Subscriber access provided by University of Newcastle, Australia
Article
Contribution of leaf litter to nutrient export during winter months in an urban residential watershed Anika Rose Bratt, Jacques C Finlay, Sarah E. Hobbie, Benjamin D. Janke, Adam C. Worm, and Kathrine L. Kemmitt Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.6b06299 • Publication Date (Web): 19 Feb 2017 Downloaded from http://pubs.acs.org on February 22, 2017
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 35
Environmental Science & Technology
1
Contribution of leaf litter to nutrient export during winter months in an urban residential
2
watershed
3
Authors: Anika R. Bratt*, Jacques C. Finlay, Sarah E. Hobbie, Benjamin D. Janke, Adam C.
4
Worm, Kathrine L. Kemmitt - Department of Ecology, Evolution and Behavior, University of
5
Minnesota, St. Paul, MN 55108
6
*Corresponding author
7 8 9
Abstract Identification of non-point sources of nitrogen (N) and phosphorus (P) in urban systems is
10
imperative to improve water quality and better manage eutrophication. Winter contributions and
11
sources of annual N and P loads from urban watersheds are poorly characterized in northern
12
cities because monitoring is often limited to warm weather periods. To determine winter export
13
of N and P, we monitored storm water outflow in a residential watershed in Saint Paul,
14
Minnesota during 2012-2014. Our data demonstrate that winter melt events contribute a high
15
percentage of annual N and P export (50%). We hypothesized that over-wintering leaf litter that
16
is not removed by fall street sweeping could be an important source to winter loads of N and P.
ACS Paragon Plus Environment
Environmental Science & Technology
17
We estimated contributions of this source by studying decomposition in lawns, street gutters, and
18
catch basins during two winters. Rates of mass and N loss were negligible during both winters.
19
However, P was quickly solubilized from decomposing leaves. Using mass balances and
20
estimates of P leaching losses, we estimated that leaf litter could contribute 80% of winter TDP
21
loading in this watershed (~40% of annual TDP loading). Our work indicates that urban trees
22
adjacent to streets likely represent a major source of P pollution in northern cities. Management
23
that targets important winter sources such as tree leaves could be highly effective for reducing P
24
loading and may mitigate eutrophication in urban lakes and streams in developed cities.
ACS Paragon Plus Environment
Page 2 of 35
Page 3 of 35
25 26
Environmental Science & Technology
Introduction Nitrogen (N) and phosphorus (P) are the most ubiquitous water pollutants1 and together
27
threaten the health of ecosystems at both local and global scales via downstream transport. Land
28
use changes associated with urban development play a dominant role in the high N and P inputs
29
to aquatic ecosystems that lead to eutrophication, especially at local scales2. The increase in
30
impervious (paved) surface cover associated with urbanization greatly influences hydrologic
31
pathways, creating a physical barrier to soils, decreasing infiltration and increasing runoff3. This
32
increase in runoff in combination with high nutrient concentrations in stormwater drives high
33
nutrient loading, especially of N and P, in urban waterways, which can be difficult to manage.
34
Point source pollution from wastewater effluent has been reduced in developed countries
35
due to improved technology and regulations; however, non-point pollution from storm sewers
36
can be an important source of nutrients as well. For example, in Baltimore, Groffman and others4
37
estimated that non-point sources contributed 25.6 kg N ha-1 yr-1, compared to 35 kg N h-1 yr-1
38
from point sources. Controlling non-point sources of N and P is difficult because they are diffuse,
39
hard to identify, and vary with space, time and climate5. This is especially true in urban centers
40
where non-point sources can range from construction sites to golf courses to private lawns and
41
include soils, fertilizer, atmospheric deposition, leaking sewers, pet waste, and leaf litter.
42
Traditional stormwater best management practices (BMPs, e.g. detention ponds,
43
vegetated swales, rain gardens) focus on volume and particulate reduction – typically
44
downstream rather than at the source – and can be expensive and are often ineffective (e.g. Song
45
et al.6). Some studies have shown that nearly half of urban P inputs are soluble7, and an
46
awareness of this fact has led to recent development of infiltration-based practices that are
47
designed to remove dissolved P (e.g. iron-enhanced sand filters; Erickson et al. 8, Arias et al.9).
ACS Paragon Plus Environment
Environmental Science & Technology
48
Despite these innovations, there is increasing interest in managing N and P at the source7, but we
49
lack a clear understanding of the relative importance of non-point sources of nutrients in urban
50
landscapes. Without this basic understanding, we cannot assess how various management actions
51
will influence nutrient retention and transport.
52
Snowmelt runoff contributes significantly to nutrient loading in streams in northern
53
climates (e.g. Mitchell et al.10, Brooks and Williams11). The sources and dynamics of nutrient
54
transport during snowmelt are relatively well studied in forested watersheds, demonstrating that
55
freeze-thaw cycles and overland flow lead to nutrient losses from organic matter leaching and
56
soil microbes12,13. In contrast, knowledge of winter biogeochemical processes in northern urban
57
watersheds is much less advanced, in part due to reduced or suspended monitoring activity
58
during winter (however, see Kaushal et al.14 and Hall et al.15). Despite emphasis on warm season,
59
some studies have shown important contributions of winter urban runoff to pollutant loads (e.g.
60
Payne et al.16, Fallon & McNellis17, Oberts18, Brezonik & Stadlemann19). A recent study also
61
showed that urban BMP performance is impaired by winter conditions but also poorly
62
quantified20. Overall, far less is known about the sources and processes that control winter
63
transport of nutrient in urban watershed compared to summer, and compared to undisturbed
64
forests.
65
Decomposition of leaf litter can play a key role in nutrient availability in freshwater
66
ecosystems, releasing soluble forms of nitrogen and phosphorus to the water column21–23. In
67
urban landscapes, streams are often buried or channelized24, reducing the capacity for biotic
68
uptake of these nutrients25. In developed, residential watersheds, trees adjacent to streets drop
69
leaves and other organic matter directly into streets where they are carried via storm drains to
70
downstream aquatic features. In cities, streets, gutters and storm drains effectively act as
ACS Paragon Plus Environment
Page 4 of 35
Page 5 of 35
Environmental Science & Technology
71
headwater streams, collecting and delivering nutrients downstream, yet they likely lack the
72
function of forested headwaters, which can retain and quickly process nutrients. The contribution
73
of leaf litter to watershed loading of nitrogen and phosphorus via these urban headwater streams
74
(i.e. street gutters) is likely important26. While decomposition of leaf litter has been studied in
75
urban landscapes (e.g. McDonnell et al.27), few studies have evaluated how leaf litter
76
decomposition on paved surfaces contributes to urban water quality28–31.
77
Municipal street sweeping is a common pollutant control measure, but its effectiveness
78
for improving water quality has not been evaluated thoroughly31,32. Prompt sweeping can be
79
especially important for P, as leaf litter leaches dissolved P quickly and can contribute to nutrient
80
loading in urban runoff33,34. Leaf litter decomposes faster in forested urban areas than in
81
undeveloped landscapes27, and rapid decomposition has been demonstrated in street gutters as
82
well35. Therefore, over time, release of N and P might occur faster than expected from estimates
83
based on reference landscapes such as forests or streams. Despite the potential importance of leaf
84
litter to local water quality, municipal street sweeping in temperate cities often occurs
85
infrequently (e.g., annually or semi-annually) in residential watersheds. The effectiveness of
86
infrequent sweeping is further hindered by the fact that the timing of leaf drop can vary greatly
87
by tree species. For example, ash trees drop their leaves within a few days in fall months, while
88
oak trees can drop leaves weeks to months later (e.g. Reich et al. 36). Due to differences in
89
phenology, climate variability, and timing of sweeping operations, some amount of leaf litter is
90
frequently not collected during fall municipal street sweeping and remains in street gutters. The
91
fate of this over-wintering leaf litter is unknown, but likely contributes to export of nutrients in
92
snowmelt.
ACS Paragon Plus Environment
Environmental Science & Technology
93
In this paper we quantify snowmelt export of N and P and evaluate the contribution of
94
leaf litter to winter N and P loads in an urban, residential, storm-drained watershed, building off
95
previous work in this watershed35. To determine winter export of N and P, we monitored storm
96
water outflow and snowmelt in a residential watershed in Saint Paul, Minnesota during 2012-
97
2014. We studied leaf litter decomposition and nutrient release in street gutters, catch basins and
98
lawns in this watershed as well. Using these data sets and a model of leaf litter inputs, we
99
estimated the potential contribution of leaf litter to this winter export. Considering the physical
100
disconnect between soils and snowmelt flowpath caused by paved infrastructure, this research
101
addresses two key issues: (1) Does winter snowmelt contribute substantially to annual nutrient
102
loading in urban, residential watersheds as it does in forested watersheds? (2) Since leaf litter is
103
known to quickly leach P and is likely an important input to urban drainage, how much of
104
snowmelt P loading can be explained by over-wintering leaf litter in street gutters?
105 106 107
Methods Site description This study was conducted in a small, residential watershed in Saint Paul, Minnesota,
108
USA. The highly urbanized, residential watershed is 0.17 km2, with a total impervious area of
109
51% (sum of street, roof and other impervious percentages) and a vegetated area of 41%35. The
110
watershed drains into an underground BMP, the Arlington-Hamline Underground stormwater
111
vault (AHUG, Fig. 1). AHUG has no ponds, but it does contain multiple rain gardens and an
112
underground trench directly connected to storm drains engineered to promote infiltration and
113
reduce particulate nutrient export; however, since all of these BMPs are not connected to the
114
watershed outlet, we considered them disconnected from the watershed and they are not included
115
in total watershed area or estimates of nutrient and runoff volumes (see Fig. 1). There is no
ACS Paragon Plus Environment
Page 6 of 35
Page 7 of 35
Environmental Science & Technology
116
baseflow35 and hydrology is flashy (median event duration is 5.0 hours), with runoff arising
117
exclusively from rainfall events or thaws (see SI for hydrographs).
118
The watershed was resurfaced, repaved, and all stormwater and sanitary infrastructure
119
was replaced in 2006. Sewage is not likely present in stormwater, as sanitary and storm sewers
120
are completely separate in the watershed. Additionally, E. coli levels are generally lower in
121
AHUG than in stormwater across other Capitol Region Watershed District (CRWD) sites37, and
122
no illicit discharges have been found in AHUG by CRWD. Molar N:P ratios observed in
123
stormwater (mean: 22.4; see SI) are several times higher than observed for sewage (~4-538).
124
Climate data
125
Daily maximum temperature and snow depth for winters 2012/13 and 2013/14 were
126
measured by the Minnesota State Climatology Office at a station 3 kilometers from AHUG
127
watershed. These data can be accessed at http://www.dnr.state.mn.us/climate/historical.
128
Watershed Nutrient Export
129
The study watershed is managed by CRWD and the city of St. Paul. CRWD monitors
130
storm drains and surface water in the watershed, including the inlet of the stormwater vault
131
(hereafter, the outlet of AHUG watershed), as temperatures permit (April –Nov.). Water samples
132
are collected with an ISCO automatic sampler (Model 6712; Lincoln, NE) equipped with an
133
area-velocity module (ISCO Model 750) to continuously record water depth and velocity in the
134
storm pipe. Sample bottles were removed within 24 hours of storm events and a single composite
135
storm sample was generated and processed within 24 hours of retrieval39.
136
We collaborated with CRWD to monitor the watershed outlet of AHUG year-round
137
beginning in May 2012. When low temperatures prevented use of an automated sampler, we
138
manually sampled events. The vault was outfitted with a flow logger year-round (ISCO Model
ACS Paragon Plus Environment
Environmental Science & Technology
139
2150 during periods when auto-sampler was removed) for a complete record of water and
140
nutrient export from this watershed. Continuous outlet sampling only overlapped with our
141
decomposition study (described below) during winter 2012/13 (Dec. – March). We also sampled
142
fresh snowfall in AHUG on 13 Feb. 2011, 14 Feb. 2011, 23 Feb. 2011, 19 Nov. 2011, 22 Feb.
143
2013, 4 March 2013, and 12 April 2013 (7 events). Fresh snow was harvested using hand tools
144
and transferred to UMN in plastic gallon bags. Snow was then melted in same bags and
145
meltwater was filtered at the University of Minnesota (UMN) within 24 hours of collection.
146
Water quality analyses
147
Water quality analyses were conducted on warm season (April - Nov.) composite
148
samples collected by CRWD and analyzed at Metropolitan Council Environmental Services and
149
Pace Analytical39. During winter, and for additional samples collected during the warm season,
150
UMN staff collected composite and manual grab samples. A total of 215 flow events, including
151
both snowmelt (81 events) and rainfall-runoff (134 events), occurred over the study period (May
152
2012 – April 2014). 116 of these events, representing 84% of the total runoff volume, were
153
sampled by grab or by composite sampling (see table S-1 in the SI). Duplicate samples were run
154
for total dissolved phosphorus (TDP), soluble reactive P (SRP), nitrate and ammonium to ensure
155
comparability across laboratories and methods. Across 71 samples, no significant differences
156
were detected between labs (Mann-Whitney-Wilcoxon test, p2mm, i.e., leaf litter) and fine (
