An Assessment of Nitrogen and Phosphorus Availability in Constructed Wetlands in the Cumberland Marsh Region, Canada

Home , Isthmus of Chignecto, Missaguash River, Northumberland Strait, Tidnish River

An assessment of nitrogen and phosphorus availability in constructed wetlands in the Cumberland Marsh Region, Canada

Maxwell J. Turner

Thesis in partial fulfilment of the

requirements for the Degree of

Bachelor of Science with

Honours in Geology

Acadia University

April, 2016

The thesis by Maxwell J. Turner

is accepted in its present form by the

Department of Earth and Environmental Science

as satisfying the thesis requirements for the degree of

Bachelor of Science with Honours

Approved by Thesis Supervisors

______Dr. Ian Spooner Date

______Dr. Mark Mallory Date

Approved by the Head of the Department

______Dr. Ian Spooner Date

Approved by the Honours Committee

______Dr. Anna Redden Date

ii I, Max Turner, grant permission to the University Librarian at Acadia University to

reproduce, loan or distribute copies of my thesis in microform, paper or electronic

formats on a non-profit basis. I, however, retain copyright in my thesis.

______Maxwell Turner

______Date

iii Acknowledgements

I would like to extend recognition to Acadia University and Ducks Unlimited Canada, whose funding and dedication to scientific research made this project possible. Nic

McLellan of Ducks Unlimited provided both in-field help and a useful supply of regional knowledge. I would like to thank the entirety of the Department of Earth and

Environmental Science for providing a supportive learning environment that allows one to feel comfortable, acknowledged, and feel the expectation for success; but a special thanks to Dr. Rob Raeside whose subtle acknowledgements truly made me feel that this was the department to which I belonged. Separate from the world of science, a tremendous thanks is given to my family, who has always supported higher learning— and to my close friends, who consistently give me their honest and blunt opinion. My supervisors Dr. Ian Spooner and Dr. Mark Mallory deserve recognition for giving me valuable advice, much of their time, and guidance for the completion of this project, but also for being among the most dynamic and interesting people I have ever met. Finally, I am so grateful to have worked alongside Amanda Loder, whose relentless optimism, passion for science, and genuine concern for improving this world has given me an example of someone who will deserve every bit of success they enjoy in life.

iv Table of Contents

Acknowledgements………………………………………………………………. iv

Table of Contents………………………………………………………………… v

List of Tables…………………………………………………………………….. vii

List of Figures……………………………………………………………………. vii

Abstract……………………………………………………………….………….. ix

Chapter 1: Introduction…………………………………………………………… 1

1.1 Purpose…………………………………………………………………. 1

1.2 Wetlands………………………………………………………………… 2

1.3 Site Description……….………………………………………………… 4

1.4 CMR Geology and Geomorphology………...………………………….. 5

1.5 Anthropogenic Activity……………….………………………………… 10

1.6 Surface and Groundwater Interactions…………………………………. 12

1.7 Nitrogen………………………………………………………………… 13

1.8 Phosphorus……………………………………………………………… 14

Chapter 2: Methods and Materials………………………………………………... 17

2.1 Field Sampling………………………………………………………….. 21

2.2 Groundwater Sampling………………………………………………… 22

2.3 Nutrient Analysis: Total Nitrogen …………………………………….. 22

2.4 Nutrient Analysis: Total Phosphorus ………………………………….. 23

2.5 One-Time Water Sampling…………………………………………….. 24

Chapter 3: Results……………….……………………………………………….. 25

3.1 Site Characteristics……………………………………….….…………. 25

3.2 CMR Water Quality Parameters………………………….……………. 33

3.3 TN Surface Water………………………………………….…………… 33

v 3.4 TP Surface Water……………………………………………………… 34

3.5 Groundwater…………………………………………………………… 36

3.6 Temporal Total Nitrogen and Phosphorus Measurements…………….. 37

3.7 Water Level and Nutrient Input……………………………………….. 43

Chapter 4: Discussion……………………………………………………………. 46

4.1 Trends: Nitrogen………………………………………………………. 47

4.2 Trends: Phosphorus…………………………………………………… 48

4.3 Sources and Pathways: Nitrogen………….………………………….. 50

4.4 Sources and Pathways: Phosphorus………………………………….. 51

4.5 Recommendations and Management Implications…………………… 53

Chapter 5: Conclusions…………………………………………………………… 55

References…………………………………………………………………… 56

vi List of Tables

Table Number Description Page

2.1 Study sites and characteristics 19

3.01 Average water parameters and chemistry (UM1) 26

3.02 Average water parameters and chemistry (LM1) 27

3.03 Average water parameters and chemistry (LM2) 27

3.04 Average water parameters and chemistry (LM3) 28

3.05 Average water parameters and chemistry (LM4) 28

3.06 Average water parameters and chemistry (LM5) 29

3.07 Average water parameters and chemistry (LM6) 29

3.08 Average water parameters and chemistry (LM7) 30

3.09 Average water parameters and chemistry (A1) 31

3.10 Average water parameters and chemistry (A2) 32

3.11 Average water parameters and chemistry (A3) 33

vii List of Figures

Figure Number Description Page

1.1 Aerial image of the CMR with regional features 5

1.2 Regional geology of the CMR 7

1.3 Soil map of CMR with previous glacial pathways 9

2.1 Images of sample sites UM1, LM5, and A3 20

2.2 Portable spectrophotometer and water vial image 23

3.01 Box plots showing mean TN and TP levels 35

3.02 Groundwater TP and TN concentrations on aerial map 36

view of the CMR

3.03 Weekly TN concentrations in the Upper Missaguash 37

3.04 Weekly TP concentrations in the Upper Missaguash 38

3.05 Weekly TN concentrations in LM7 38

3.06 Weekly TN concentrations in the Lower Missaguash’s 39

newly constructed wetlands

3.07 Weekly TP concentrations in LM7 40

3.08 Weekly TP concentrations in the Lower Missaguash’s 41

newly constructed wetlands

3.09 Weekly TN concentrations in the Amherst Marsh 42

3.10 Weekly TP concentrations in the Amherst Marsh 43

3.11 Daily precipitation amounts over the study period 44

3.12 Temporal graphs mean weekly water level changes 45

and nutrient concentrations

viii Abstract

The Cumberland Marsh Region (CMR), located on the coast of the Bay of Fundy, is a major feeding ground for waterfowl and contains significant coastal wetland systems.

At this site there is concern over the occurrence of wetland senescence (the apparent decline of an ecosystem’s productivity), which appears to be a limiting factor in the viability of the CMR’s constructed wetlands, and results in management challenges. This study focuses on evaluating nitrogen (N) and phosphorous (P) in both surface and groundwater, and assessing the impact of landscape variability, and temporal and quantitative variations of natural and anthropogenic nutrient sources within altered and constructed wetlands. Water analyses were carried out weekly over a 3-month period on ten constructed freshwater wetlands, one altered freshwater wetland and three ground water sites. Results indicate that the impact of regional N and P sources on the sampled wetland sites is relatively low, and P loading is primarily autochthonous. These observations are supported by low N readings (<3 mg/L) at all sites with little seasonal variation, and higher (eutrophic) P levels that fluctuate without external input (40—300

µg/L) most notably in newly constructed wetlands. Although land use may not be an important contributor to N levels in surface water, a relatively high (5 mg/L) ground water reading suggests that anthropogenic sources exist. These results imply that nutrient sources, inputs, and pathways should be examined when choosing a potential wetland site to avoid the monetary and ecological losses associated with senescent constructed wetlands.

Chapter 1: Introduction

1.1 Purpose

The Isthmus of Chignecto is an important feeding ground for migratory and residential fauna and encompasses significant important coastal wetland systems.

Nutrients and water quality are critical factors that often govern productivity in aquatic ecosystems, and warrant investigation in constructed wetlands (Austin-Smith, 1998;

Craft, 1997). The purpose of this study is to investigate the variability of total nitrogen

(TN) and total phosphorus (TP) levels in constructed wetland systems, and assesses the influence of landscape variability on TN and TP levels in the Cumberland Marsh

Region’s (CMR) coastal wetlands.

The following questions will be addressed:

• What are the concentrations of TN and TP in selected constructed wetlands in the

Amherst and Missaguash marshes during June–August 2015?

• Do adjacent landscapes have a substantial impact on the nutrient loading to

constructed wetlands?

• Do groundwater concentrations of TN and TP suggest that groundwater flow may

contribute nutrients to constructed wetlands?

• Do internal site characteristics such as water chemistry and underlying soils/strata

affect TN and TP values?

• Which hydrologic pathway in constructed wetlands is dominant, how does it affect

nutrient levels, and will substantial rainfall mobilize allochthonous nutrients?

• What is the effect of anthropogenic activity on CMR landscape and water quality?

1.2 Wetlands

Wetlands are aquatic ecosystems that are inundated (periodically, seasonally or permanently) by surface water and/or groundwater, and may be characterized by high hydraulic flux and high nutrient concentration (Cook and Hauer, 2007). They are often cited as the most productive of all ecosystems, and contain abundant macrophytes and unique ecological niches (Mitsch and Gosselink, 2015). The simplest classification of wetlands can be separated into open and closed systems. Open wetlands are more common, and receive nutrients from upstream sources, runoff and overbank flooding.

Conversely, closed wetlands receive nutrients from atmospheric sources, mainly in the form of precipitation and rock weathering (Austin-Smith, 1998). Wetlands, particularly open wetlands, are efficient at removing nutrients from non-point sources by intercepting water, sediments and nutrients from adjacent uplands, and are helpful in removing harmful contaminants from water over a large area through biogeochemical processes

(Loder et al., 2016; Lockbay and Connor, 1999). In addition to nutrient (both N and P) and metal filtration, wetland ecosystems provide flood and storm attenuation, wildlife habitat and food supply, buffers to climate change, and ecotourism (Mitsch and

Gosselink, 2015).

Coastal wetlands form within a narrow geographical zone between land and sea, and have developed responses to the specific ranges and rates of sea-level change, sediment supply rates, and specific patterns of storm frequency (Michener et al., 1997;

Mitsch and Gosselink, 2015). Throughout history, humans frequently settled in coastal regions due to the accessibility of land and sea resources, thus coastal wetlands are very

2 susceptible to human modification which has led to major inputs of toxic materials, nutrients and sediments into these systems. Although wetlands are capable of removing many of these toxins from surface water, erosion from underlying bedrock is a natural occurrence that may increase the availability of toxic trace metals (Loder 2014; Michener et. al 1997).

Human alteration of wetlands has historically led to the widespread degradation of resident flora and fauna, as well as reducing water quality and/or flood attenuation over large areas (Mitsch et al., 1998; Zampella and Laidig, 2003). Man-made wetlands (which will be referred to as constructed wetlands or impoundments in this thesis), such as those in the CMR, are often built with the intent to rehabilitate ecosystems lost from past landscape modification (Mitsch et al., 1998; Zampella and Laidig, 2003). Although these wetlands are constructed with the intent to improve biodiversity, studies have shown that poorly designed wetlands will not compensate for habitat loss, and many knowledge gaps exist on what biogeochemical interactions are important to maximize productivity in constructed wetlands (Zelder, 2000; Winter, 1999).

Impoundments are shallow bodies of water (<2 m) enclosed by a dyke and manipulated with a water leveler. In the CMR, Ducks Unlimited Canada (DUC) is in charge of emplacing impoundments (Mansell et al., 1998). Construction of impoundments is a process that begins with a borrow pit dug near a potential wetland site; material removed from the pit is then used to build dykes that either enclose or partially enclose the impoundment, the low permeability dyke soils then retain precipitation, surface flow and groundwater (Maillet, 1997; Mansell et al., 1998; Payne

1992). A water control/outlet structure is installed through an adjacent dyke to manage

3 outflows and control water level (Payne, 1992).

1.3 Site Description

The CMR is located on the Isthmus of Chignecto (22 km in width), which also contains the Nova Scotia/New Brunswick Border. It connects mainland Canada to peninsular Nova Scotia, and it separates the Cumberland Basin’s cold, macrotidal waters of the south from the Northumberland Strait’s warmer northern waters (MacDonald and

Clowater, 2007; Monbet, 1997). The CMR is an expanse of low-lying land, reaching a maximum height of 90 m above sea level, and comprises a variety of wetland systems which include salt marshes, natural lakes, brackish ponds and constructed freshwater wetlands (Austin-Smith, 1998; MacDonald and Clowater, 2007). The isthmus is productive waterfowl habitat, as it a resting point along the major Atlantic Migratory

Flyaway, and thus is an important area for conservation and research (Dunnington, 2011).

There are four nationally protected areas of the CMR: Amherst Point Migratory Bird

Sanctuary, John Lusby National Wildlife Area, Tintamarre National Wildlife Area and

Chignecto National Wildlife Area (Environment Canada, 2015). The CMR was farmed in the past and in modern times; due to this, a network of channels and roads can direct surface flows containing particulate and dissolved forms of P (Kadlec, 1999).

The CMR is characterized by both maritime and continental climates, and experiences hot summers and cold winters. Precipitation is significant year round with high rainfall during the spring and autumn seasons (MacDonald and Clowater, 2007).

Figure 1.1. Aerial image of the CMR watersheds. The orange dashed line represents rolling structural ridges, the white dashed line represents the Nova Scotia-New Brunswick border and the yellow numbers represent sampling sites (image courtesy of Google Earth, 2015).

1.4 CMR Geology and Geomorphology

The Tantramar, Missaguash, and Amherst marshes are the three major watersheds that span the CMR (Austin-Smith, 1998). Each marsh is separated by structural ridges that trend northeast-southwest (Figure 1.1), from the Cumberland Basin to the

Northumberland Strait (Gussow, 1953; White, 2012). Five rivers (Tantramar, Aulac,

Missaguash, Laplanche, and Nappan) bisect the CMR and flow southerly from the northern upland region. The uplands contain fresh water marshes, bogs, and lakes while downstream is dominated by cultivated dykelands and salt marshes. The downstream region is influenced by the Bay of Fundy, which hosts some of the highest tides in the

5 world, with 160 billion tonnes of seawater entering and exiting the basin during the span of a single day (Austin-Smith and Bowes, 2000; Austin-Smith, 1998; Shaw et al. 2010).

Most of the CMR’s bedrock is composed of Carboniferous strata, consisting of coarse grained “soft” sandstone dating 300 million years. A thick cover of glacial till overlies the bedrock and underlies the modern wetland soils (Maillet et al., 1999; White,

2012). The strata (Figure 1.2) is characterized as one of the following: late Carboniferous

Pictou and Cumberland Group sedimentary rocks, early to late Carboniferous Mabou

Group sedimentary rocks, or the early Carboniferous Windsor Group chemical sedimentary rocks (Keppie, 2000). The Upper Windsor Group comprises mudstone, sandstone, minor gypsum and shallow marine limestones; and the Middleborough and

Shepody Formations (part of Mabou Group and relevant to study site) contains floodplain mudstone, fluvial sandstone, with rare conglomerates and limestone (Keppie, 2000).

Regional bedrock of the CMR is gently folded, with dipping synclines and anticlines that trend northeast-southwest (White, 2012).

Figure 1.2. Nova Scotian geology of the CMR. The Amherst Point Marsh is on the boundary of Windsor Group and Mabou Group rocks, and the Missaguash Marshes are entirely within the Cumberland Group range. The dashed line represents the Nova Scotia- New Brunswick Border.

Modern soils of the CMR (Figure 1.3) can be categorized into three main types:

Acadian soils, marshland peat bogs, and Masstown soils (Davis and Browne, 1996;

7 Austin-Smith Jr. and Bowes, 2000). Acadian soils are composed of red-brown or grey silty clay loams with imperfect drainage and little horizon development due to the constant influx of marine sedimentation (Davis and Browne, 1996; Austin-Smith Jr. and

Bowes, 2000). However, once drained and cultivated, they may be used for productive agricultural land—as evident by the Acadian soils found on coastal reclaimed dykeland, which extend several kilometres inland (Davis and Browne, 1996; White, 2012).

Marshland peat bog soils are located inland from areas bearing Acadian soils, and contain remnants of old forests are found between layers of peat (White, 2012; Davis and

Browne, 1996; Austin-Smith Jr. and Bowes, 2000). They have high organic content with poor drainage, and form on flat, low-lying marsh terrain (White, 2012; Davis and

Browne, 1996; Austin-Smith Jr. and Bowes, 2000). Masstown soils are poorly drained due to the hummocky nature of the terrain to which they belong (adjacent to marshland peat bog soils). The surface layer contains ≤ 15 cm of organic matter and is underlain by

~25 cm of thick sandy "A" horizon sediments (Davis and Browne, 1996).

Nova Scotia contains a complex array of glacial sediments, which are shown in ice flow records during the Wisconsinan glaciation period, and are depicted by Stea et al.

(1998) as ice flow events. Phase 1 occurred during the early-mid Wisconsinan, as ice flowed eastward across the Maritimes and into the Scotian Shelf (Stea et al. 1998; White,

2012). Phase 2 took place during the late Wisconsinan, as southeastern ice flow from the

Prince Edward Island region produced the north-south and northwest-southeast alignment of mainland Nova Scotia’s geomorphic features (Stea et al., 1998; White, 2012). Phase 3 was likely a period of deglaciation and southward ice flow (Stea et al., 1998). A proposed

Phase 4 (Chignetco Phase) suggests another period of freezing, when glaciation would

8 once again cover Nova Scotia (Stea et al., 1998). The Isthmus of Chignecto was ice free at least 13000 years ago (Dunnington, 2011; Stea et al. 1998). A warming trend that followed the Wisconsinan glaciation period resulted in the growth of vegetation, such as small shrubs, that enhanced slope stability to the topography province-wide (Dunnington,

2011). This stability reduced minerogenic input into lakes, and thus increased lake productivity and sedimentation rates (White, 2012).

Figure 1.3. Soil map of the CMR showing approximate glacial pathways of the Wisconsinan glaciation period phases 1, 2, and 3 (adapted from Englehardt, 2013).

9 The deposition of Fundy sediments occurs preferentially near banks of rivers and edges of salt marshes. This, along with a slight coastal topographical high, suggest that the Cumberland Basin was a lake during part of the Holocene epoch. These sediments were later covered in salt marsh sediments as relative sea level rose (Dunnington, 2011).

Today, the silty till plain and hummocky ground moraine glacial deposits attest to the area’s glaciated past (Dunnington, 2011).

Salt marshes form on submerging coast lines when the sedimentation rate exceeds the rate of land submergence (Austin-Smith, 1998). The Chignecto Basin, which is located at the head of the Bay of Fundy (Figure 1.1), is an example of a submerging coastline—with sea level rising approximately 3 mm per year (Austin-Smith, 1998; Shaw et al. 2010). During the Wisconsinan ice age, large glaciers in the Hudson Bay depressed the middle of the continent (Greenburg et al., 2012). Massive weight of the large scale glaciation caused the margins (the modern Bay of Fundy’s location) to tilt upward, reaching the edge of the continental shelf (Greenburg et al., 2012). Today, the middle of the continent is experiencing isostatic rebound, and the margins are submerging— resulting in a transgression (Greenburg et al., 2012).

1.5 Anthropogenic Activity

Chignecto Bay was named by the Mi’kmaq and translated means “Great Marsh

District”. The Mi’kmaq took advantage of the highly productive ecosystems and high numbers of waterfowl and shellfish for hunting and fishing before the arrival of the

European’s in the mid-1600s (MacDonald and Clowater, 2007).

The arrival of the Acadians in the 1670s marked the beginning of a widespread transformation of the region. The salt marsh that once dominated the CMR was altered to

10 agriculture land with the use of dyking practices and draining the region’s salt marshes to access the underlying productive soils (Austin-Smith, 1998). 70% of the region was converted to fields for hay cultivation (specifically valuable marsh hay, Spartina), and land for livestock grazing (MacDonald and Clowater, 2007). The Acadian dykes were not only built on the coast, but also upland and along rivers and creeks (Austin-Smith, 1998).

Today, the CMR is characterized as predominately rural and contains many small- scale agriculture operations that are predominantly livestock farms and blueberry farms

(MacDoanld and Clowater, 2007). The towns of Sackville, NB (72.1 people/km2) and

Amherst, NS (787.0 people/km2) are the main urban centres in the area, each located several kilometres outside of the study areas (MacDoanld and Clowater, 2007). Seven two lane roads transect the CMR, including the Trans-Canada Highway. There are three main locations in which industrial forestry is prevalent: north of the Tintamarre National

Wildlife Area, the boundary of the Missaguash Marsh and Tidnish River watershed, and upland of the Missaguash Marsh, located on the Nova Scotia-New Brunswick Border— these industries mainly employ clear cutting techniques (MacDonald and Clowater,

2007). Anthropogenic activities have the potential to effect nutrient balance in the CMR’s wetland systems (Zheng and Stevenson, 2006). Nutrient accumulation and storage from agricultural runoff will alter productivity, species diversity, and water quality (Zheng and

Stevenson, 2006).

Modern constructed wetlands in the CMR were built and managed through a joint effort between provincial/federal applications and Ducks Unlimited Canada (DUC). This was done in an effort to reclaim wetland habitat lost by dyking practices of the past

(Englehardt, 2013).

11 1.6 Surface and Groundwater Interactions

Lakes, streams, rivers and wetlands play an integral role in regional groundwater interactions, as they are affected by groundwater flow systems, geological characteristics of underlying beds, and local climate (Winter, 1999). Topography often determines surface and groundwater recharge and discharge areas—a recharge zone will collect precipitation in a topographic high and transport the water to the discharge area, which is the topographic low (Winter, 1999). However, groundwater from local flow systems tend to discharge into nearby wetland systems, regardless of the topography (Winter, 1999).

Wetlands receive water from three main sources: precipitation, lateral surface flows, and groundwater, which are all influenced by the degree of connection, salinity gradients, degree of slope or channel gradient, position in the landscape, and watershed size (U.S. EPA. 2002).

Water quality of a wetland is determined by the geological formation of its hydraulic catchment and whether inflow includes wastewater (Kansiime et al, 2007).

Regarding water quality parameters in differing topographic settings—Zampella and

Laidig (2003) found significant difference between the mean shore slope, pH, specific conductance, and total dissolved carbon (TDC) between a natural and excavated pond.

Steeper slopes, pH, and specific conductance were higher in excavated ponds than their natural counterparts (Zampella and Laidig, 2003).

12 1.7 Nitrogen

Nitrogen (N) is a macronutrient that often limits productivity in aquatic systems, particularly in brackish and saline ecosystems. It is not clear whether N is the limiting nutrient in aquatic systems falling between brackish and fresh, such as a fresh water wetland underlain by salt marsh sediments (Mitsch and Gosselink, 2015). Total Nitrogen

(TN), a water quality parameter often analyzed to assess productivity and is the sum of

+ nitrate (NO3), nitrite (NO2), ammonia (NH4 ) and organic nitrogen (Mitsch and Gosselink,

2015). A range of 2 mg/L—6 mg/L in TN concentration is usually tolerated in drinking water, but depending on local governing standards, up to 45 mg/L is acceptable (Heath,

1983).

Most N enters aquatic ecosystems through the biological fixation of atmospheric N

(N2), which comprises the majority (78%) of the Earth’s total atmosphere (Mitsch and

Gosselink, 2015). However, aquatic ecosystems are also susceptible to anthropogenic sources of TN that can enter a system from wastewater treatment plants, runoff from fertilized lawns and croplands, failing septic systems, runoff from animal manure and storage areas, and industrial discharges that contain corrosion inhibitors (Schlesginer,

2009). The levels of nutrients entering an aquatic system are controlled by the hydrogeological catchment of a region, with discharge areas (topographic low) accumulating more N than areas of higher elevation (Schlesginer, 2009; Winter, 1999).

Denitrification is a permanent loss of N in a system, and is limited by nitrate

(Austin-Smith, 1998). Availability/retention of N in wetlands is controlled by biological processes that belong to a succession of events (for example, organic matter accumulation related to denitrification). In a newly constructed salt marsh, vegetation

13 may rapidly attain 100% cover, leading to fast N retention (Austin-Smith, 1998; Zedler,

2000). Nitrogen retention begins with initial vegetation growth, and for the first 1-3 years, low amounts of denitrification will occur. After 5-10 years have passed, the constructed salt marsh will begin to accumulate sufficient organic carbon to cause denitrification. Once this threshold has been reached, the same amounts of N accumulated will be removed (Zedler, 2000).

1.8 Phosphorus

Phosphorus (P) is a macronutrient that often limits productivity in freshwater lakes.

It is common for a wetland ecosystem to have P concentrations high enough to be considered hypereutrophic (under the lake trophic level classification), due to their high productivity—with bogs being noted as the unproductive exception to wetland ecosystems (Taylor, 2005; Mitsch and Gosselink, 2015). Total nutrient concentrations

(e.g., TP and TN) are considered the best chemical indicators of trophic status; measuring the TP concentration in an aquatic system is normally done by calculating the sum of the reactive, condensed, and organic phosphorous (Taylor, 2005). Total P includes both organic (such as ortho-phosphate, the chemically active dissolved form of P) and inorganic species of P (ASA Analytics, 2016; Environmental Laboratories, 2016; Taylor,

2005).

Phosphorus has been a bio-critical nutrient that has shaped evolution, but it was often in short supply (Filippelli, 2008). Before human existence, P cycling through ecosystems was often slow and sporadic, and usually relied on tectonic processes that would heighten chemical erosion of terrestrial rocks (Filippelli, 2008). Crustal erosion is the only way P can naturally enter ecosystems, due to the element lacking a stable

14 atmospheric gas phase (unlike N). Although P comprises only 0.09%wt of the Earth’s crust, higher concentrations of the element can be released through the degradation of high P-bearing minerals, such as apatite. However, the P cycle is often driven by anthropogenic activity today (Filippelli, 2008).

Phosphorus may enter a system in a variety of ways, both natural and anthropogenic. Major anthropogenic sources include human and animal wastes, soil erosion, detergents, septic systems and runoff from farmland or fertilized lawns. Internal loading is a natural phenomenon and causes P fluctuations in a watershed during the course of a year. When the bottom of a fresh water body enters a state of anoxia

(frequently during late summer or winter months), a chemical reaction at the sediment- water interface causes the unavailable P at geochemical adsorption sites to release into the water column, which is quickly used by biota uptake (Dunne and Reddy, 2005;

Filippeli, 2008, Wilcock et al., 2007; Zheng and Stevenson, 2006). The availability of P is regulated by geochemical processes, such as adsorption, and does not cycle as easily as

N, which relies mainly on biological pathways (Craft, 1997).

Removal of P in wetlands depends on the presence of factors that cause the removal of dissolved inorganic phosphorus (DIP). Microbial and plant uptake, soil adsorption, and incorporation of organic P into soil peat are three main processes that remove DIP in a system (Richardson, 1985). Microbial uptake is often the initial DIP removal factor, and rooted emergent wetland vegetation will uptake significant levels of phosphate. However, tissue death will cause 35-75% of plant P to be released back into the system, making this factor only a short term nutrient sink (Richardson, 1985). Phosphorus has the ability to bind and become locked up at various geochemical adsorption sites which include

15 aluminum, iron and calcium minerals, clay minerals, and some organics (Dunne and

Ruddy, 2005; Richardson, 1985).

Previous research indicates that the first 1-3 years of a constructed wetland’s life, the P entering the system is quickly removed (Craft, 1997). Removal is faster during this time due to sediment deposition, resulting in sorption and deposition of P over time

(Craft, 1997; Austin-Smith, 1998). Once these sorption sites become saturated with P, retention levels decrease and thus surface water levels may increase. Although open wetlands are effective at removing P, their effectiveness declines over time due to the nature of P sorption sites becoming saturated and become unable to retain further P

(Craft, 1997).

Phosphorus cycling in wetlands is controlled by the water, soil and biosphere.

Movement through surface water flow is most common, with subsurface flow occurring depending on soil type and regional geology. Certain chemical sediments, such as limestone, allow a greater interaction between surface and subsurface flow systems (Craft

1997; Dunne and Reddy, 2005; Filippeli, 2008). Large amounts of annual rainfall and single storm events affect the water elevation and storages within a region (such as

CMR). Total P levels tend to increase lower in a watershed compared to upstream locations, this is due in part to the cumulative effect of a higher concentration of anthropogenic activity in the lower portions of watershed, but also to natural changes in rivers which occur during their lifespan (Environment Canada, 2016).

16 Chapter 2: Materials and Methods

Water collection and sampling of ten constructed wetlands (impoundments) and one altered wetland was conducted between June 2015—August 2015. Total N and Total

P concentration levels were recorded as part of an overarching productivity assessment of the CMR. Change in surface water level was recorded to interpret water level fluxes in these wetland systems. Water chemistry parameters (pH, dissolved oxygen, conductivity, total dissolved solids, temperature, salinity and secchi depth) were recorded on site, and the elemental analysis of TN and TP was conducted at the Beaubassin Research Centre.

Each site was sampled once per week, with the exception of the Lower Missaguash

Marsh’s new impoundments (LM1, LM2, LM3, LM4, LM5, and LM6—Figure 3.06) which were sampled bi-weekly until the third-to-last week, when weekly sampling was resumed. Sampling took place between 9:00am and 12:00pm for consistency.

The eleven freshwater sites examined in the CMR were situated in one of three geographic “zones”: Upper Missaguash, Lower Missaguash and Amherst Point (Figure

1.1). Each zone has varying underlying strata, soil type, hydrologic catchment, and anthropogenic use. The Upper Missaguash contains one relatively old (~41 years) constructed wetland site, and is located in an area with several nearby small-scale agricultural operations. The Lower Missaguash’s seven sites are located close to the

Chignecto Basin and are potentially being affected by salt water flux. Six are newly constructed impoundments that were flooded in April 2015, and one is 30 years old. The

Missaguash sites are underlain by Carboniferous sandstone—with the Upper Missaguash site located on land that was once a bog, and the Lower Missaguash site on land once dominated by salt marsh. The Amherst Point sites were constructed in the 1970s, and are

17 partially underlain by the Upper Windsor Group (chemical sediments). Amherst Point 3

(A3) is connected and receives inflow from a nearby lake, thus is considered an altered wetland (Figure 3.09; Figure 3.10). Additionally, it was dyked and had a water control structure installed by DUC and the Canadian Wildlife Services (CWS).

Table 2.1. Abbreviations: Amherst Marsh (A), Lower Missaguash Marsh (LM), Upper Missaguash Marsh (UM). Abbreviated names will be used throughout this thesis.

Figure 2.1. Wetlands is the CMR. (1) UM1: constructed wetland, (2) LM5: constructed wetland, (3) A3: altered wetland

20 2.1 Field Sampling

Weekly water samples were collected at central, deep points of the Lower

Missaguash impoundments, and offshore the Amherst Point Bird Sanctuary and Upper

Missaguash impoundments. Samples were collected in white translucent polyethylene bottles with a capacity of 0.5 L that had been acid sterilized and rinsed in triplicate with wetland water. Water samples collected from the Lower Missaguash impoundments were reached by inflatable boat, and obtained approximately 0.40 m under the water’s surface.

Samples from the Amherst and Upper Missaguash sites were reached by attaching a polyethylene bottle to a metal painter’s pole, which extended a maximum of 3.66 m. The painter’s pole was used to overcome the difficulty of transporting an inflatable boat to impoundments with difficult accessibility. After the samples were collected, they were stored and covered in a cooler pack at approximately 4℃. Before the analyses took place, the samples were removed from the cooler pack and brought to room temperature.

Water chemistry data was collected on site using a Hanna Instruments 9828 multi parameter probe. The parameters recorded included dissolved oxygen content, pH, conductivity, total dissolved solids, salinity, and temperature. Turbidity was measured on site by using a secchi disk which was slowly lowered into the water at the sampling site.

When the disk was no longer visible, a reading in meters was recorded. As the disk was brought back to the surface, a second reading was recorded when it became visible again.

These two readings were averaged and recorded for the secchi depth measurement. When the bottom was visible, the value was recorded as “greater than wetland depth” (e.g. >1 m). Over the sampling season, if a specific site received both secchi depth values and

“bottom visible” values, the mean value was recorded by averaging both sets of values.

21 Wetland water depth relative change was recorded using a 0.91 m pole marked with depth measurements in 1-cm increments. These data can be used to understand the effect of surface flow and rainfall inputs at each site. All data was recorded using a Microsoft

Excel spread sheet.

2.2 Groundwater Sampling

One-time groundwater samples were collected from three drilled wells in each of the geographic marsh zones. Water quality parameters were excluded in the assessment of groundwater, and the samples followed the same analytical procedure that the surface water samples followed for nutrient analysis.

2.3 Nutrient Analysis: Total Nitrogen

Nutrient analysis was accomplished using a spectrophotometer, which works by providing a source of light (lamp) that strikes a diffraction gradient, working like a prism to scatter light (Hach, 2016). Only a specific wavelength is able to reach the exit slit, which interacts with the sample. A detector measures the transmittance and absorbance of light by the sample—a value is given based on the comparison to the blank (Hach, 2016).

The persulfate digestion method was used to measure TN levels (detection limit of

0.5 mg/L—25.0 mg/L). A TN persulfate reagent powder pillow was added through a plastic funnel to several TN hydroxide digestion reagent vials (Hach, 2016). The samples collected earlier in the day were added to individual vials (2 mL) and 2 mL of demineralized water was added for preparation of a blank sample. The vials were shaken for 30 seconds and placed into a digital reactor block set at 105°C. Thirty minutes was allotted for heating, after which, they were removed and left to cool to room temperature

(Hach, 2016). Total N reagent powder pillows were added through plastic funnel to the

22 vials. Then, 2 mL of each sample and blank were pipetted into separate TN regent C vials, followed by 10 vial inversions (Hach, 2016). The vials were then placed in a Hach

DR/2400 portable spectrophotometer.

Figure 2.2. Hach DR/2400 portable spectrophotometer (background) and vials containing UM1 water samples tested for TN (foreground)

2.4 Nutrient Analysis: Total Phosphorus

The USEPA PhosVer® 3 with Acid Persulfate method was used to asses TP levels.

5 mL of each sample and persulfate powder were added to separate TP test vials. The vials were shaken until all powder was dissolved, and then placed in a digital reactor block set at 105°C. The vials were heated for 30 minutes, then removed and allowed to cool to room temperature (Hach, 2016). 2 mL of 1.54 N sodium hydroxide standard

23 solution was added to watch vial and promptly inverted. One PhosVer 3 powder pillow was then added and shaking for 20-30 seconds began (Hach, 2016). Once each vial had been shaken, a 2-minute reaction took place before being assessed by the spectrophotometer.

Duplicates, triplicates and standards were taken regularly. The standard solutions used were a 5.0 mg/L nitrate standard and a 1.0 mg/L phosphate standard.

2.5 One-Time Water Quality

One-time water samples were taken at each surface water site, and sent to Quebec

National Research Institute to analyze chlorine (Cl), sodium (Na), calcium (Ca) and sulfur (S) concentrations. This was done to understand general concentrations of other cations and anions within the constructed wetlands, and assess if potential interplay between N or P was occurring. The Schimadzo technique was used to analyze Cl concentrations while the remaining elements used the Autoanalsyer Lachat Colorimetry method (Shimadzu, 2016).

24 Chapter 3: Results 3.1 Site Characteristics

The eleven sites chosen were characterized by water chemistry (pH, DO, temperature, turbidity, salinity, TDS), physical parameters (age of construction, geological location, morphology and anthropogenic activity) and nutrient levels (TP,

TN). The study region was partitioned into three areas, two within the Missaguash Marsh

(Upper, Lower) and one within the Amherst Marsh. Each site will be described in terms of water chemistry, field observations, nutrient level and seasonal variation.

3.11 Upper Missaguash Marsh

The Upper Missaguash Marsh is situated in close proximity to the Missaguash

River. The marsh is positioned on the convex portion of a regional antiform, and is enclosed by rolling structural ridges. Rural settlements and small agricultural operations are proximal to constructed wetlands within the marshland, making these wetlands potential catchment sites for surface runoff. The impoundments were constructed in the

1960s and 1970s on salt marsh sediment and were the first wetlands constructed in the

CMR (Austin-Smith, 1998).

Upper Missaguash 1 (UM1)

UM1 is located approximately 150 m west of the Nova Scotia/New Brunswick border. The impoundment was constructed in 1974, dug into former agriculture land, and partially enclosed by dykes. The site may receive overflow from other sources due to its partially enclosed nature and proximity to a man-made canal. Notable features surrounding the site include a fish ladder (constructed 30-40 years prior to the study), a canal that extends laterally adjacent to UM1, and several abandoned agriculture fields

(Figures 3.03; Figure 3.04). Between 4 August, 2015—12 August, 2015 UM1 had a

25 significant proportion (~30-40%) of its vegetation cleared by a DUC cookie cutter machine (Austin-Smith, 1998). The amount of sediment re-introduced into the water column by the machine is unknown. The site is accessible by a dirt road off of Highway

16.

Table 3.01. Upper Missaguash 1: Average Water Parameters and Chemistry

3.12 Lower Missaguash Marsh

The Lower Missaguash Marsh is located adjacent to the Missaguash River, between the Cumberland Basin and Trans-Canada Highway. Aulac, New Brunswick is a popular resting and refueling site and is the closest settlement to the marsh. The marsh’s coastline is heavily dyked, protecting the land from the Bay of Fundy’s macro-tides. The landscape is covered by grassy fields and (constructed) wetlands, with minimal tree cover. The marsh is situated on former farmland, which has been re-established as the Beaubassin

Research Centre (Figure 3.05; Figure 3.06; Figure 3.07; Figure 3.08).

Lower Missaguash 1 (LM1)

LM1 is the northernmost cell of the six newly flooded wetlands in the Lower

Missaguash Marsh—which are adjacent to each other, and lie separated by dykes in a continuous, lateral fashion (Figure 3.06; Figure 3.08). Collectively, the cells are bordered by a small channel on their eastern side. LM1 has not reached the water level operating height set by DUC (~1 m) upon construction. During the study, the 180 m by 100 m site

26 was composed of a small, shallow channel surrounded by salt marsh sediments and vegetation. The channel itself is the remnant of a former Missaguash River intercept and has a width of approximately 10 m. Compared to its neighboring cells, aquatic vegetation was sparse, and less wildlife was noted.

Table 3.02. Lower Missaguash 1: Average Water Parameters and Chemistry

Lower Missaguash 2 (LM2)

LM2 is approximately 210 m by 90 m and is bordered by LM1 (north) and LM3

(south). The cell reached the water level operating height set by DUC (~45 cm) and was completely flooded at the time of sampling.

Table 3.03. Lower Missaguash 2: Average Water Parameters and Chemistry

Lower Missaguash 3 (LM3)

LM3 is approximately 185 m by 120 m and is bordered by LM2 (north) and LM4

(south). The cell reached the water level set by DUC (~22.5 cm) and was completely flooded at the time of sampling.

Table 3.04. Lower Missaguash 3: Average Water Parameters and Chemistry

Lower Missaguash 4 (LM4)

LM4 has the longest plot of all newly constructed Lower Missaguash sites and is approximately 160 m by 26 m. The cell is bordered by LM3 (north) and LM5 (south).

LM4 is completely flooded and is noted for having the largest area of open water of the six newly flooded Lower Missaguash Marsh impoundments (Figure 3.06; Figure 3.08).

Table 3.05. Lower Missaguash 4: Average Water Parameters and Chemistry

Lower Missaguash 5 (LM5)

LM5 is 265 m by 65 m and is bordered by LM4 (north) and LM6 (south). The small channel that bordered the four previous cells stops at LM5 (Figure 3.06; Figure 3.08).

The cell is completely flooded (~22.5 cm in depth) and it was noted (during the study season) for containing the greatest variety of wildlife out of the six newly constructed cells. Nesting waterfowl and smaller birds appeared frequently throughout the sampling season. LM5 had a markedly greater conductivity, Na and Cl- values than LM1, LM2,

28 LM3, and LM4.

Table 3.06. Lower Missaguash 5: Average Water Parameters and Chemistry

Lower Missaguash 6 (LM6)

LM6 is the southernmost cell and most proximal to the Cumberland Basin. At the time of sampling, it was considered an “unsuccessful” impoundment as only ~30% of the

250 m by 60 m cell flooded. The flooded portion consists of a channel ~3 m in width that runs parallel to the dykes that enclose the cell (Figure 3.06; Figure 3.08). LM6 recorded a markedly greater conductivity, Na and Cl- than LM1, LM2, LM3, LM4 and LM5. It also recorded the lowest average pH reading, and largest seasonal fluctuation as the pH increased from 5.65 in late May to 6.58 in late August.

Table 3.07. Lower Missaguash 6: Average Water Parameters and Chemistry

Lower Missaguash 7 (LM7)

LM7 was flooded (~0.5 m in depth) and completely dyked 30 years prior to the

29 study. It is disconnected from the other Lower Missaguash Marsh impoundments, is approximately 330 m by 350 m in area, and borders the Trans-Canada Highway. The site is accessible by a dirt road off of Brown Road, and is routinely accessed by Canadian

Wildlife Services (CWS) for bird counts and tags. LM7 is noted for thick vegetation cover, rendering some flooded areas inaccessible by boat in the latter half of the sampling season (Figure 3.05; Figure 3.07). Dissolved oxygen levels were markedly lower in LM7 than all other sites.

Table 3.08. Lower Missaguash 7: Average Water Parameters and Chemistry

3.13 Amherst Point Marsh

The Amherst Point Marsh is located in a provincial park (protected area) south of

Amherst, NS. The marsh has significant tree coverage and contains several constructed and natural wetlands, lakes, and ponds. These surface waters are accessible by a series of walking trails and bridges throughout the provincial park. The park was built on abandoned agriculture land during the 1970s, and similar land can be observed surrounding the park. Small scale agriculture operations (e.g., hay production and livestock housing) exist within the region.

A salt mining facility operated by Sifto Canada is located approximately 1.5 km from the study site, and produces much of the province’s table salt. Production is driven

30 by solution mining, where hot water is pumped underground and dissolves the evaporite rocks that comprise the Windsor Group strata. A salt brine is then removed which produces refined salt (Rosar, E.C. 1993). The park itself is bordered by a secondary highway connecting Amherst to the nearby community of Nappan, and is located approximately 2 km from the Cumberland Basin. The secondary highway that runs northwest is on a slight topographic high, 10-15 m above the provincial park, creating a slight barrier from the ocean.

Amherst Point 1 (A1)

A1 is accessible from a small dirt pathway off of Southampton Road. The road provides a barrier between the site and nearby agricultural fields. The perimeter of A1 is dyked, and borders several other impoundments (e.g., A2). The site is partially dyked and has a water level set at ~0.5 m, but it is unknown if any portion the dyke is breached allowing water flow between adjacent impoundments (Figure 3.09; Figure 3.10).

Table 3.09. Amherst Point 1: Average Water Parameters and Chemistry

Amherst Point 2 (A2)

A2 is accessible from a walking trial within the Amherst Point Bird Sanctuary. The cell’s perimeter is partially dyked (water level at ~0.5 m) and borders adjacent impoundments and forested landscape. The northwest dyke is breached, allowing water

31 flow between adjacent impoundments (Figure 3.09; Figure 3.10). A series of telephone lines stand at the impoundment’s southern border.

Table 3.10. Amherst Point 2: Average Water Parameters and Chemistry

Amherst Point 3

A3 is accessed by a walking trial within the Amherst Point Bird Sanctuary (Figure

3.09; Figure 3.10). The wetland is close to its natural configuration, and has only been slightly altered through the construction of nearby impoundments and the addition of an outlet that is dyked and contains a water leveler. The site receives inflow from a culvert attached to Layton’s Lake, the lake includes deep (up to 30 m) sinkholes that were formed through the dissolution of underlying gypsum deposits in the Windsor Group strata (Howell and Kerekes, 1987). Its outflow channel was dyked by the Canadian

Wildlife Service in 1983 for the installation of a water leveler (water level kept at ~0.5 m). A3 contained markedly higher conductivity, Ca and S values than adjacent sites.

Sampling did not begin until week 4 of the season.

Table 3.11. Amherst Point 3: Average Water Parameters and Chemistry

3.2 CMR Water Quality Parameters

Water chemistry in the CMR was variable amongst the study sites. Some of the tested parameters were highly sensitive to the time of day and current weather conditions

(temperature, dissolved oxygen). The majority of the sites had sufficient dissolved oxygen readings to be characterized as aerobic, with an overall average of 6.44 mg/L.

However, the altered wetland (A3) had markedly higher dissolved oxygen concentrations, averaging 9.31 mg/L over the season. The pH readings ranged from 5.16 (LM6) to 9.76

(A1), with LM6 exhibiting markedly lower pH. Conductivity was highly variable amongst the sample sites and ranged from the detection limit (which was 0 µs/cm) to 941

µs/cm (LM6). A3, LM5, and LM6 exhibited markedly higher conductivity values, and conductivity was not detected in LM7 and UM1 during this study.

3.3 Total Nitrogen in Surface Water

Collectively, TN concentrations were low in the CMR constructed wetland’s surface water. The range of recorded values in each site was also low and only varied by

1-2 mg/L throughout the season. No apparent seasonal trend was noted in TN values. The highest recorded value (5.1 mg/L at A1), is below the MAC recommended for drinking water (Heath, 1983). It was noted that TN was higher in newly flooded impoundments,

33 and lower in wetlands over 30 years old. Notable values were tested on 18 August, 2015 in sites A1 and A2, producing concentrations that were markedly higher than previous values.

3.4 Total Phosphorus in Surface Water

High TP concentrations were observed in the CMR’s surface water sites. The lowest recorded value of 0.04 mg/L (A3) was still high enough to warrant a eutrophic classification by the Environment Canada’s tropic status classification guide (0.035—

0.100 mg/L). TP concentrations were highest and most variable in the Lower

Missaguash’s newly constructed wetlands, which gradually increased to hypereutrophic levels (>1.00 mg/L) by the end of the summer. Concentrations were markedly lower in the Amherst and Upper Missaguash Marshes, where all sites were constructed/altered more than 30 years prior to the study. Water clarity, which was evaluated by secchi disk measurement, decreased as TP levels increased throughout the season in the Lower

Missaguash’s newly constructed wetlands.

34 Newly (<1 year) 0.5 Flooded

Flooded > 30 0.4 Years Ago

0.3

0.2 Total Phoshporus (mg/L) Phoshporus Total

0.1

A1 A2 A3 LM1 LM2 LM3 LM4 LM5 LM6 LM7 UM1 Site

5.0 Newly (<1 year) Flooded

4.0 Flooded > 30 Years Ago

3.0

2.0 Total Nitrogen (mg/L) Nitrogen Total

1.0

A1 A2 A3 LM1 LM2 LM3 LM4 LM5 LM6 LM7 UM1

Site

Figure 3.01. TP (top) and TN (bottom) concentrations measured in constructed and altered wetlands in the CMR during the summer of 2015. LM1 and LM6 are controls for flooding. The samples were collected between 3 June, 2015—25 August, 2015.

35 3.5 Groundwater

The highest levels of TN in the CMR were detected in groundwater in the Upper

Missaguash Marsh (5.0 mg/L). Total N concentrations in groundwater samples were at or below the detection limit in the Lower Missaguash Marsh and Amherst Marsh. Total P concentrations in groundwater varied across the CMR, ranging from levels that would be considered meso-eutrophic in the Lower Missaguash Marsh (0.03 mg/L) to eutrophic levels in the Amherst Marsh (0.09 mg/L). The lowest TP value of the season was recorded in the Lower Missaguash’s groundwater.

Figure 3.02. One-time groundwater samples showing TN and TP concentration levels throughout the CMR.

36 3.6 Temporal Total Nitrogen and Phosphorus Measurements

For the purpose of the temporal assessment of TN and TP in the Lower

Missaguash, sites LM1 and LM6 will be considered “controls for flooding” and will not be included in the overarching assessment of nutrient availability within the Lower

Missaguash Marsh. Sediment disturbance occurred upon construction, but water levels did not reach operating levels (between 22.5 cm and 45 cm), nor did it flood vegetation.

This could potentially skew results, as the other newly constructed Lower Missaguash sites (LM2-5) reached desirable operating levels.

3.61 Upper Missaguash Marsh

TN levels in the Upper Missaguash Marsh do not appear to follow a seasonal trend, other than a slight increase noted in early July. Surface water TN levels were consistently lower than groundwater.

Figure 3.03. Weekly TN concentrations in the Upper Missaguash Marsh.

Total P fluctuation was also less pronounced than in the Lower Missaguash site, peaking in late July then tapering off at the end of the sampling period.

Figure 3.04. Weekly TP concentrations in the Upper Missaguash Marsh.

3.62 Lower Missaguash Marsh

Total N in the newly constructed Lower Missaguash sites slightly decrease slightly in week 4 of the study (late June) followed by an overall increase the following week in early July (Figure 3.06). However, the changes are only slight, as overall TN concentrations were low in the CMR.

Figure 3.05. Weekly TN concentrations in the Lower Missaguash Marsh: site LM7 (old).

38 Figure 3.06. Weekly TN concentrations in the Lower Missaguash Marsh (excluding LM7).

39 Seasonal variability of TP was greatest in the Lower Missaguash Marsh. Values gradually increased to hypereutrophic levels, and peaked around week 8 (late July) of the study. The sites then experienced a nutrient level decline during the final weeks of the study (late August), with a TP resurgence in LM3 during the last week (Figure 3.08).

Figure 3.07. Weekly TP concentrations in the Lower Missaguash Marsh: site LM7 (old).

Figure 3.08. Weekly TP concentrations in the Lower Missaguash Marsh (excluding LM7).

41 3.63 Amherst Point Marsh

There does not appear to be any temporal correlation with TN concentrations in the

Amherst Point Marsh. Levels reach their peaks at differing times throughout the season in each site (Figure 3.09).

Figure 3.09. Weekly TN concentrations in the Amherst Point Marsh.

Total P concentrations follow a similar, but less pronounced trend of a gradual increase then tapering off as the sample period concluded. A1 did not have a significant peak of TP concentration during the sampling season (Figure 3.10).

Figure 3.10. Weekly TP concentrations in the Amherst Point Marsh.

3.7 Water Level and Nutrient Input

Water levels greatly fluctuated in the CMR wetlands during the study period, with precipitation likely being the main driver of water level change. Between 21 June,

2015—22 June, 2015 the region experienced a large rainstorm event from the remnants of Tropical Storm Bill. Regional data from the Nappan, NS weather station indicates that

108.8 mm of rain fell over a 24-hour span (see Figure 3.11), however, local reports from nearby communities estimate up to 147.2 mm of rainfall (Environment Canada, 2016).

This was the largest regional rainfall event in 2015, and the 24-hour rainfall was similar to the total rainfall amounts expected for the month of June within the CMR. Affected communities experienced localized flooding, especially in low-lying areas. Main roads in

Sackville, NB were closed due to the flooding, and a nursing home in Amherst, NS required an evacuation (The Weather Man, 2016).

Figure 3.11. Daily precipitation values during the study period. The June 21 spike is attributed to a large rainfall event caused by Tropical Storm Bill.

The event caused significant water level changes throughout the study sites.

Amherst Point Marsh experienced the greatest amount of water input during this event, averaging a 27 cm rise in water level within the wetlands following the rainfall. The control sites, LM1 and LM6, experienced the least amount of change, rising only 2 and 5 cm respectively. There does not appear to be a correlation between nutrient concentrations and water input, particularly when comparing seasonal water level change and TP levels. However, a minor decrease in TN concentrations after the rainfall event in the old constructed wetlands, followed by an increase in TN concentration that reached the greatest values that summer was noted in the new Lower Missaguash Marsh wetlands

(Figure 3.12).

Figure 3.12. Comparing mean weekly water level change within CMR marshes (top) to TN and TP concentrations in new and old constructed wetlands. Yellow areas highlight the June 21-June 22 rainfall event.

45 Chapter 4: Discussion

The success of a constructed wetland is subjective to the application in which it is designed for—wetlands built throughout the CMR are often constructed to compensate for habitat loss and improve biological diversity (Austin-Smith, 1998; Brooks et al. 2005;

MacDonald and Clowater, 2007). Productivity, which is related to wetland senescence, is a limiting factor in improving diversification. However, in order to gain perspective on why a wetland is productive or senescent, one must consider nutrient regimes within the system. A high macronutrient concentration may improve productivity in one ecosystem niche, while inhibiting productivity in another (Zedler, 2000). Nutrient regimes are essential in characterizing wetland productivity, as nutrient sources, pathways, and depositional environments directly relate to the landscape in which they exist. Previous studies have shown that poorly designed and constructed wetlands are unlikely able to compensate for natural habitat loss, therefore, understanding the hydrological, geological, and biological influences on productivity is essential (Kansiime et al., 2007; Winter,

1999; Zedler, 2000). Research on the loading, pathways, deposition and trends of nitrogen and phosphorus in the CMR’s constructed wetlands has indicated that:

I. Phosphorus is likely autochtonous, as significant surface flow (through a major

rainfall event) did not increase measurable P concentrations.

II. Lack of variability of TN levels over the study period suggests that anthropogenic

sources of N are not the dominant contributor to TN concentrations in surface

water.

III. Underlying sediments and strata influence P availability in the CMR.

46 4.1 Trends: Nitrogen

The availability and cycling of N in wetlands is controlled by alteration of biological processes as the system ages (Craft, 1997). Nitrogen is readily released and retained in vegetation and through denitrifying bacteria due to its stable gaseous phase

(N2), which P lacks (Dunne and Reddy, 2005). Studies indicate that constructed wetlands are effective at nitrogen removal, and that TN removal rates within wetlands may average

513 mg Nm-2 day-1 (Bachand and Horne, 2000). A wetland with high TN removal rates could decrease a temporal nutrient source of TN availability in the water column.

A lack of variability in TN concentrations over the study period was noted in all surface water sites within the CMR (Figure 3.01). This may be attributed to the strong, year-round N regulation within the studied wetland systems. Small fluxes of N throughout the season may be the result of growth and decay of vegetation rather than external input, as additional N loading is often accompanied by a complementary removal rate increase (Craft, 1997). Additionally, there does not appear to be a significant variation in TN levels between the newly constructed wetlands (LM1—6) and established wetlands (old), with only slightly higher values being recorded in LM1—6 (Figure 3.12).

Research by Craft (1997) has shown that N retention (which decreases TN concentration in the water column) begins when emergent vegetation is established, and organic matter starts to accumulate in the soil. Emergent vegetation noted in the new Lower Missaguash sites may indicate that the N is being removed from the water column, thus reducing the variability of TN measurements in wetlands of differing ages.

Lack of TN variability may also indicate that N is not the limiting nutrient in the

CMR’s constructed wetlands, as N is more frequently limiting in brackish and saline

47 wetland environments (Taylor, 2005). Salinity values in the CMR wetlands ranged from

0.0—0.47 PPT which indicates that all sites were freshwater (Eilers et al., 1990). Higher conductivity values (seasonal average 0.26 and 0.32 PPT, respectively) in sites LM5 and

LM6 may be attributed to seawater penetration and/or the release of seawater salts from underlying salt marsh sediment (Table 3.06; Table 3.07).

4.2 Trends: Phosphorus

Unlike N, TP concentrations produced temporal trends and notable variability between constructed wetlands of different ages (Figure 3.01; Figure 3.12). In wetlands, P availability is regulated primarily by geochemical processes, such as sorption and precipitation. Most P in wetlands is chemically bound to clays and organics within the sediments, rather than existing in a soluble, aqueous form (Craft, 1997; Dunne and

Ruddy, 2005). Total P increased in all sites during study period (Figure 3.04; Figure 3.07;

Figure 3.08; Figure 3.10) with the newly constructed wetlands reaching hypereutrophic

(>0.10 mg/L TP) values (Environment Canada, 2016). The seasonal increase can be attributed to several internal processes, mainly those affecting P exchange at the water/lakebed sediment interface. For instance, oxygen fluctuation often drives P exchange at the sediment interface, and DO recordings indicate that there was sufficient oxygen within the CMR surface water to promote these fluctuations (Ruddy and Duddy,

2005). Wind driven currents can disturb P loaded sediments, especially in sites that are close to the coastline and lack significant tree cover (e.g., Lower Missaguash and Upper

Missaguash Marshes). Research indicates that internal loading, the process that causes geochemically bound P to release from their adsorption sites into the overlying water column, normally occurs late summer as oxygen levels drop (Environmental

48 Laboratories, 2016). This study is in agreement with findings from previous research, as

TP concentrations increased significantly between July 1—16 (week 6—7: Figure 3.08).

The higher TP concentrations and greater (seasonal) fluctuations recorded in the newly constructed wetlands can be attributed to several factors. Craft (1997) indicates that TP availability is generally higher (in suspension) in new (1-3 year old) wetlands due to higher sedimentation rates and lack of aquatic vegetation. Established vegetation, as noted in LM7 and UM1, provides sediment stability with intricate rooting systems, and stems and leaves also restrict hydraulic movement and shoreline erosion, thus reducing P availability. Phosphorus levels in newly constructed wetlands are generally initially high

(often hypertrophic status) and then become stabilized as emergent vegetation roots itself

(Craft, 1997; Dunne and Reddy, 2005; Zedler, 2000). Although all newly constructed

Lower Missaguash sites show these trends of seasonal fluctuation, there is still variability amongst the sites. Sites LM1 and LM6 had consistently lower initial and final TP concentrations; this may be attributed to the sites not being fully flooded which may have resulted in less TP fluctuation. As seen in Figure 3.08, LM1 and LM6 had nearby shoreline vegetation which would have been killed and may have released TP that may have contributed to the surface water concentrations had the sites reached their maximum water levels. Similar vegetation existed in sites LM2, LM3, LM4, and LM5 but was killed by water level rise. This decomposing vegetation within the sites/plots likely released P back into the water column as the plant material decomposed.

Turbidity is an indicator of trophic status within freshwater systems (Environment

Canada, 2016). It was noted in sites LM2—6 that as TP levels increased throughout the summer, water clarity and light penetration decreased. This suggests that increased P

49 concentrations may give rise to pelagic productivity as extant vegetation decays and releases nutrients. The relationship between these two qualities may indicate that P is a limiting nutrient in the CMR wetlands, as increased P seems to correlate with higher pelagic activity. Additionally, it is well documented that P is often the limiting nutrient in freshwater systems (Taylor, 2005).

4.3 Sources and Pathways: Nitrogen

Identifying N sources may be difficult without understanding the effectiveness of a wetland system’s ability to remove and retain nutrients. Total N values may not solely indicate if external N is loading into a system, as it has been noted that nutrient removal rates likely equal external loading rates, thus reducing seasonal variability (Craft, 1997;

Wilcock et al., 2011). Denitrification is the primary process in removing nitrate from overlying wetland waters and increases under anaerobic conditions (Bachand and Horne,

2000; Wilcock et al., 2011). Due to sufficient DO levels within the study sites, denitrification may be occurring at a slow rate, however more research is required to better understand conditions at the redox sediment/water interface.

Nitrogen often enters aquatic systems from runoff events by overland flow, or from rivers and streams flowing directly into wetland systems. The constructed wetlands in this study lacked these connecting streams, and research by Wilcock et al. (2011) suggests that this can result in a lower supply of N and other particulates. Low water conductivity in the old wetlands in the Missaguash Marshes suggests that water input at these sites is predominately by precipitation.

High detection in TN groundwater from the Upper Missaguash (Figure 3.03) can be attributed to nearby agricultural activity including fertilizer and manure applications.

50 Although significant anthropogenic N sources exist in the CMR, land use may not be an important nutrient contributor to surface water, as TN concentrations remained low throughout the study in all study sites, despite a major rainstorm event (Figure 3.11). The rainstorm event resulted in a 10% increase in volume in the sites, suggesting that overland flow and runoff may be increasing water levels in wetlands but are not contributing to N concentrations. Week 4 (the major rainfall event) and week 8 show an increase in mean water level, and a decrease in TN concentration. This is likely attributed to dilution of TN concentrations in the surface water, as the runoff does not contain a N source.

4.4 Sources and Pathways: Phosphorus

The main source of inorganic P is from the erosion of crustal rocks—a long-term process which relies on major tectonic or global events e.g., the uplift of a granitic pluton/glacial movement (Filippelli, 2008). The advent of phosphate loaded detergents and other human byproducts has increased P loading, and can now happen at a much faster rate (Dunne and Reddy, 2005; Filippelli, 2008; Wilcock et al., 2007; Zheng and

Stevenson, 2006). Although possible anthropogenic sources of P exist in the CMR, such as towns and agricultural fields, land use does not appear to be an important P contributor. Total P concentrations did not seem to be affected by the early summer rainfall event (Figure 3.12), which would have been sufficient to mobilize and increase the input of land-derived sources of TP into each wetland if anthropogenic P was a significant source (Filippelli, 2008). Due to the nature of P loading and retention, there is less variability associated with the biological cycling of P, as geochemical processes are the main factors in P loading, mobilization and retention; thus it can be concluded that

51 most P in the CMR wetlands is autochthonous (Craft, 1997).

Phosphorus availability among the CMR constructed wetlands may vary due to underlying strata and age. Phosphorus variability between old and new wetlands can be attributed to sediment stability, as increased stability in aging wetlands is related established (old) vegetation, with roots holding sediments together and stems preventing water disturbances that may release P into surface water. These differences in P concentrations may also be governed the capacity of a wetland system to remove water column P through physical, chemical and biological processes in wetlands sediments.

This process is regulated by various properties including: pH, redox potential, Al, Ca content of soils, organic matter content, P loading, hydraulic loading, and ambient P content of soils (Craft, 1997; Runne and Duddy, 2005). High conductivity values in site

A3 may be attributed to underlying gypsum (CaSO4· 2H2O) in Windsor Group strata

(Figure 1.2). Small, circular indentations noted along Layton’s Lake’s northwestern perimeter indicate possible gypsum sinkholes. These values coincide with the highest calcium (Ca) and sulfur (S) values recorded at A3 (120 and 82.8 mg/L, respectively).

When both soluble P and Ca2+ are available in sufficient concentrations within a system, insoluble phosphate minerals, such as calcium phosphate1 and hydroxyapatite have the potential to precipitate (Runne and Duddy, 2005). This geochemical process may explain the difference in TP levels between A3 (seasonal average: 0.061 mg/L), and adjacent sites A1 and A2 (seasonal average: 0.081 mg/L and 0.081 mg/L, respectively) where the higher Ca concentration may be increasing P retention.

2+ 2- 1. Ca + HPO4 → CaHPO4 [precipitate]

52 Aluminum (Al) and iron (Fe) are also elements that P commonly interacts with in forming insoluble mineral phosphates, it has been shown that wetlands with high amounts of Al or Fe loadings can retain high amounts of P. However, this does not seem to be a concern in the CMR, as values of Fe and Al are low (Craft, 1997; Dunne and

Reddy, 2005). Further analysis may be required to better understand P cycling.

Although the high conductivity values in A3 may be bedrock related, it should be noted that previous studies indicate that Layton’s Lake (which A3 receives inflow from) is ectogenically meromictic, having once received periodic seawater intrusions from the

Cumberland Basin resulting in salt within sediments that may release into overlying water (Howell and Kerekes, 1987). This suggests that even higher conductivity values may be recorded at greater depths that are below the saline-fresh interface.

4.5 Recommendations and Management Implications

As a result of the research conducted in this study, it is apparent that the construction or modification of wetlands needs to take into account the nutrient pathways and availability. Although it appears that anthropogenic sources are not an important contributor to nutrient loading in the wetland sites studied, local processes (erosion, water level) and physical properties (geology, slope) have a significant impact on nutrient availability. Constructing a wetland within sediments or strata that may contain higher concentrations of elements that have interaction potential with macronutrients may significantly influence wetland productivity. For example, the interplay between calcium and phosphorus binding and creating compounds that are not bioavailable. Additionally, understanding the behaviors of nutrients within a system—from initial construction to older age—is important before labelling a system as “unsuccessful” or senescent. If it is

53 deemed that a wetland is falling into senescence, and it is determined which nutrient regimes are affecting the productivity, physical alteration may be a viable option. For instance, manually disturbing wetland sediments is a possible productivity-increasing application, as disturbed sediments have high potential to reincorporate phosphorus back into the water column.

Future research by Loder (2016) will employ the paleolimnology method to reconstruct productivity levels in the CMR’s wetlands by analyzing wetland sediments proxies to infer changes and variability in the systems. Sedimentary chlorophyll α levels, carbon/nitrogen ratios and other elemental concentrations will be used to infer productivity on a temporal and spatial scale, which will further our understanding of wetland senescence.

54 Chapter 5: Conclusions

• The CMR comprises constructed coastal wetland systems that may decline in

productivity as they age.

• Productivity is directly related to nutrient regimes, and phosphorus and nitrogen are

essential macronutrients which exist within the CMR. Phosphorus appears to be the

limiting nutrient, and low salinity levels indicate that the constructed wetlands are

freshwater (with anomalously high levels attributed to sea water penetration or the

release seawater salts from underlying salt marsh sediment in select sites).

• Phosphorus is likely autochthonous, as TP concentrations did not increase after a

significant early summer rainfall event that was sufficient to mobilize any

allochthonous TP sources.

• Decreasing TP concentrations as a constructed wetland system ages might be related

to increased sediment and slope stability as P binds to clays and organics within the

soils, and/or plant/animal uptake.

• A lower overall TP concentration may be attributed to mineralization/deposition

events through chemical reactions with cations released from underlying sediments

and strata (e.g. calcium released from gypsum in the Windsor Group deposit).

• Nitrogen levels appear to be stable throughout the CMR’s constructed wetlands, as

low variability indicates land use is not a dominant contributor.

• Further research is required to understand knowledge gaps associated with wetland

senescence and nutrient cycling.

55 References

Alcock, F.J. 1949. Geological Map of the Maritime Provinces, New Brunswick, Nova Scotia and Prince Edward Island. Geological Survey of Canada, 910A.

ASA Analytics. 2016-01-05 (date accessed). http://rmbel.info/total-phosphorus/ Total Phosphorus.

Austin-Smith Jr., P. 1998. Tantramar dykeland wildlife habitat status report. Department of Natural Resources and Energy.

Austin-Smith Jr., P. and Bowes, S., 2000. Tantramar dykeland wildlife habitat strategy. Department of Natural Resources and Energy, March 2000.

Bachand, P.A.M. and Horne, A.J. 2000. Denitrification in constructed free-water surface wetlands: II. Effects of vegetation and temperature. Ecological Engineering, 14: 17-32.

Benito, G., Perez del Campo, P., Gutierrez-Elorza, M. and Sancho, C. 1995. Natural and human-induced sinkholes in gypsum terrain and associated environmental problems in NE Spain. Environmental Geology, 25: 156-164.

Brooks, R. P., Wardrop, D.H., Cole, C.A. and Campbell, D.A. 2005. Are we purveyors of wetland homogeneity? A model of degradation and restoration to improve wetland mitigation performance. Ecological Engineering, 24: 331-340.

Cook, B.J. and Hauer, R.F. 2007. Effects of Hydrologic Connectivity on Water Chemistry, Soils, and Vegetation Structure and Function in an Intermontane Depressional Wetland Landscape. Wetlands, 27: 719-738.

Craft, C.B. 1997. Dynamics of nitrogen and phosphorus retention during wetland ecosystem succession. Wetlands Ecology and Management, 4: 177-187.

Davis, D. and Browne, S. 1996. The Natural History of Nova Scotia, Volumes I & II. The Nova Scotia Museum and Nimbus Publishing, Halifax. Dunne, E.G. and Reddy, K.R. 2005. Phosphorus in biogeochemistry of wetlands in agricultural watersheds. Nutrient management is agricultural watersheds: A wetlands solution, 4: 105-129 Dunnington, D. 2011. Using Paleolimnological Methods to Track late Holocene Environmental Change at Long Lake, New Brunswick-Nova Scotia Border Region, Canada. BScH. Thesis, Acadia University, Wolfville, N.S.

Eilers, J. M., Sullivan, T. J. and Hurley, K. C. 1990. "The most dilute lake in the world?". Hydrobiologica, 199: 1–6.

56 Ells, R.W. 1887. A history of New Brunswick geology. Gazette printing company.

Englehardt, H. 2013. Lead accumulation in constructed open water wet ecosystems in the Border Marsh region of New Brunswick and Nova Scotia. BScH. Thesis, Acadia University, Wolfville, N.S.

Environment Canada. 2016-02-29 (date accessed). http://climate.weather.gc.ca/climateData/dailydata_e.html? timeframe=2&Prov=&StationID=42083&dlyRange=2003-10-29%7C2016-02-28 &cmdB2=Go&cmdB1=Go&Year=2015&Month=6&cmdB1=Go# Daily Data Report for June 2015.

Environment Canada. 2015-09-20 (date accessed). https://www.ec.gc.ca/ap-pa/default.asp?lang=En&n=989C474A-1 Network of Protected Areas.

Environment Canada. 2016-01-20 (date accessed). https://www.ec.gc.ca/eaudouce-freshwater/default.asp? lang=En&n=0A77A85E-1&printfullpage=true Phosphorus in Canada’s Ecosystems.

Environmental Laboratories, Inc. 2016-01-05 (date accessed). http://rmbel.info/total-phosphorus/ Total Phosphorus.

Filippelli, G.M. 2008. The Global Phosphorus Cycle: Past Present, and Future. Elements, 4: 89-95.

Green, D.F. 1985. Chignecto marshes: bird and hay county. Canadian Geographic, 101: 44-50.

Greenberg, D.A., Blanchard, W., Smith, B. and Barrow, E. 2012. Climate Change, Mean Sea Level, and High Tides in the Bay of Fundy. Atmosphere-Ocean, 50: 261-276

Gussow, W.C. 1953. Carboniferous stratigraphy and structural geology of New Brunswick, Canada. American Association of Petroleum Geologists Bulletin, 37: 1713-1816.

Hach. 2016-03-22 (date accessed). http://ca.hach.com/chemistries-reagents-and- standards/family-chemistry?productCategory=22218734032 Chemistries, Reagents and Standards.

Heath, R.C. 1983. Basic ground-water hydrology: U.S. Geological Survey Water-Supply Paper 2220, 86 p.

Howell, G.D. and Kerekes, J.J. 1987. Primary Production of Two Small Lakes in Atlantic Canada. Proc. NS. INST. SCI., 37: 71-88.

Kadlec, R. H. 1999. Transport of Phosphorus in Wetlands, in O’Connor, G.A., Reddy, K. R., Schelske, C. L., Phosphorus Biogeochemistry in Subtropical Ecosystems: Lewis Publishers, p. 355-375.

Kansiime, F., Kateyo E., Mucunguzi, P. and Oryem-Origa, H. 2007. Nutrient Status and Retention in Pristine and Disturbed Wetlands in Uganda: Management Implications. Wetlands Ecology Management, 15: 453-476.

Keppie, J.D. 2000. Geological Map of the Province of Nova Scotia. Nova Scotia Department of Natural Resources Minerals an Energy Branch, ME 2000-1 (scale 1:500 000).

Lockbay, B.G. and Conner, W.H. 1999. N:P Balance in Wetland Forests: Productivity across a Biogeochemical Continuum. The Botanical Review, 65: 171-185.

Loder, A.L. 2014. Examination of metals in gastropods to determine the potential for accumulation in the Border Marsh Region. BScH. Thesis, Acadia University, Wolfville, N.S

Loder, A.L., Mallory, M.L., Spooner, I., McLauchlan, C., Englehardt, P., McLellan, N. and White, C. 2016. Bioacculumation of Lead and Arsenic in Gastropods Inhabiting Salt Marsh Pons in Coastal Bay of Fundy, Canada. Water Air Soil Pollution. 227:75: 1-12.

MacDonald, A. and Clowater, R. 2007. Natural ecosystem connectivity across the Chignecto Isthmus – opportunities and challenges. Canadian Parks and Wilderness Society, New Brunswick and Nova Scotia. Maillet, J.L., MacKinnon, C.M., and Pollard, J.B. 1999. Re-examination of the relationship between constructed impoundment age and water-bird use. Canadian Wildlife Service, Environment Conservation Branch, Technical Report Series Number 290.

Michener, W.K., Blood, E.R., Bildstein, K.L., Brinson, M.M. and Gardner, L.R. 1997. Climate change, hurricanes and tropical storms, and rising sea level in coastal wetlands. Ecological Applications, 7: 770–801.

Mitsch, W.J., Cronk, J.K., Wu, X., Nairn, R.W. and Hey, D.L. 1995. Phosphorus Retention in Constructed Freshwater Riparian Marshes. Ecological Applications, 5: 830-854.

Mitsch, W.J., and Gosselink, J.G. 2015. Wetlands, 5th Edition. Edited by John Wiley & Sons Inc., New Jersey.

58 Monbet, Y. 1992. Control of Phytoplankton Biomass in Estuaries: A Comparative Analysis of Microtidal and Macrotidal Estuaries. Estuaries, 15: 563-571.

Payne, N.F. 1992. Techniques for wildlife habitat management of wetlands. College of Natural Resources, University of Wisconsin. McGraw-Hill Inc., pp. 11–128.

Richardson, C.J. 1985. Controlling Phosphorus Retention Capacity in Fresh water Wetlands. Science, New Series, 228: 1424-1427.

Rosar, E.C. 1993. Method and Apparatus for Solution Mining. United States Patent 5,246,273.

Schlesginer, W.H. 2009. On the fate of anthropogenic nitrogen. Proceedings of the National Academy of Sciences of the United States of America, 106: 203-208.

Shaw, J., Amos C.L., Greenburg, D.A., O’Reilly, C.T., Parrott, D.R. and Patton, E. Catastrophic tidal expansion in the Bay of Fundy, Canada. 2010. Canadian Journal of Earth Science, 47: 1079-1091.

Shimadzu. 2016-01-20 (date accessed). http://www.shimadzu.com/an/ftir/support/tips/letter6/aimtech.html Microscope Measurement Techniques.

Stea, R.R., Piper, D.J.W., Fader, G.B.J. and Boyd, R. 1998. Wisconsinan glacial and sea-level history of Maritime Canada and the adjacent continental shelf: A correlation of land and sea events. Geological Society of America Bulletin, 110: 821-845.

Taylor, D. 2005. A Water Quality Assessment of Blair Lake, Cumberland County, Nova Scotia. Nova Scotia Department of Environment & Labour.

Trueman, G.J. 1899. The marsh and lake region at the head of the Chignecto Bay. Bulletin of the Natural History Society of New Brunswick, 14: 93-104.

U.S. EPA. 2002. Methods for Evaluating Wetland Condition: Wetlands Classification. Office of Water, U.S. Environmental Protection Agency, Washington, DC, 43 p.

The Weather Man. 2016-02-29 (date accessed). https://awd1970.wordpress.com/tag/sackville/ June 2015 - Cool and Wet.

White, H. 2012. Paleolimnological Records of Post-Glacial Lake and Wetland Evolution from the Isthmus of Chignecto Region, Eastern Canada. M.Sc. Thesis, Acadia University, Wolfville, N.S.

59 Wilcock, R.J., Müller, K., van Assema, G.B., Bellingham, M.A. and Ovenden, R. 2011. Attenuation of Nitrogen, Phosphorus and E.Coli Inputs from Pasture Runoff to Surface Waters by a Farm Wetland: the Importance of Wetland Shape and Residence Time. Water Air Soil Pollution, 223: 499-509.

Winter, T.C. 1999. Relation of streams, lakes, and wetlands to groundwater flow systems Hydrogeology Journal, 7: 28-45.

Wu, X., & Mitsch, W. J. 1998. Spatial and temporal patterns of algae in newly constructed freshwater wetlands. Wetlands, 18: 9–20.

Zampella, R.A. and Laidig, K.J. 2003. Functional Equivalency of Naturaland Excavated Coastal Plain Ponds. Wetlands, 23: 860-876.

Zedler, J.B. 2000. Progress in Wetland Restoration Ecology. Trends in Ecology & Evolution, 15: 402-407.

Zedler, J.B. and Callaway, J.C. 1999. Tracking Wetland Restoration: Do Mitigation Sites Follow Desired Trajectories? Restoration Ecology, 7: 69-73.

Zhao, H., Cui, B., Zhang, H., Fan, X., Zhang, Z., & Lei, X. 2010. A landscape approach for wetland change detection (1979-2009) in the Pearl River Estuary. Procedia Environmental Sciences, 2: 1265–1278.

Zheng, J. and Stevenson, R. J. 2006. Algal assemblages in multiple habitats of restored and extant wetlands. Hydrobiologia, 561: 221-238.