thermally the crater was characterized by three concentric regions: a hot outer annulus of loose fine material at 150–400°C, an inner cold annulus of blocky lava at 40–80°C, and a warm central core at 100–200°C comprising younger, hotter lava. Intermittent explosions resulted in thermal renewal of some surfaces, mostly across the outer annulus where loose, fine, fill material was ejected to expose hotter, underlying, material. Surface heat flux densities (radiative + free convection) were dominated by losses from the outer annulus (0.3–1.5 × 10⁴  s⁻¹m⁻²), followed by the hot central core (0.1–0.4 × 10⁴ J s⁻¹m⁻²) and cold annulus (0.04–0.1 × 10⁴ J s⁻¹m⁻²). Overall surface power output was also dominated by the outer annulus region (31–176 MJ s⁻¹), but the cold annulus contributed equal power (2.41–7.07 MJ s⁻¹) as the hot central core (2.68–6.92 MJ s⁻¹) due to its greater area. Cooled surfaces (i.e. the upper thermal boundary layer separating surface temperatures from underlying material at magmatic temperatures) across the central core and cold annulus had estimated thicknesses, based on simple conductive model, of 0.3–2.2 and 1.5–4.3 m. The stability of the thermal structure through time and between explosions indicates that it is linked to a deeper structural control likely comprising a central massive plug, feeding lava flow from the SW rim of the crater, surrounded by an arcuate, marginal fracture zone through which heat and mass can preferentially flow.     

Seasonal and interannual variation of the phytoplankton and copepod dynamics in Liverpool Bay

The seasonal and interannual variability in the phytoplankton community in Liverpool Bay between 2003 and 2009 has been examined using results from high frequency, in situ measurements combined with discrete samples collected at one location in the bay. The spring phytoplankton bloom (up to 29.4 mg chlorophyll m⁻³) is an annual feature at the study site and its timing may vary by up to 50 days between years. The variability in the underwater light climate and turbulent mixing are identified as key factors controlling the timing of phytoplankton blooms. Modelled average annual gross and net production are estimated to be 223 and 56 g C m⁻² year⁻¹, respectively. Light microscope counts showed that the phytoplankton community is dominated by diatoms, with dinoflagellates appearing annually for short periods of time between July and October. The zooplankton community at the study site is dominated by copepods and use of a fine mesh (80 μm) resulted in higher abundances of copepods determined (up to 2.5 × 10⁶ ind. m⁻²) than has previously reported for this location. There is a strong seasonal cycle in copepod biomass and copepods greater than 270 μm contribute less than 10% of the total biomass. Seasonal trends in copepod biomass lag those in the phytoplankton community with a delay of 3 to 4 months between the maximum phytoplankton biomass and the maximum copepod biomass. Grazing by copepods exceeds net primary production at the site and indicates that an additional advective supply of carbon is required to support the copepod community.     

Observation and modeling of NPP for Pinus elliottii plantation in subtropical China

Based on the stem analysis of 59 individuals of Pinus elliottii in combination with tree biomass models, we calculated annual biomass increment of forest plots at Qianyanzhou Ecological Station, Chinese Academy of Sciences in subtropical China. In addition, canopy layer and community NPP were calculated based on 12 years’ litter fall data. NPP of the 21-year-old forest was estimated by using the BIOME BGC model; and both measured NPP and estimated NPP were compared with flux data. Community biomass was 10574 g · m⁻²; its distribution patterns in tree layer, shrub layer, herbaceous layer, tree root, herbaceous and shrub roots and fine roots were 7542, 480, 239, 1810, 230, 274 and 239 g · m⁻², respectively. From 1999 to 2004, the average annual growth rate and litter fall were 741 g · m⁻² · a⁻¹ (381.31 gC · m⁻² · a⁻¹) and 849 g · m⁻² · a⁻¹ (463 gC · m⁻² · a⁻¹), respectively. There was a significant correlation between annual litter fall and annual biomass increment; and the litter fall was 1.19 times the biomass increment of living trees. From 1985 to 2005, average NPP and GPP values based on BGC modeling were 630.88 (343.31–906.42 gC · m⁻² · a⁻¹) and 1 800 gC · m⁻² · a⁻¹ (1351.62–2318.26 gC · m⁻² · a⁻¹). Regression analysis showed a linear relationship (R ²=0.48) between the measured and simulated tree layer NPP values. NPP accounted for 30.2% (25.6%–32.9%) of GPP, while NEP accounted for 57.5% (48.1%–66.5%) of tree-layer NPP and 41.74% (37%–52%) of stand NPP. Soil respiration accounted for 77.0% of measured tree NPP and 55.9% of the measured stand NPP. NEE based on eddy covariance method was 12.97% higher than the observed NEP. Supported by the National Key Basic Research Special Foundation of China (Grant No. 2002CB4125), International Joint Research Project under Ministry of Science and Technology of China (Grant No. 2006DFB91920)     

CO<Subscript>2</Subscript> and He degassing at El Chichón volcano,Chiapas, Mexico: gas flux,origin and relationship with local and regional tectonics

During 2007–2008, three CO₂ flux surveys were performed on El Chichón volcanic lake, Chiapas, Mexico, with an additional survey in April 2008 covering the entire crater floor (including the lake). The mean CO₂ flux calculated by sequential Gaussian simulation from the lake was 1,190 (March 2007), 730 (December 2007) and 1,134 g m⁻² day⁻¹ (April 2008) with total emission rates of 164 ± 9.5 (March 2007), 59 ± 2.5 (December 2007) and 109 ± 6.6 t day⁻¹ (April 2008). The mean CO₂ flux estimated from the entire crater floor area was 1,102 g m⁻² day⁻¹ for April 2008 with a total emission rate of 144 ± 5.9 t day⁻¹. Significant change in CO₂ flux was not detected during the period of survey, and the mapping of the CO₂ flux highlighted lineaments reflecting the main local and regional tectonic patterns. The ³He/⁴He ratio (as high as 8.1 R _A) for gases in the El Chichón crater is generally higher than those observed at the neighbouring Transmexican Volcanic Belt and the Central American Volcanic Arc. The CO₂/³He ratios for the high ³He/⁴He gases tend to have the MORB-like values (1.41 × 10⁹), and the CO₂/³He ratios for the lower ³He/⁴He gases fall within the range for the arc-type gases. The high ³He/⁴He ratios, the MORB-like CO₂/³He ratios for the high ³He/⁴He gases and high proportion of MORB-CO₂ (M = 25 ±15%) at El Chichón indicate a greater depth for the generation of magma when compared to typical arc volcanoes.     

The dynamics of a double-cell hydrothermal system in triggering seismicity at Somma-Vesuvius: results from a high-resolution radon survey (revisited)

Data collected at Somma-Vesuvius during the 1998–1999 radon surveys have been revisited and reinterpreted in light of recent geophysical and geochemical information. The duration of selected radon anomalies, together with the decay properties of radon, have been used to estimate the permeability and porosity of rocks of the deep hydrothermal system. The current local cyclic seismicity is explained by means of a double convective-cell model. Convective cells are separated by a low-permeability horizon located at about 2–2.5 km below sea level. Fluids convecting within the upper cells show temperatures ranging 300–350°C. Rock permeabilities in this sector are estimated on the order of 10⁻¹² m², for porosities (ϕ) of about 10⁻⁵ typical of a brittle environment where fluid velocities may reach ∼800 m/day. Fluid temperatures within the lower cells may be as high as 400–450°C, consistent with supercritical regimes. The hydrodynamic parameters for these cells are lower, with permeability k ∼ 10⁻¹⁵ m², and porosity ranging from 10⁻⁶ to 10⁻⁷. Here, fluid motion toward the surface is controlled by the fracture network within a porous medium approaching brittle–ductile behaviour, and fluid velocities may reach ∼1,800 m/day. The low-permeability horizon is a layer where upper and lower convecting cells converge. In this region, fluids (convecting both at upper and lower levels) percolate through the wallrock and release their brines. Due to self-sealing processes, permeability within this horizon reaches critical values to keep the fluid pressure near lithostatic pressure (for k ∼ 10⁻¹⁸ m²). Deep fluid pressure buildups precede the onset of hydrothermally induced earthquakes. Permeability distribution and rock strength do not exclude that the next eruption at Somma-Vesuvius could be preceded by a seismic crisis, eventually leading to a precursory phreatic explosion. The coupling of these mechanisms has the potential of inducing pervasive failure within rocks of the hydrothermal shell, and may be a prelude to a magmatic eruption. It is finally emphasised that the integrated analysis of seismic and geochemical data, including radon emissions, could be successfully used in testing temperature distributions and variations of porosity and permeability in active geothermal reservoirs.     

Steric Sea-Level Change and its Impact on the Gravity Field caused by Global Climate Change

It is sometimes assumed that steric sea-level variations do not produce a gravity signal as no net mass change, thus no change of ocean bottom pressure is associated with it. Analyzing the output of two CO₂ emission scenarios over a period of 2000 years in terms of steric sea-level changes, we try to quantify the gravitational effect of steric sea-level variations. The first scenario, computed with version 2.6 of the Earth System Climate Model developed at the University of Victoria, Canada (UVic ESCM), is implemented with a linear CO₂ increase of 1% of the initial concentration of 365 ppm and shows a globally averaged steric effect of 5.2 m after 2000 years. In the second scenario, computed with UVic ESCM version 2.7, the CO₂ concentration increases quasi-exponentially to a level of 3011 ppm and is hold fixed afterwards. The corresponding globally averaged steric effect in the first 2000 years is 2.3 m. We show, due to the (vertical) redistribution of ocean water masses (expansion or contraction), the steric effect results also in a small change in the Earth’s gravity field compared to usually larger changes associated with net mass changes. Maximum effects for computation points located on the initial ocean surface can be found in scenario 1, with the effect on gravitational attraction and potential ranging from 0.0 to −0.7·10⁻⁵ m s⁻² and −3·10⁻³ to 6·10⁻³ m² s⁻², respectively. As expected, the effect is not zero but negligible for practical applications.     

Soil organic carbon storage and soil CO<Subscript>2</Subscript> flux in the alpine meadow ecosystem

High-resolution sampling, measurements of organic carbon contents and ¹⁴C signatures of selected four soil profiles in the Haibei Station situated on the northeast Tibetan Plateau, and application of ¹⁴C tracing technology were conducted in an attempt to investigate the turnover times of soil organic carbon and the soil-CO₂ flux in the alpine meadow ecosystem. The results show that the organic carbon stored in the soils varies from 22.12×10⁴ kg C hm⁻² to 30.75×10⁴ kg C hm⁻² in the alpine meadow ecosystems, with an average of 26.86×10⁴ kg C hm⁻². Turnover times of organic carbon pools increase with depth from 45 a to 73 a in the surface soil horizon to hundreds of years or millennia or even longer at the deep soil horizons in the alpine meadow ecosystems. The soil-CO₂ flux ranges from 103.24 g C m⁻² a⁻¹ to 254.93 gC m⁻² a⁻¹, with an average of 191.23 g C m⁻² a⁻¹. The CO₂ efflux produced from microbial decomposition of organic matter varies from 73.3 g C m⁻² a⁻¹ to 181 g C m⁻² a⁻¹. More than 30% of total soil organic carbon resides in the active carbon pool and 72.8%281.23% of total CO₂ emitted from organic matter decomposition results from the topsoil horizon (from 0 cm to 10 cm) for the Kobresia meadow. Responding to global warming, the storage, volume of flow and fate of the soil organic carbon in the alpine meadow ecosystem of the Tibetan Plateau will be changed, which needs further research. Supported by the National Natural Science Foundation of China (Grant Nos. 40231015, 40471120 and 40473002) and the Guangdong Provincial Natural Science Foundation of China (Grant No. 06300102)     

Mineralogical analyses and in vitro screening tests for the rapid evaluation of the health hazard of volcanic ash at Rabaul volcano, Papua New Guinea

The continuous ash and gas emissions from the Tavurvur cone in Rabaul caldera, Papua New Guinea, during 2007–08, raised concerns regarding how exposure would affect the respiratory health of nearby populations and impact on the environment. As part of a formal evaluation of the effects of volcanic emissions on public health, we investigated the potential health hazard of the ash using a suite of selected mineralogical analyses and in vitro toxicity screening tests. The trachy-andesitic ash comprised 2.1–6.7 vol.% respirable (sub-4 μm diameter) particles. The crystalline silica content was 1.9–5.0 wt.% cristobalite (in the bulk sample) with trace amounts of quartz and/or tridymite. Scanning electron microscopy showed that the ash particles were angular with sparse, fibre-like particles (∼3–60 μm max. diameter) observed in some samples, which we confirmed to be CaSO₄ (gypsum, at <6 wt.% in the bulk samples) and not asbestiform fibres. The ash specific surface area was low (0.1–2.7 m² g⁻¹). The leached solution from one of the ash samples was slightly acidic (pH 5.6), but did not contain high levels of toxic metals (such as F, Cu, Zn, Mn, As, Ni and Cd) when compared to previously tested volcanic ash leachates. Ash samples generated potentially-harmful hydroxyl radicals through an iron-mediated catalytic reaction, in the range of 0.15–2.47 μmol m⁻² (after 30 min of reaction). However, measurement of particle oxidative capacity (potential oxidative stress reaction using ascorbic acid) and silica-like injury to red blood cells (erythrolysis assay, i.e. measurement of cell death) nevertheless revealed low biological reactivity. The findings suggest that acute exposure to the ash would have a limited potential to exacerbate pre-existing conditions such as asthma or chronic bronchitis, and the potential for chronic exposure leading to silicosis was low.     

Belowground carbon balance and carbon accumulation rate in the successional series of monsoon evergreen broad-leaved forest

The belowground part of terrestrial ecosystem is a huge carbon pool. It is believed that of the total 2500Gt carbon stored in global terrestrial ecosystem, soil carbon storage within the 1 m surface layer ac- counts for 2000Gt, which is 4-fold of vegetation car- bon storage[1,2]. Compared with the carbon in the vegetation, carbon in the deep soil layers is much more stable, and it will stay in soil profile permanentlyunless geological vicissitude occurs. Essentially, forest restoration is the…     

Uncertainty analysis of CO<Subscript>2</Subscript> flux components in subtropical evergreen coniferous plantation

We present an uncertainty analysis of ecological process parameters and CO₂ flux components (R _eco, NEE and gross ecosystem exchange (GEE)) derived from 3 years’ continuous eddy covariance measurements of CO₂ fluxes at subtropical evergreen coniferous plantation, Qianyanzhou of ChinaFlux. Daily-differencing approach was used to analyze the random error of CO₂ fluxes measurements and bootstrapping method was used to quantify the uncertainties of three CO₂ flux components. In addition, we evaluated different models and optimization methods in influencing estimation of key parameters and CO₂ flux components. The results show that: (1) Random flux error more closely follows a double-exponential (Laplace), rather than a normal (Gaussian) distribution. (2) Different optimization methods result in different estimates of model parameters. Uncertainties of parameters estimated by the maximum likelihood estimation (MLE) are lower than those derived from ordinary least square method (OLS). (3) The differences between simulated R_eco, NEE and GEE derived from MLE and those derived from OLS are 12.18% (176 g C·m⁻²·a⁻¹), 34.33% (79 g C·m⁻²·a⁻¹) and 5.4% (92 g C·m⁻²·a⁻¹). However, for a given parameter optimization method, a temperature-dependent model (T_model) and the models derived from a temperature and water-dependent model (TW_model) are 1.31% (17.8 g C·m⁻²·a⁻¹), 2.1% (5.7 g C·m⁻²·a⁻¹), and 0.26% (4.3 g C·m⁻²·a⁻¹), respectively, which suggested that the optimization methods are more important than the ecological models in influencing uncertainty in estimated carbon fluxes. (4) The relative uncertainty of CO₂ flux derived from OLS is higher than that from MLE, and the uncertainty is related to timescale, that is, the larger the timescale, the smaller the uncertainty. The relative uncertainties of R_eco, NEE and GEE are 4%−8%, 7%−22% and 2%−4% respectively at annual timescale. Supported by the National Natural Science Foundation of China (Grant No. 30570347), Innovative Research International Partnership Project of the Chinese Academy of Sciences (Grant No. CXTD-Z2005-1) and National Basic Research Program of China (Grant No. 2002CB412502)     

Phytoplankton abundance and contributions to suspended particulate matter in the Ohio,Upper Mississippi and Missouri Rivers

Main channel habitats of the Ohio, Missouri, and Upper Mississippi Rivers were surveyed during the summers of 2004, 2005 and 2006 using a probability-based sampling design to characterize inter-annual and inter-river variation in suspended chlorophyll a (CHLa) and related variables. Large (fivefold) differences in CHLa were observed with highest concentrations in the Upper Mississippi (32.3 ± 1.8 μg L⁻¹), intermediate values in the Missouri (19.7 ± 1.1 μg L⁻¹) and lowest concentrations in the Ohio (6.8 ± 0.5 μg L⁻¹). Inter-annual variation was small in comparison to inter-river differences suggesting that basin-specific factors exert greater control over river-wide CHLa than regional-scale processes influencing climate and discharge. The rivers were characterized by variable but generally low light conditions as indicated by depth-averaged underwater irradiance <4 E m⁻² day⁻¹ and high ratios of channel depth to euphotic depth (>3). Despite poor light conditions, regression analyses revealed that TP was the best single predictor of CHLa (R ² = 0.40), though models incorporating both light and TP performed better (R ² = 0.60). Light and nutrient conditions varied widely within rivers and were inversely related, suggesting that riverine phytoplankton may experience shifts in resource limitation during transport. Inferred grazing and sedimentation losses were large yet CHLa concentrations did not decline downriver indicating that growth and loss processes were closely coupled. The contribution by algae to suspended particulate organic matter in these rivers (mean = 41%) was similar to that of lakes (39%) but lower relative to reservoirs (61%).