Lake and Reservoir Management Development of a ...

This article was downloaded by: [Kangwon National University] On: 18 March 2015, At: 08:31 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Lake and Reservoir Management Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/ulrm20

Development of a phosphorus budget for Lake Mead a

b

c

d

Li Ding , Imad A. Hannoun , E. John List & Todd Tietjen a

Flow Science Incorporated, 310 Neff Ave, Harrisonburg, VA, 22801

b

California Lake Management Society, 1726 Three Springs Rd, McGaheysville, VA, 22840

c

Flow Science Incorporated, 48 S Chester Ave, Ste 200, Pasadena, CA, 91106

d

Southern Nevada Water Authority, River Mountains Water Treatment Facility, PO Box 99954, Las Vegas, NV, 89193 Published online: 22 Apr 2014.

Click for updates To cite this article: Li Ding, Imad A. Hannoun, E. John List & Todd Tietjen (2014) Development of a phosphorus budget for Lake Mead, Lake and Reservoir Management, 30:2, 143-156, DOI: 10.1080/10402381.2014.899656 To link to this article: http://dx.doi.org/10.1080/10402381.2014.899656

PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions

Lake and Reservoir Management, 30:143–156, 2014  C Copyright by the North American Lake Management Society 2014 ISSN: 1040-2381 print / 2151-5530 online DOI: 10.1080/10402381.2014.899656

Development of a phosphorus budget for Lake Mead Li Ding,1,∗ Imad A. Hannoun,2 E. John List,3 and Todd Tietjen4 1

Flow Science Incorporated, 310 Neff Ave, Harrisonburg, VA 22801 California Lake Management Society, 1726 Three Springs Rd, McGaheysville, VA 22840 3 Flow Science Incorporated, 48 S Chester Ave, Ste 200, Pasadena, CA 91106 4 Southern Nevada Water Authority, River Mountains Water Treatment Facility, PO Box 99954, Las Vegas, NV 89193 2

Downloaded by [Kangwon National University] at 08:31 18 March 2015

Abstract Ding L, Hannoun IA, List EJ, Tietjen T. 2014. Development of a phosphorus budget for Lake Mead. Lake Reserv Manage. 30:143–156. Phosphorus is the growth-limiting nutrient for phytoplankton in Lake Mead. Multiple agencies have developed an extensive phosphorus dataset with low detection limit since 2007 by obtaining monthly or biweekly samples at 19 monitoring stations throughout the lake and at all the major inflows. Approximately 2000 phosphorus grab samples were collected and analyzed as part of this effort. We developed a phosphorus budget for Lake Mead during 2007–2008 using these measured phosphorus data and evaluated its accuracy using the water and bromide budget developed in this study. Based on the phosphorus budget, total phosphorus (TP) loading from the Colorado River dominated external TP loading to Lake Mead, accounting for 98% of total external loading. The orthophosphate (Ortho-P) portion that is soluble and bioavailable accounted for 1000 μg/L (Rosen et al. 2012) and could be detected downstream as far away as a reservoir in San Diego County, California (CRRSCO 2008). Although multiple factors contributed to the occurrence of the algal bloom (LaBounty and Burns 2005), the main driving factor was attributed to increased phosphorus concentrations at the water surface during early spring 2001 (Rosen et al. 2012). This abundance of phosphorus, combined with warming water, promoted a rapid growth of the algal species in 2001 (LaBounty and Burns 2005, Rosen et al. 2012). Since 2001, the surrounding WWTPs have enhanced treatment for phosphorus removal, which resulted in a steadily decreasing phosphorus discharge from wastewater effluent (Holdren and Turner 2011). Although the algal abundance in the area close to the WWTP discharges is decreasing, the importance of this phosphorus reduction in relation to other phosphorus sources in the lake is still unknown because of a lack of a phosphorus budget. Determining the amount of phosphorus loading from LVW to the lake, and comparing this phosphorus loading to the loadings from other sources, became one of the key drivers of this study. 144

The unprecedented drop in water levels in Lake Mead since 1999 (Holdren and Turner 2011) provided additional motivation for constructing a phosphorus budget. The lowering lake water levels reduces the volume of water available to dilute incoming phosphorus loads and may lead to rising phosphorus levels and algal production in the lake. An accurate phosphorus budget can provide reasonable estimates of rising phosphorus levels and facilitate quick and appropriate responses from the lake management agencies. The recent discovery of adult quagga mussels (Dreissena rostriformis bugensis) in Lake Mead has significant impacts on water quality, lake biology, and infrastructure (Holdren and Turner 2010, Wong et al. 2010). The study of water quality impacts of quagga mussels can benefit from the development of a phosphorus budget. For example, the phosphorus budget can be used to determine whether the lowering chlorophyll a concentrations in the lake in recent years can be attributed to the decline of phosphorus loadings or invasion of quagga mussels (Wong et al. 2010). The objectives of this study were to (1) develop a phosphorus budget in Lake Mead and provide essential information for sound management of this lake, and (2) provide insight into the fate of major phosphorus loadings into the lake. Two distinct features are associated with this study. First, we used a large number of available high-quality phosphorus and flow data to study a phosphorus budget in one of the largest reservoirs in the world. Second, we constructed a bromide budget in the lake to estimate errors in the calculated phosphorus budget. Bromide served as a conservative tracer, while the residual of calculated bromide mass balance provides an indication of errors in the water mass balance and, to a certain extent, the phosphorus budget.

Study area Lake Mead, formed in 1935 by the construction of Hoover Dam on the Colorado River, is the largest reservoir in the United States and one of the largest reservoirs in the world. The dam is located ∼40 km southeast of the City of Las Vegas, Nevada. The lake extends upstream from the dam along approximately 110 km of the Colorado River valley and straddles the states of Nevada and Arizona. At full capacity, the lake holds ∼37 km3 of water, has a surface area of 660 km2, and a maximum depth of 180 m (LaBounty and Burns 2007). Lake Mead serves as an important water supply for almost 2 million residents of Southern Nevada and provides recreational opportunities to more than 8 million yearly visitors to the Lake Mead National Recreation Area (Tietjen and Holdren 2010). The lake also provides flood control, electricity through hydropower generation, and water storage for millions of downstream users in Arizona, California, Nevada, and Mexico.

Downloaded by [Kangwon National University] at 08:31 18 March 2015

Development of a phosphorus budget for Lake Mead

Figure 1. Lake Mead, Nevada, including all main inflows, outflows, and basins. Solid circles represent monitoring stations (CR = Colorado River, LVB = Las Vegas Bay, VR = Virgin River).

The lake comprises 3 main basins (Fig. 1): (1) the first immediately upstream to the Hoover Dam is Boulder Basin; (2) the next basin upstream of Boulder Basin consists of the Virgin Basin, Temple Basin, and the Overton Arm (referred as Virgin+Temple Basin hereafter); and (3) the most upstream basin is the Gregg Basin. These 3 basins are connected by 2 narrow channels: The Narrows connects Boulder Basin and Virgin+Temple Basin, and Virgin Canyon connects Virgin+Temple Basin and Gregg Basin. There are 4 main inflows into Lake Mead (Fig. 1): the Colorado River enters the Gregg Basin; the Muddy and Virgin rivers discharge into the Overton Arm, which is connected to the northern part of the Virgin+Temple Basin; and LVW flows into the northwest part of the Boulder Basin. WWTPs from surrounding areas discharge into the LVW. The 2 principal outflows include releases through Hoover Dam to

supply the lower Colorado River and withdrawals by the Southern Nevada Water Authority (SNWA). SNWA withdraws through intakes located at the east side of Saddle Island in Boulder Basin to supply municipal, irrigation, and industrial demand in the Las Vegas metropolitan area. Basic Management Industrial (BMI) also withdraws a small amount of water through its intakes near Saddle Island. Because the BMI intake is located close to the SNWA intakes, the withdrawals were designated as “SNWA/BMI outflow.” The average water residence times for the Gregg, Virgin+Temple, Boulder Basin, and the lake as a whole (estimated as lake volume divided by Colorado River flow rate) are ∼2, 12, 8, and 22 months, respectively. Lake Mead is a deep, subtropical, monomictic lake (LaBounty and Burns 2005) that is thermally stratified from April to December and is generally well-mixed from January 145

Ding et al.

Downloaded by [Kangwon National University] at 08:31 18 March 2015

to March. During the stable summer stratification period, the epilimnion, metalimnion, and hypolimnion are about 15, 20, and up to 120 m thick, respectively, depending on the depth of the reservoir and location (LaBounty and Burns 2007). Lake Mead does not completely destratify every year. For example, LaBounty and Burns (2007) reports that Boulder Basin completely destratified every other year on average between 1991–2007. During the same period, the dissolved oxygen in the near-bottom water layer in Boulder Basin remained >2 mg/L at all times, despite this incomplete destratification (LaBounty and Burns 2007). Algal growth in Lake Mead is strongly phosphorus limited (Paulson and Baker 1983, Piechota et al. 2002, LaBounty and Burns 2005). Algae biomass typically peaks in spring and summer and declines in the winter (Fig. 2; FSI 2010). Thus, the algal growing season (GS) is between 1 April and 30 September, while the rest of the year is the non-growing season (Non-GS) according to the Lake Mead water quality standard (A. Preston et al., 2013, Flow Science Incorporated, unpubl.). The dominant algal species in Lake Mead is Pyramichlamys spp., a Chlorophyta or green algae (Rosen et al. 2012). Chlorophyll a concentrations remain relatively low in the open water but are higher near the lake inflows (Holdren and Turner 2010). The Virgin+Temple Basin, the lower half of the Overturn Arm, and the upper end of Boulder Basin are oligotrophic to mesotrophic, while the areas near inflows are eutrophic to hypereutrophic (Holdren and Turner 2010). The chlorophyll a concentrations are highest in the area near LVW and in Las Vegas Bay (LVB; Fig. 2 and 3; FSI 2010) due to a large amount of bioavailable phosphorus entering from LVW (LaBounty and Burns 2005).

Conceptually, the mass budget for any substance (i.e., water, phosphorus, and bromide in this study) can be constructed by considering the area of interest (i.e., basin of the lake in this study) as a single “black box” using the following equation (Chapra 1997): (1)

where Mx /t is the rate of observed change of the mass of material x in the water body over time period t; Ix and Ox are the inflow and outflow mass flow rates of the material; Sx is the sum of rate of change of internal sources (e.g., phosphorus release from sediment) and sinks (e.g., phosphorus settling to the bottom of the lake); and Ex is the residual of the mass balance. For phosphorus, Sx is unknown and can be inferred from equation 1 by calculating all other terms in the 146

Ex is not zero, however, because of inaccuracy associated with field data and approximations (e.g., interpolation) used to develop the budget calculation. Thus, the calculated source and sink term in the phosphorus budget includes this non-zero residual term. The significance of this residual term relative to the total loading (or the true source and sink term that was expected to be of the same order of magnitude as the total loading in Lake Mead) is estimated by evaluating both the water and bromide budget in the lake that were constructed in the same fashion as the phosphorus budget. Because the internal sources/sinks terms were near zero for water and bromide, a factor validated by small relative errors (RE) in the calculated water and bromide budget, the residual terms (i.e., Ex ) for water and bromide can be calculated directly by computing the values of Ix , Ox , and Mx /t in equation 1 using field data. The RE were calculated for water and bromide budgets as Ex expressed as a percentage of Ix (i.e., as a percentage of the sum of all inflow mass). The small RE of the water budget indicated an accurate water budget, a prerequisite for an accurate phosphorus balance. The RE of the bromide mass budget provided an additional check of the water budget and validation of approximations used to develop the phosphorus budget. Thus, a small RE in the bromide budget indicates that the major inflows and outflows are included in the calculation, and errors associated with approximations are relatively small. The budgets developed in this study were presented in annual averaged daily values to be consistent with common lake water quality regulations such as total maximum daily loading.

Water budget

Methods

Mx + Ox = Ix + Sx + Ex , t

equation using measured inflow, outflow, and in-lake data if Ex is zero. A positive value for Sx indicates an internal source while a negative value indicates an internal sink.

The water budget for Lake Mead includes inflows, outflows, precipitation, evaporation, groundwater seepage, and changes in water storage. During any given period, the sum of outflows, evaporation, and groundwater seepage, and changes in water storage should balance the sum of inflows and precipitation. Change in water storage is defined as the difference in lake water volume between the end and the beginning of the period of concern. The groundwater seepage and precipitation (precipitation was estimated as 100 m water depth) were used at the Narrows. Measured phosphorus concentrations were averaged in these layers and then combined with flows to calculate phosphorus loadings for each layer. The sum of the phosphorus loads for these layers produced the overall phosphorus loadings through each channel. A phosphorus retention coefficient (R), defined as the ratio of the net amount of phosphorus retained in the lake sediment to the sum of external loadings, was calculated for each of these basins and for the whole lake. Following equation 1, R can be written as:   Mx + Ox − Ix − Ex /Ix R = −Sx /Ix = − t   Mx =− (2) + Ox − Ix /Ix + RE. t The RE of the phosphorus budget represents the error in the calculation of R. In this study, the RE of the bromide budget was used to provide an indication of the RE of the phosphorus budget associated with an unbalanced water budget and approximation approach. A positive retention coefficient indicates that a fraction (i.e., R) of the phosphorus external loading is deposited into the lake sediments; a negative retention coefficient reflects the sediment release and downstream transport of previously retained phosphorus. The significance of calculating the retention coefficient in the lake is to determine whether the lake is a nutrient sink

(positive R) or nutrient source (negative R) as well as the importance of this source or sink in related to the total external loadings. To closely examine the settling patterns of phosphorus particles from the Colorado River, ratios of TP concentration to Ortho-P concentration were calculated during the winter months (in this case, defined as Nov–Feb) at the stations in Gregg Basin including CRLM, CR394.0, CR390.0, and CR380.0. Considering there was minimal loss of Ortho-P due to limited algal consumption during the winter months, the ratio of TP to Ortho-P should remain constant in Gregg Basin if settling of phosphorus particles did not occur. The reduction in this ratio indicated the loss of TP due to particle settling. Note that the ratios of TP to Ortho-P presented here were averaged vertically at these stations to account for all the TP and Ortho-P over the whole water column. Not included in the current analysis was the effect of infrequent storm events on the overall phosphorus budget because the existing phosphorus monitoring frequency is insufficient to capture these phenomena.

Bromide budget All terms except the residual in the mass budget equation (i.e., equation 1) for bromide were calculated based on field data. The bromide data as well as the sampling locations, depths, collection times, analytical methods, and protocols were obtained from the SNWA website (http://www.snwa.com/apps/wq database/index.cfml). Bromide budgets in 2007 and 2008 were computed, and the RE were obtained for each basin as well as for the whole lake. The calculation of the bromide budget used the same approach and same units used for the phosphorus budgets (with the exception that the internal source term Sx is set to zero). In addition, the bromide dataset came from the samples obtained on the same day at the same monitoring stations as the phosphorus dataset. Thus, the RE of the bromide mass budget provided an additional check of the water budget and validation of approximations used to develop the phosphorus budget.

Results Water budget Colorado River inflow and Hoover Dam outflow dominated the water budgets for 2007 and 2008 (Table 1). The average flow rate for the Colorado River was approximately 378 m3/s and contributed about 97% of all inflow to the lake. The next largest inflow was the LVW with a 2007–2008 average flow rate of about 8.2 m3/s. The 2007–2008 average flow rate 149

Ding et al.

Downloaded by [Kangwon National University] at 08:31 18 March 2015

Table 1. Annual water budget in Lake Mead and its individual basins for 2007 and 2008 in Lake Mead, NV. Symbols are inflows (I) and outflows (O), evaporation loss (S), rate of change of mass (M/t), and residual (E) (see equation 1). There are no data for the flows through Virgin Canyon and the Narrows, and these flows were derived from equation 1. The relative error (RE) is the residual expressed as a percentage of the total inflow rate; all other terms are flow rates in m3/s.

Water Budget Terms (m3/d)

Type Symbol

2007

2008

Colorado River Outflow through Virgin Canyon Evaporation Loss Rate Water Volume Change Rate

Gregg Basin I O S M/t

360 362 −2.7 −4.8

397 396 −2.4 −1.3

362 0.1 2.7 379 −18.0 −32.3

396 0.1 3.4 392 −16.4 −8.5

379 8.3 365 19.7 −8.3 −14.5

392 8.1 374 18.3 −7.6 −3.9

Inflow through Virgin Canyon Muddy River Virgin River Outflow through the Narrows Evaporation Loss Rate Water Volume Change Rate

Virgin + Temple Basins I I I O S M/t

Inflow through the Narrows LVW Hoover Dam Outflow SNWA/BMI Outflow Evaporation Loss Rate Water Volume Change Rate

Boulder Basin I I O O S M/t

Colorado River Muddy River Virgin River LVW Hoover Dam Outflow SNWA/BMI Outflow Evaporation Loss Rate Water Volume Change Rate) Residual Relative Error

The Whole Lake I I I I O O S M/t E RE

through Hoover Dam was about 370 m3/s and consisted of 95% of the outflow. The remaining 5% of the outflow was the withdrawal through intakes operated by SNWA/BMI with a 2007–2008 average flow rate of 19.0 m3/s. The residuals were −2.3% (−8.7 m3/s) and −0.9% (−3.8 m3/s) of the measured total inflows in 2007 and 2008, respectively (Table 1). These small residuals suggested that measured flow rates were reasonably accurate and can be used to develop reliable bromide and phosphorus loading estimates. Note that the measured Colorado River inflow rates reported by USGS were rounded to the nearest 100 cfs (or 2.8 m3/s), and the residual errors for the water budget were of the same order of magnitude as the rounding error. 150

360 0.1 2.7 8.3 365 19.7 −29.2 −51.5 −8.7 −2.3%

397 0.1 3.4 8.1 374 18.3 −26.2 −13.7 −3.8 −0.9%

Phosphorus budget The phosphorus budget for Gregg Basin showed that the Colorado River TP load into the Gregg Basin was quite large, and that the TP flux leaving the basin was an order of magnitude smaller (Table 2). In addition, the Ortho-P loads from the Colorado River made up