Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Examining selected soil properties on the lower Fraser River delta following four-year grassland set-aside Porter, Maria Teresa 2021

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
24-ubc_2021_may_porter_maria.pdf [ 3.85MB ]
Metadata
JSON: 24-1.0395492.json
JSON-LD: 24-1.0395492-ld.json
RDF/XML (Pretty): 24-1.0395492-rdf.xml
RDF/JSON: 24-1.0395492-rdf.json
Turtle: 24-1.0395492-turtle.txt
N-Triples: 24-1.0395492-rdf-ntriples.txt
Original Record: 24-1.0395492-source.json
Full Text
24-1.0395492-fulltext.txt
Citation
24-1.0395492.ris

Full Text

EXAMINING SELECTED SOIL PROPERTIES ON THE LOWER FRASER RIVER DELTA FOLLOWING FOUR-YEAR GRASSLAND SET-ASIDE by  Maria Teresa Porter  B.Sc., The University of British Columbia, 2014  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Soil Science)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  January 2021  © Maria Teresa Porter, 2021   ii  The following individuals certify that they have read, and recommend to the Faculty of Graduate and Postdoctoral Studies for acceptance, a thesis/dissertation entitled:  Examining soil physical properties in the Fraser River delta following four-year grassland set-aside   submitted by Mara Teresa Porter in partial fulfillment of the requirements for the degree of Master of Science in Soil Science  Examining Committee: Dr. Maja Krzic, Associate Professor, Faculty of Land and Food Systems & Faculty of Forestry Supervisor  Dr. Sean Smukler, Associate Professor, Faculty of Land and Food Systems Supervisory Committee Member  Dr. Gary Bradfield, Professor Emeritus, Faculty of Science Supervisory Committee Member Dr. Art Bomke, Professor Emeritus, Faculty of Land and Food Systems Additional Examiner   iii  Abstract Intensive cultivation, combined with poor drainage and heavy precipitation, has led to soil degradation in the lower Fraser River delta region (LFRD) of British Columbia, an important agricultural area of the province. The Delta Farmland & Wildlife Trust (DF&WT), a non-profit organization in the LFRD, partners with farmers to implement conservation practices such as grassland set-asides (GLSA). Farmers seed fields with grasses and clover and leave them to rest for one to four years. GLSA programs have been used worldwide, but program goals, GLSA duration, and the geographical contexts of these programs vary widely. The objectives of this study were to evaluate the effects of three- and four-year GLSAs on selected soil properties in the LFRD. Samples were collected from seven operating farm sites in 2018, and five in 2019. Sites were classified as productive or unproductive, and each comprised of a GLSA seeded in 2015 and an adjacent field under annual crop rotation (ACR). Previously collected data from 2015 and 2016 was included in the analyses. At depth 0–7.5 cm, GLSA had greater aggregate stability than ACR at productive and unproductive sites. GLSA had significantly lower bulk density and higher aeration porosity than ACR at depth 0–7.5 cm after the three years, but not after four years. The improvements to aggregate stability under GLSA were observed in 2015 and 2016 and were maintained thereafter. In 2018 and 2019, at depth 0–7.5 cm, aeration porosity decreased at ACR but not at GLSA. Total carbon and total nitrogen were significantly higher in GLSA than ACR after four years only at depth 30–45 cm. GLSA did not yield significant improvements in soil properties at depths below 7.5 cm, nor did they fully restore compacted soils. Still, GLSA remain a useful management practice for farmers in the LFRD and can improve soil quality in the first year of GLSA while protecting and maintaining soil structure from further degradation in the third and fourth years.  iv  Lay Summary The lower Fraser River delta region (LFRD) is an important agricultural area for British Columbia; however, intensive cultivation, poor drainage, and heavy winter precipitation have caused the soil degradation in this region. The Delta Farmland & Wildlife Trust (DF&WT), a non-profit organization, works with farmers to establish grassland set-asides (GLSAs). These are fields that have been removed from agricultural production, seeded with grass and clover, and left to rest for one to four years. The objective of this study was to evaluate the effects of GLSA on soil quality. I found that GLSAs have better soil structure compared to cropped fields in the years three and four, and that improvements from GLSA take place in year one. These results demonstrate the benefits of GLSA and may assist local farmers in selecting fields for GLSA and length of GLSA to support the preservation of soil quality in the LFRD. v  Preface This thesis represents unpublished work which I conducted with assistance from undergraduate students and advisors. I was the lead investigator in the study and was responsible for all major areas of research question formation, 2018 and 2019 data collection, data analysis and thesis composition. Field selection for this project was led by Dru Yates and Christine Terpsma. Jason Lussier provided data from 2015 and 2016. Alexandra Mullins, Alexander Kramer, Patricia Hanuszak, and Meghan McIllfaterick provided field assistance in sample collection. Alexander Kramer, Patricia Hanuszak, Skylar Kylstra, and Phyllis Fang assisted in laboratory processing of samples. Several other volunteers also provided assistance in sample collection and laboratory processing. Jason Lussier, Lewis Fausak, and Chantel Chizen, provided support with field and laboratory protocols included in this study. The Sustainable Agricultural Landscape lab coordinators, Katie Neufeld, Paula Porto, and Carson Li provided support with field and laboratory equipment. Dr. Maja Krzic was the supervisory author on this project and was closely involved in all aspects of the study. The study was conducted in collaboration with Dr. Sean Smukler and Drew Bondar, who were closely involved in project development. Dr. Sean Smukler and Dr. Gary Bradfield provided project guidance and thesis edits. Dr. Art Bomke also provided thesis edits. vi  Table of Contents  Abstract ......................................................................................................................................... iii Lay Summary ................................................................................................................................ iv Preface ............................................................................................................................................ v Table of Contents .......................................................................................................................... vi List of Tables ................................................................................................................................. ix List of Figures ................................................................................................................................ x List of Abbreviations ................................................................................................................... xii Acknowledgements ..................................................................................................................... xiii Chapter 1: General Introduction ................................................................................................. 1 1.1 Agriculture and Land Issues in the Fraser River Delta Region ....................................... 1 1.2 Conservation Practices and the Delta Farmland and Wildlife Trust ............................... 4 1.3 Grassland Set-asides ........................................................................................................ 5 1.3.1 Grassland Set-asides in Canada and British Columbia ........................................... 7 1.3.2 Grassland Set-Aside Research: Worldwide ............................................................. 9 1.3.3 Grassland Set-Aside Research: Lower Fraser River delta, BC ............................. 11 1.4 Study Objectives and Hypotheses ................................................................................. 15 Chapter 2: Materials and Methods ............................................................................................ 17 2.1 Study Sites ..................................................................................................................... 17 Chapter 3: Results and Discussion ............................................................................................. 24 3.1 Comparing GLSA to ACR ............................................................................................ 27 3.1.1 Aggregate Stability ................................................................................................ 27 vii  3.1.2 Soil Bulk Density .................................................................................................. 31 3.1.3 Aeration Porosity ................................................................................................... 33 3.1.4 Total Soil Carbon and Nitrogen ............................................................................ 37 3.1.5 Conclusions of GLSA and ACR comparison ........................................................ 39 3.2 Changes in Soil Properties Over Four Years of GLSA Relative to Baseline Soil Conditions .................................................................................................................................. 40 3.2.1 Aggregate Stability ................................................................................................ 40 3.2.2 Soil Bulk Density .................................................................................................. 45 3.2.3 Aeration Porosity ................................................................................................... 48 3.2.4 Total Soil Carbon and Nitrogen ............................................................................ 53 3.2.5 Conclusions of Changes in Soil Properties Over Four Years of GLSA Relative to Baseline Soil Conditions ....................................................................................................... 53 Chapter 4: General Conclusions and Recommendations for Future Research .................... 56 4.1 General Conclusions ...................................................................................................... 56 4.2 Recommendations for Future Research ......................................................................... 59 References .................................................................................................................................... 61 Appendices ................................................................................................................................... 68 Appendix A Crop history on eight study sites. Site 7 was not part of my study but was included in the larger study initiated in 2015 and is included here. PT=Potatoes; P= Peas; GLSA=Grassland set-aside; B= Barley; W=Wheat; BN= Beans; C= Corn; T= Turnips, F=Fallow. ................................................................................................................................... 68 Appendix B Soil baseline properties of fields entering the larger study in spring 2015. All properties were determined at the 0-15 cm depth with the exception of mean weight diameter viii  (MWD) and bulk density (determined at 0-7.5 cm depth), and Na+ (determined at 0–30 cm depth). Source: Lussier 2018. .................................................................................................... 69 Appendix C Percentage of total dry biomass by vegetation group (grass, clover, and weeds/other) in productive and unproductive grassland set-asides (GLSA) at each study site.70 Appendix D Average temperature, rainfall, snowfall, and precipitation in the LFRD from September 2014 to April 2019. Data is from the Delta - Tsawwassen Beach weather station (Environment and Climate Change Canada no date a). ............................................................. 71 Appendix E Total carbon at productive annual crop rotation (ACR) and grassland set-aside (GLSA) fields in spring 2015 and 2019 at 0–15, 15–30, 30–45, and 45–60 cm depths. No significant differences were found (α = 0.10). .......................................................................... 73 Appendix F Total nitrogen at productive annual crop rotation (ACR) and grassland set-aside (GLSA) fields in spring 2015 and 2019 at 0–15, 15–30, 30–45, and 45–60 cm depths. No significant differences were found (α = 0.10). .......................................................................... 74 Appendix G C:N ratio at productive annual crop rotation (ACR) and grassland set-aside (GLSA) fields in spring 2015 and 2019 at 0–15, 15–30, 30–45, and 45–60 cm depths. No significant differences were found (α = 0.10). .......................................................................... 75 ix  List of Tables Table 3.1 Mean weight diameter (MWD) and proportion of total soil in four aggregate size classes in spring, summer and fall of 2018 and spring 2019 for productive and unproductive ACR and GLSA fields. The standard error of the mean is shown in brackets. Different lowercase letters indicate a significant difference due to treatment (i.e., ACR vs. GLSA at each sampling time). Values followed by different letters are significantly different at p<0.10. ......................... 30 Table 3.2 Soil aeration porosity and bulk density for productive and unproductive ACR and GLSA fields. The standard error of the mean is shown in brackets. Different lowercase letters indicate a significant difference due to treatment (i.e., ACR vs. GLSA at each sampling time) (α = 0.10), while uppercase letters indicate significant differences between treatment × site type (α = 0.10) where the interaction was significant. Letters cannot be compared between the different times or depths. .............................................................................................................................. 36 Table 3.3 Mean total carbon, total nitrogen, and C:N ratio at productive (n=3) and unproductive (n=2) ACR and GLSA fields in spring 2019. The standard error of the mean is shown in brackets. Lowercase letters indicate significant differences due to treatment when there was no interaction between treatment and site type while uppercase letters indicate significant differences between treatment × site type (α = 0.10). ................................................................... 39  x  List of Figures  Figure 2.1 Overview of management history and soil characteristics of seven study sites located at operational farms in the lower Fraser River Delta (LFRD), British Columbia. ........................ 23 Figure 3.2 Mean weight diameter of water stable aggregates (MWD) at productive and unproductive annual crop rotation (ACR) and grassland set-aside (GLSA) fields in spring 2015, 2016, 2018, and 2019. Different letters denote significant differences based on pairwise contrasts of estimated marginal means using Bonferroni’s adjustment at p <0.05. ..................................... 43 Figure 3.3 Fraction of total soil in the 2–6 mm aggregate size class at productive and unproductive annual crop rotation (ACR) and grassland set-aside (GLSA) fields in spring 2015, 2016, 2018, and 2019. Different letters denote significant differences based on pairwise contrasts of estimated marginal means using Bonferroni’s adjustment at p <0.05. ..................................... 44 Figure 3.4 Fraction of total soil in the <0.25 mm aggregate size class annual crop rotation (ACR) and grassland set-aside (GLSA) fields in spring 2015, 2016, 2018, and 2019. Different letters denote significant differences by year (averaged across site type) based on pairwise contrasts of estimated marginal means using Bonferroni’s adjustment at p <0.05. .......................................... 45 Figure 3.5 Mean bulk density at annual crop rotation (ACR) and grassland set-aside (GLSA) fields in spring 2015, 2016, 2018, and 2019. Different letters denote significant differences by year (averaged across site type) based on pairwise contrasts of estimated marginal means using Bonferroni’s adjustment at p <0.05. .............................................................................................. 48 Figure 3.6 Mean aeration porosity at productive and unproductive annual crop rotation (ACR) and grassland set-aside (GLSA) fields at depth 0–7.5 cm in spring 2016, 2018, and 2019. xi  Different letters denote significant differences based on pairwise contrasts of estimated marginal means using Bonferroni’s adjustment at p <0.05. ......................................................................... 50 Figure 3.7 Mean aeration porosity at productive annual crop rotation (ACR) and grassland set-aside (GLSA) fields at depth 7.5–15 cm in spring 2016, 2018, and 2019. Different letters denote significant differences based on pairwise contrasts of estimated marginal means using Bonferroni’s adjustment at p <0.05. .............................................................................................. 51 Figure 3.8 Mean aeration porosity at unproductive annual crop rotation (ACR) and grassland set-aside (GLSA) fields at depth 7.5–15 cm in spring 2016, 2018, and 2019. Different letters denote significant differences by year (averaged across treatment) based on pairwise contrasts of estimated marginal means using Bonferroni’s adjustment at p <0.05. .......................................... 51 Figure 3.9 Mean aeration porosity at productive annual crop rotation (ACR) and grassland set-aside (GLSA) fields at depth 15–30 cm in spring 2016, 2018, and 2019. Different letters denote significant differences based on pairwise contrasts of estimated marginal means using Bonferroni’s adjustment at p <0.05. .............................................................................................. 52 Figure 3.10  Mean aeration porosity at unproductive annual crop rotation (ACR) and grassland set-aside (GLSA) fields at depth 15–30 cm in spring 2016, 2018, and 2019. Different letters denote significant differences by year (averaged across treatment) based on pairwise contrasts of estimated marginal means using Bonferroni’s adjustment at p <0.05. .......................................... 52  xii  List of Abbreviations  ACR Annual crop rotations BC British Columbia BBCC Boundary Bay Conservation Committee  CAP Common Agricultural Policy CCP Conservation Cover Program CRP Conservation Reserve Program DFI Delta Farmers Institute DF&WT Delta Farmland and Wildlife Trust GLSA Grassland set-aside LFRD Lower Fraser River delta MWD Mean weight diameter PCP  Permanent Cover Program USA United States of America USDA United States Department of Agriculture  xiii  Acknowledgements This research project took place on the traditional and ancestral territories of the Kwantlen Nation and Stó:lō people, the šxʷməθkʷəy̓əmaɁɬ təməxʷ (Musqueam) First Nation, Stz’uminus, and sc̓əwaθenaɁɬ təməxʷ Tsawwassen First Nations.  A treaty agreement was reached between the Tsawwassen First Nation, the government of Canada, and the government of British Columbia, effective in 2009. While the farms in my study are not located within the treaty lands, some are located within an area where the Tsawwassen First Nation shall receive first right of refusal. Their traditional territory spans around 10,000 km2 (or one million hectares). Through the treaty they regained sovereignty over roughly 0.07% (724 hectares) of this territory. There have never been any other treaties on these lands with any other nation. My study deals with soil degradation, which was not an issue in the region prior to colonization. I wish freedom, justice, rights, and restitution to all First Nations, Métis, and Inuit peoples.  To Dr. Maja Krzic, thank you for opening my undergraduate mind to the wonder of soils, and for being my supervisor. You have been abundantly patient and encouraging. Your unwavering dedication to teaching is evident, and I am lucky to count myself among your students. Thank you for leading with kindness and confidence, and for being a mentor and role model far beyond the scope of this study.  Thanks to my advisory committee for your guidance. Dr. Sean Smukler, thank you for the close look you gave my tables and figures, and for furthering my recommendations. Dr. Gary Bradfield, thank you for lending your thoroughness and keen eye for diction and syntax to this work. Dr. Art Bomke, thank you for your generosity and interest in this study from the beginning. To have you as my external examiner was a privilege; it felt like a lottery win. xiv  To Drew Bondar and the DF&WT, thank you for including me in this collaboration. Thanks also to the farmers. Your very involvement in this project and the GLSA Stewardship Program demonstrates your care for the land and soil.  Thanks to Dr. Valerie Lemay, for your generous and kind guidance as I trudged through my statistical analyses. Any depth to my understanding of statistics I owe to you, and any statistical silliness that persists in this document is my own doing.  Jason Lussier, thank you for taking me out to see the sites, and for helping me out in countless ways. Thanks to Alex Kramer and Skylar Kylstra, for your dedicated field and lab assistance on this project as work-learn students and to Trish Hanuszak for the contributions you made through your directed study. Thanks as well to the many people who volunteered with me, including Alexandra Mullins, Phyllis Fang, and Meghan McIllfaterick. To my colleagues, thank you for showing me the ropes in the lab and for your encouragement and solidarity. Most of all, and unexpectedly, thank you for your friendship. May it endure far beyond our degrees.  Jessica Poon, my friend and editor, thank you. Your criticism is unparalleled in its helpfulness and entertainment value. To all my friends, thank you for reminding me what life is all about and for saving the day on many occasions. To my family thank you for your love, confidence, support, and good company throughout my entire life. And to Tatenda, thank you for being here with me.  This research was supported by the Delta Farmland and Wildlife Trust, the Mitacs Accelerate Program, the British Columbia Investment Agriculture Foundation, and the Natural Sciences and Engineering Research Council of Canada.  1  Chapter 1: General Introduction 1.1 Agriculture and Land Issues in the Fraser River Delta Region The lower Fraser River delta region (LFRD) of British Columbia (BC) has some of the province’s best agricultural land due to a combination of the mild and temperate climate, flat topography, relatively long growing season, and medium-textured, fertile soils. In 2016, agriculture in the municipalities of Delta and Richmond within the LFRD accounted for 7.5% of the province’s total gross farm receipts (Statistics Canada 2016a). Agriculture in the area includes 9,175 ha of field production—0.47% total field production in BC (Statistics Canada 2016b). The main crops are (by area) 27% berries, 24% forage and hay, 15% potatoes, 15% other field vegetables, and 7% silage corn (Statistics Canada 2016c and d).  Although soils in the LFRD are quite fertile, they are also characterized by poor natural drainage (Bertrand et al. 1991) due to a high water table and flat topography. The issue of poor natural drainage is compounded further by salinity. Farmers irrigate using water from the Fraser River, which can be influenced by the salt wedge entering from the Salish Sea (Thiel et al. 2015). Because sodium can cause the dispersion of soil aggregates and prevent flocculation (Litalien and Zeeb 2020), this ocean water intrusion plays a role in soil structural degradation. Furthermore, soluble salts that move towards the soil surface with the rising water table may get trapped in the upper layers of the soil profile due to drainage issues and surface sealing (the formation of a dense layer of clay particles at the soil surface as a result of the disintegration of aggregates). In response to a 2015 survey, 76% of farmers (n=17) in LFRD, reported that salinity was an issue on their farms (Thiel et al. 2015). When soil structure is degraded, fields take longer to dry out in the spring, which presents a dilemma for farmers. They must either delay planting 2  and subsequent harvest or cultivate wet soils (Hermawan and Bomke 1996). Neither option is desirable.  All of these problems combined with heavy precipitation during fall and winter make the soils of the LFRD highly susceptible to degradation. The rainfall from October through April is around 730 mm which represents 78% out of 930 mm of annual precipitation (Environment and Climate Change Canada no date a). Raindrop impact damages soil structure and can cause surface sealing, especially when fields are left bare over winter (Krzic et al. 2000). Soil structural degradation exacerbates infiltration and drainage issues, causing surface ponding, which in turn increases degradation caused by farm traffic and tillage (Paul and de Vries 1979; Hermawan and Bomke 1996). Indeed, soil degradation has been observed in the region; 76% of farmers reported that drainage was an issue on their farms, primarily in the fall and winter (Thiel et al. 2015). For decades, a variety of socioeconomic factors—insecure land tenure and market shifts, for instance—have pushed farming in the LFRD towards highly intensive annual crop rotations and lowered incentives to manage for long-term soil conservation (Fraser 2004). In Richmond, agriculture was mostly mixed vegetable and small fruit production until the 1950s, after which farms transitioned to industrial agriculture and monocropping (Steves 2007). Until the 1960s, agriculture in Delta included grazing cattle and horses (for logging operations) on grass pasture (DF&WT 2017a). Cattle and dairy moved to the eastern part of the Fraser Valley and as cash cropping became dominant, grasses were largely removed from crop rotation (DF&WT 2017a). By the 1980s, only 10% of agricultural land in Delta was in forage production (DF&WT 2017a), while by the early 1990s, over 50% of farmland in the region was used for potato, bean, corn, and pea production.  3  In 1995, the Fraser Valley fruit and vegetable processing plant closed, while potato production increased as process markets for potatoes were unaffected, leading to shorter crop rotations in the area (Bomke et al. 1996). More recently, farmers have reported growing potatoes on a one- to five-year rotation with other crops, mainly peas, beans, corn, wheat, barley, and other vegetables (Thiel et al. 2015). While rotations under four years were once considered short for potatoes (Vos et al. 1989), rotations beyond that length of time are at present rare in the LFRD. Since potatoes take roughly as much space (15% of total field production) as all other field vegetables combined (Statistics Canada 2016d), it is likely that most rotations are on the shorter side of the one- to five-year range. Potato quality is judged strongly by potato shape; hence, farmers perform numerous tillage operations to achieve uniform soil tilth and promote round potato development. Thus, potato production requires more intensive cultivation than other crops. Additional cultivation is often taxing on soil structure (Bertrand et al. 1991).  Development and lease structures have also put pressure on farmland in the LFRD. In 1956, the Richmond Council re-zoned 5,059 ha of farmland to industrial and residential use—about 48% of total farmland in Richmond at the time (Steves 2007). In 1968, the province of BC expropriated 1,641 ha of land to create an industrial park for the Roberts Bank terminal of the Port of Vancouver (Western Producer 1994; Fraser 2004). This project was not completed, and the province continued to lease out this land on one-year leases that could be cancelled with a notice of 90 days (Fraser 2004), leaving farmers in extremely precarious financial positions. In the early 1990s, the leases changed to 20-year terms. By then, a quarter century of short-term leases had contributed to a legacy of soil degradation. A proposed expansion of the Roberts Bank of the Port of Vancouver in Delta, consisting of a new container terminal to be built over subtidal waters, is currently under review (Vancouver Fraser Port Authority 2018). In 2016, about 40% of 4  the total farm area in Delta and Richmond was leased or rented (Statistics Canada 2016e). Fraser (2004) found that in Delta, long-term leases did not provide the same incentive for implementing conservation practices as land ownership.  1.2 Conservation Practices and the Delta Farmland and Wildlife Trust Soil conservation practices carried out by farmers in the region include sub-surface drainage installation, surface laser levelling, cover cropping, and grassland set-aside (GLSA) use. Drainage installation reduces the length of time a soil is saturated, which has been linked to improved aggregate stability (Hermawan and Bomke 1996) and can also alleviate salinity issues. While less effective than sub-surface drainage, laser-levelling can prevent water from pooling in low-lying areas (Thiel et al. 2015). Due to the costs associated with laser-levelling and drainage installation, land tenure is a barrier to the implementation of these practices by farmers who do not own the land they farm (Fraser 2004). Cover cropping over the winter provides physical protection of soils from raindrop impact and can enhance soil structure and nutrient cycling (Hermawan and Bomke 1997; Liu et al. 2005). Soil degradation can lead to a delayed harvest and may eclipse the window during which cover crops can be seeded. Meanwhile, GLSAs have been implemented in the area since 1994, and participation in this practice has been equally adopted by land-owner and tenant farmers (Fraser 2004).   Support for farmers implementing GLSAs in the LFRD is offered through the Delta Farmland and Wildlife Trust (DF&WT). The DF&WT is a non-profit organization that works with farmers to promote the annual preservation of over 3,500 acres (1,400 ha) of farmland and wildlife habitat on the LFRD through co-operative land stewardship programs (DF&WT 2017b). It was established in 1993 by the Boundary Bay Conservation Committee (BBCC) and the Delta 5  Farmers Institute (DFI). The BBCC, established in 1988, is an organization that seeks to promote public awareness of the importance of the Fraser River Estuary ecosystem (BBCC 2008). The DFI, founded in 1898, is a producer group focused on education and advocacy by and for farmers in Delta (Kennedy 2017). Recognizing their common goals—preserving wildlife habitat and farmland—they formed the DF&WT, whose board is made up of equal representation from the two groups. The DF&WT was founded based on the recognition that farmland can provide habitat for wildlife that urban and industrial development cannot, and that the preservation of farmland can benefit both the farming community and wildlife. Funding for the DF&WT comes primarily from the YVR Stewardship Fund. This perpetual endowment fund was provided by Environment Canada to offset the 179 ha of wildlife habitat lost with the expansion of the Vancouver International Airport on Sea Island in the early 1990s (Principe 2001). It is held in trust for the DF&WT by the Vancouver Foundation. The DF&WT offers several programs to support farmers in the Municipalities of Delta and Richmond in implementing management practices such as laser levelling, winter cover cropping, and hedgerow, grass margin and GLSA establishment.  1.3 Grassland Set-asides A set-aside is a field removed from agricultural production (Clarke 1992), while a GLSA is a field removed from production, seeded with grasses, and left to rest. Typically, what might distinguish a GLSA from a covered ley or fallow field is the payment the farmer receives for establishing a GLSA, or the long duration of the set-aside. The reasons for GLSA payments range from providing ecosystem services and soil conservation to lowering crop production as an economic measure.  6  The 1980s saw the introduction of GLSA programs in North America and Europe in attempts to reduce agricultural surpluses and conserve biodiversity (Clarke 1992; Baer et al. 2000). In the United States of America (USA), the United States Department of Agriculture (USDA) Conservation Reserve Program (CRP) was created with the goal of addressing soil erosion (Dunn et al. 1993, Osborn 1993). Lands entered in the CRP were and continue to be in set-aside on 10- or 15-year contracts (USDA 2019). While the majority of these are grasslands, also included in the CRP are treed areas, riparian buffers, and wetlands (Osborn 1993). As of December 2019, 8.9 million hectares were in the CRP (USDA 2019). The European Commission’s Common Agricultural Policy (CAP) was enacted to decrease cereal surplus (European Commission 2004). This program ran from 1988 to 2008 and included both rotational GLSAs and non-rotational GLSAs, the latter of which had a minimum duration of five years (European Commission 2004). In 1993, CAP GLSAs became mandatory for farmers receiving EU subsidies (Tscharntke et al. 2011). At first, the majority of CAP set-asides were left to natural succession, but by the mid-1990s, farmers increasingly began sowing these with cover crops and even energy crops (Tscharntke et al. 2011). Before the CAP was abolished, around 5-15% of arable land was in GLSA (Tscharntke et al. 2011). Switzerland maintained the obligation that farmers keep 7% of their land in set-aside as Ecological Conservation Areas (Tscharntke et al. 2011) in order to receive subsidies (Aviron et al. 2009). These are primarily (78%) meadows (Albrecht et al. 2007) and to a lesser degree, consist of wildflower-strip fallows (Aviron et al. 2009).  7  1.3.1 Grassland Set-asides in Canada and British Columbia GLSA programs in Canada were also introduced in the 1980s. Agriculture and Agri-food Canada ran the Permanent Cover Program (PCP) in Alberta, Saskatchewan, Manitoba, and the Peace River Region of BC. Between 1988 and 1993, fields were set aside on 10- or 21-year contracts for the purpose of addressing soil degradation on agricultural lands (Bregha and Moffet 1995). Ontario ran its own Provincial Land Management Assistance Program, which signed 5-, 10-, and 15-year contracts from 1992 to 1994 to establish grass and tree cover to protect soils. Over 90% of these were 15-year agreements (National Soil Conservation Program no date). Another provincial set-aside program, the Conservation Cover Program (CCP), ran from 2001 to 2003 in Saskatchewan. Over 200,000 ha were converted to perennial cover through the CCP (McNeil 2013). From 2003 to 2008, Agriculture and Agri-food Canada signed 10-year set-aside contracts through the Greencover Canada program. Through this program, farmers set aside fields of a minimum of 40 acres (16.2 ha) in size and established native and non-native perennial grasses and shrubs, with a higher payment for native species (McNeil 2013). In 2018, the Ontario Species at Risk Partnerships on Agricultural Lands’ Grassland Stewardship Program made three-year agreements with producers to establish native and non-native grasslands to provide habitat for bobolink and other nesting grassland birds (Species at Risk Partnership on Agricultural Lands 2018). The DF&WT’s GLSA Stewardship Program is the only GLSA program currently active in Canada. The goals of the DF&WT GLSA Stewardship Program are to provide habitat for wildlife (e.g. nesting grounds for birds and small mammals, and hunting grounds for raptors), and to improve soil structure and increase soil organic matter (DF&WT 2019). Farmers can apply to enter a maximum of 50 acres (20.2 ha) or up to 50% of the land they farm or own into the GLSA 8  Stewardship Program, whichever is lower. This area is seeded with a standard DF&WT seed mix and then left fallow for a minimum of one and a maximum of four years (DF&WT 2019). Farmers receive $400/acre/year (~$964/ha/year). They may harvest the GLSA once per year, but payment is reduced to $200/acre/year (~$482/ha/year) if they do so (D. Bondar, Personal communication, Sept. 2020). The payment rates were previously set at $300/acre/year (~$741/ha/year) and $150/acre/year (~$370/ha/year) if hayed (DF&WT 2019) and increased in 2020. From 2008 to 2018, an average of 211 ha (522 acres) were enrolled in the GLSA Stewardship Program each year, comprised of 35% first year GLSA, 29% second year GLSA, 19% third year GLSA, 11% fourth year GLSA, and 6% GLSA in their fifth year or longer (D. Bondar, Personal communication, Sept. 2020).  The DF&WT GLSA seed mix is composed of 25% (by seed weight) orchard grass (Dactylis glomerata L.), 28% tall fescue (Festuca arundinacea Schreb.), 30% short fescue (Festuca rubra subsp. commutata Gaudin and F. rubra subsp. rubra L.), 15% timothy grass (Phleum pratense L.), and 2% red clover (Trifolium pratense L.) (DF&WT 2019). While not native to the area, the species in the mix were selected in part based on their habitat potential (Principe 2001). Before the 1890s, the natural vegetation of the LFRD was mainly herbaceous and shrubby, and the herbaceous portions consisted of grassland and marshes (Environment Canada and Canadian Wildlife Services 1998). Species were also selected based on their suitability to the climate and soil conditions, and their potential to improve soil (Principe 2001).  The flexibility and relatively short duration of the DF&WT GLSA Stewardship program is likely necessary for this program’s success, given the economic pressures and land tenure issues in the region. Farmers participate in the program for three main reasons: to improve degraded soil, to transition their fields to organic production, and/or to maintain soil quality by 9  including GLSA in their regular crop rotation (Lussier et al. 2019). As a result, fields entering the GLSA program vary widely in terms of baseline soil quality, ranging from degraded and unproductive to productive (Lussier et al. 2019).   1.3.2 Grassland Set-Aside Research: Worldwide Many studies have found GLSAs and comparable management practices to have positive effects on soil quality. Total soil carbon has been found to increase after converting cropped land to grass (Guo and Gifford 2002). After five years of GLSA, CRP sites in the Great Plains of the USA had significantly higher soil organic carbon than paired cropped sites (Gebhart et al. 1994). Total carbon and total nitrogen were higher in 2.5-year GLSAs in the CRP relative to paired cropped sites, and trends of higher carbon and nitrogen were observed in 4.5-, 5.3-, 5,5-, 6-, and 6.5-year GLSAs relative to cropped pairs (Karlen et al. 1999). A study comparing GLSA in the CRP to adjacent cropped fields found that on a site-by-site basis four out of six GLSAs of two, four, and six years had higher soil organic content than adjacent cropped fields, while no significant differences were found when results were averaged across sites (Bowman and Anderson 2002). Another study comparing 20 four- to seven-year GLSA sites in the CRP to adjacent cropped sites (in wheat-fallow rotation) reported that total nitrogen was 10% higher in GLSA than adjacent cropped sites (Staben et al. 1997). There is some evidence that regularly including short-term GLSAs in crop rotation could help maintain or even improve soil structure. Set-aside duration did not have a clear effect on soil structure in a study of the CRP (Karlen et al. 1999). With the exception of 4.5-year GLSAs in North Dakota, where no effects of GLSA were observed, percent water stable aggregates tended to be higher in GLSAs compared to cropped fields at all other study locations (Karlen et 10  al. 1999). Differences in Iowa were significant at 2.5-year GLSAs paired with cropped fields that were cultivated annually, but not significant at six-year GLSAs paired with minimum-tillage cropped fields (Karlen et al. 1999). Percent water-stable aggregates were also significantly higher in 6.5-year GLSAs in Minnesota compared to their cropped pairs and GLSAs were found to have a significantly higher mean aggregate size (Karlen et al. 1999). Bulk density measured at 7.5 cm depth (with the exception of North Dakota, where it was measured at 20 cm depth), was only significantly lower at the Minnesota GLSAs and at the 2.5-year GLSAs in Iowa relative to their cropped pairs (Karlen et al. 1999). Additional sites in Washington showed higher bulk density at 15 cm depth in 5.5-year GLSAs than cropped pairs (Karlen et al. 1999). These results suggest that aggregate stability can respond rapidly (2.5 to 6.5 years) under GLSA management, while changes in bulk density are more variable and may take longer than 6.5 years. As this study was carried out in various U.S. States and on different soils, it demonstrates the need to assess GLSA within their regional contexts.  The length of time needed under set-aside before seeing improvements in porosity and bulk density varies in other contexts as well. Pranagal et al. (2007) tracked bulk density and aeration porosity in a set-aside and two rotational cropping systems over a ten-year period. The study took place in Poland on sandy loam Podzol. The GLSA fields were removed from crop production, seeded with an oat and barley mix, and left to undergo natural succession for ten years. While bulk density was initially higher in the set-aside field than in the cropped fields, it was lower in the set-aside after four years. Pranagal et al. (2007) observed that bulk density in the GLSA decreased over time, especially in the fifth year, while bulk density tended to increase in the cropped fields. Total porosity increased in the GLSA over time. Average bulk density over the ten-year period was significantly lower in the set aside relative to ACR fields, while average 11  porosity was significantly higher in the GLSA (Pranagal et al. 2007). Meanwhile, a study of cropping systems in Norway examined aeration porosity and bulk density at plots that were managed under four-year rotations, some of which included one- to three-year grass fallows (Riley et al. 2008). After 15 years, bulk density was significantly higher and aeration porosity significantly lower compared to baseline in only one ACR treatment. While these parameters did not change significantly over time in any other treatments, bulk density was lowest in the treatment that had the longest GLSA duration (i.e., three years). Rotations that included two- to three-year set-asides tended to have a higher or similar aeration porosity than they did at baseline (Riley et al. 2008).   1.3.3 Grassland Set-Aside Research: Lower Fraser River delta, BC Set-asides vary in terms of their objectives, duration, and inclusion of associated management practices (e.g., fertilizer application, mowing, seeding of grasses and/or grass–legume mixtures, tillage). Consequently, set-aside programs should be evaluated within the geographical, agronomic, and socioeconomic context of the country or region where they are implemented (Kleijn and Baldi 2005). Indeed, the shorter duration of the GLSAs in the LFRD and the potential influence of soil type, climate, and other factors on GLSA effectiveness call for local studies. While many GLSA studies carried out in the LFRD have focused on effects of GLSAs to wildlife (Merkens 2004; Fairbrother et al. 2006), early and recent studies have focused on effects of GLSAs on soil quality and nutrient availability for crops following GLSA (Hermawan and Bomke 1996; Principe 2001; Yates et al. 2017; Lussier et al. 2019; Lussier et al. 2020; Fausak 2019; Walji et al. 2020).  12  In an early regional study, Hermawan and Bomke (1996) examined soil quality improvements to a degraded site under different management practices on silty clay loam Humic Luvic Gleysol in Delta, BC. The treatments consisted of GLSA comprised of 95% tall fescues (Festuca arundinacea Schreb.) and 5% timothy grass (Phleum pratense L.) and summer cash cropping with winter cover systems under two drainage regimes. Aggregate stability improved significantly under GLSA in both drained and undrained treatments by the spring of the second year, but not under cash crop and clover cover systems (Hermawan and Bomke 1996). Aggregate stability was higher at the drained sites compared to the undrained sites, except under GLSA in the fall of the second season (Hermawan and Bomke 1996). This finding suggests that aggregate stability under GLSA is less sensitive to water content fluctuations than under cash cropping. In addition, the 3-year GLSA had higher aggregate stability relative to the cash crop with winter cover crop systems (Hermawan and Bomke 1996).  Other local studies have yielded variable responses to GLSA treatment, perhaps due to variable field management, duration of GLSA, and baseline soil quality in the LFRD. In a study by Principe (2001), mean weight diameter (MWD) of water stable aggregates did not increase after one year of GLSA at a site on Westham Island, Delta, perhaps because MWD was high (3.1 mm). In that study, the proportion of water stable aggregates did increase significantly after one year of GLSA, from to 70% to 73% (Principe, 2001). Yates et al. (2017) compared soil properties of two, three, four and six-year GLSA fields and adjacent potato fields on silt loam to silty clay loams at operational farms in Delta, BC. The six-year GLSA site was highly degraded before GLSA establishment (Yates et al. 2017). At the end of the six years of GLSA, MWD was greater than at the adjacent cropped field, at 1.6 mm compared to 0.8 mm (Yates et al. 2017). Aeration porosity was significantly higher at the six-year GLSA relative to the paired cropped 13  field, at 0.10 m3 m-3 (the highest of all fields) compared to 0.08 m3 m-3 (Yates et al. 2017). Mean weight diameter (1.2 mm vs. 1.0 mm) and aeration porosity (0.09 m3 m-3) did not differ between the four-year GLSA and its pair (Yates et al. 2017); however, bulk density at this site was significantly lower under GLSA compared to the cropped field. No significant differences in aeration porosity, bulk density, or aggregate stability between two and three- year GLSAs and adjacent cropped fields were detected (Yates et al. 2017). The results of these studies suggest that prior and concurrent management practices can influence effects of GLSA on soil quality and outline the value of baseline soil property assessments. In a subsequent GLSA study by Lussier et al. (2019), baseline properties prior to GLSA establishment were assessed and GLSA and adjacent fields under annual crop rotation (ACR) were compared during the first two growing seasons of GLSA. Out of eight fields entering the GLSA Stewardship Program, two were deemed unproductive based on baseline (spring 2015) soil properties, namely, high exchangeable sodium, low MWD, low total carbon, and high bulk density (Lussier et al. 2019). The state of these unproductive sites was likely the result of soil degradation, and these sites were likely most affected by the regional degradation issues described in section 1.1. Furthermore, GLSA vegetation did not establish well at unproductive sites and did not show any significant soil improvements due to GLSA. At productive sites, MWD was greater at GLSA relative to ACR after two seasons. In the spring, MWD at productive GLSAs was 21% higher than at adjacent ACR fields (Lussier et al. 2019). In the summer, MWD was 14% higher at GLSAs than ACR fields (Lussier et al. 2019). In the fall, after crop harvest, MWD was 19% higher in GLSAs relative to ACR fields, aeration porosity was 24% higher, and bulk density was 7% lower (Lussier et al. 2019). These results suggest that two 14  growing seasons of GLSA is sufficient to improve soil structure at productive sites in the FRD, but not at unproductive sites.  Findings by Fausak (2018) suggest that three years of GLSA may lead to some improvements of degraded soils. Fausak established study plots at one unproductive and one productive site from the Lussier et al. (2019) study described above. In the spring of 2017, plots with two-year GLSA above-ground biomass either removed or incorporated were established, and nitrogen fertilization treatments were also applied. Plots were then seeded with beans. The following year, additional plots with or without the incorporation of three-year GLSA above-ground biomass were added to the study and potatoes were planted in all plots. Aggregate stability samples were collected pre-planting, mid-season, and post-harvest. At the productive site, in plots with no added nitrogen, MWD was higher where above-ground biomass incorporated compared to where it was removed in the first year following two-year GLSA (Fausak 2018). Differences in MWD between treatments were less pronounced at the unproductive site. In 2018, MWD was significantly higher following three-year GLSA incorporation relative to the second year following two-year GLSA incorporation at the productive site (Fausak 2018). A similar trend was observed at the unproductive site (Fausak 2018). The results of this study suggest that improvements to aggregate stability may only persist for one year following GLSA incorporation at productive sites.  The results from studies on GLSA in the LFRD to date underline the importance of baseline assessments in the evaluation of GLSA impacts on soil structure. While some study results suggest that GLSAs can improve structure in just one to two years (Hermawan and Bomke 1996; Principe 2001; Yates et al. 2017; Lussier et al. 2019), other results suggest that more time may be required, particularly at degraded sites (Yates et al. 2017; Fausak 2018; 15  Lussier et al. 2019). Further examination of three- and four-year GLSAs with baseline assessment is required to better understand their effects on soil physical quality in the LFRD.   1.4 Study Objectives and Hypotheses The objectives of this study were as follows:   Objective 1: Evaluate three- and four-year GLSAs relative to paired ACR fields at productive and unproductive sites in the LFRD based on soil aggregate stability, aeration porosity, bulk density, and (only for four-year GLSAs), total soil carbon and nitrogen.  Hypotheses:  1.1 Aggregate stability will be higher in GLSA fields relative to paired ACR fields during the third year (2018) and fourth year (2019) of GLSA at both productive and unproductive sites.  1.2 Bulk density will be lower in GLSA fields relative to paired ACR fields in 2018 and 2019 at both productive and unproductive sites.   1.3 Aeration porosity will be higher in GLSA fields relative to paired ACR fields in 2018 and 2019 at both productive and unproductive sites. 1.4 Total carbon and nitrogen will be higher in GLSA fields in 2019 relative to paired ACR fields at both productive and unproductive sites. Objective 2: Evaluate the effects of three- and four-year GLSAs and ACR on soil aggregate stability, bulk density, and total soil carbon and nitrogen (after four years only), at productive and unproductive sites relative to soil properties before GLSA establishment in the spring of 2015 and after one year of GLSA in the spring of 2016.   16  Hypotheses:  2.1 Aggregate stability will be higher in GLSA fields in 2018 and 2019 compared to 2015 and 2016 at both productive and unproductive sites. 2.2 Bulk density will be lower in GLSA fields in 2018 and 2019 compared to 2015 and 2016 at both productive and unproductive sites.  2.3 Aeration porosity will be higher in GLSA fields in 2018 and 2019 compared to 2016 at both productive and unproductive sites. 2.4 Total carbon and nitrogen will be higher in GLSA fields in 2019 compared to 2015 at productive sites. 2.5 ACR fields will not be improved.      17  Chapter 2: Materials and Methods 2.1 Study Sites This study took place from spring 2018 to spring 2019 on seven study sites located on operating farms in the LFRD in BC; and was part of a larger study initiated in 2015 (Krzic et al. 2020). The region is characterized by mild humid winters and warm dry summers, with mean annual precipitation of 927 mm in Delta and 1,082 mm in Richmond (Environment and Climate Change Canada no date a, Environment and Climate Change Canada no date b). The topography of the region is flat to slightly undulating, with an elevation of 2 m above sea level. Soils on the study sites developed from Fraser River deltaic deposits and were classified as Humic Luvic Gleysols, Orthic Gleysols, Orthic Humic Gleysols, Rego Gleysols, and Rego Humic Gleysols (Luttmerding 1981). Soil textures ranged from silt loam to silty clay loam (Lussier 2018) and natural drainage was classified as moderately poor to very poor (Luttmerding 1981).  A total of seven sites (ranging in size from 2 to 11 ha), were included in this study. Six sites were located in the municipality of Delta (Sites 2, 3, 4, 5, 8, and 9) and one site in the municipality of Richmond (Site 1). Each site consisted of one GLSA field and an adjacent field managed for annual crop rotation (ACR). The adjacent ACR pairs were selected on the basis of similar soil type and management history (Lussier et al. 2019). Fields were grouped by productivity based on baseline (i.e., 2015) soil quality and GLSA vegetation growth (Lussier et al. 2019). Out of the seven sites, two were deemed unproductive on the basis of high bulk density, low mean MWD, high exchangeable sodium percentage, and low GLSA biomass during the first season of GLSA establishment (Lussier et al. 2019); the remaining five sites were deemed productive. In May 2018, the GLSA biomass was incorporated into the soil at Site 1 and Site 9. These productive sites were planted with potatoes and data from these sites were excluded 18  from 2019 analyses. The other five sites remained in GLSA until May 2019 when their GLSA biomass was tilled into the soil.  All GLSA fields in the study were part of the DFWT GLSA Stewardship Program. GLSA fields were taken out of production in September of 2014 and seeded in 2015 with the standard GLSA seed mix that consisted of 30% short fescues, 28% tall fescue (Festuca arundinacea Schreb.), 25% orchard grass (Dactylis glomerata), 15% timothy (Phleum pratense L.), and 2% red clover (Trifolium pratense L.) by seed weight.  Because the study was conducted on operating farms, management practices and history varied among sites. Sites were under conventional or organic management, some had been laser-levelled, and some received manure applications (Fig. 1). Crops grown at each ACR also varied and included the following: potatoes, peas (Pisum sativum L.), barley (Hordeum vulgare L.), turnips (Brassica rapa subsp. rapa L.), and corn (Zea mays L.) (Appendix A).  2.2 Sampling and Analyses  At each GLSA and ACR field, four sub-plots (radius 6 m) were randomly generated using a global positioning system in 2015 (Lussier et al. 2019). Sub-plots were kept a minimum of  10 m away from field edges to avoid edge effect. In spring 2018, some sub-plots were found to be too close to field edges, due to ditch dredging and tractor work, and were moved to be 16 m away from the field edge. One sub-plot in the Site 4 GLSA field was found to be positioned on a former road, and therefore would not appropriately reflect field responses to GLSA. Data from this subplot was excluded from the analyses.  19  2.2.1 Baseline and First Year Data Baseline soil samples were collected in April 2015, before GLSAs were seeded, at the start of the larger study that included my study. A summary table of baseline properties can be found in Appendix B. Aeration porosity was not assessed in April 2015. Soil samples were also collected in spring 2016 as part of the larger study. Data from spring 2015 and spring 2016 were included in my analyses in order to address Objective 2 of this study.  2.2.2 Aggregate Stability Aggregate stability samples were collected at 0–7.5 cm depth. Two composite samples were collected from each sub-plot. Sampling took place in spring 2018 (pre-tillage in ACR fields), summer 2018 (during the growing season), fall 2018 (post ACR field harvest but before cultivation), and spring 2019 (pre-tillage). As part of the larger study, aggregate stability samples were also collected in 2015 and 2016, and this data will be included in my analysis. Aggregate stability determination was conducted following the wet-sieving method described by Nimmo and Perkins (2002), with variations as described by Wallace et al. (2009). The results for aggregate stability were expressed as the MWD, which is the summation of a series of Di × Wi products where Di is the mean diameter of each size fraction and Wi is the proportion of the sample weight occurring in the corresponding size fraction.      20  2.2.3 Soil Bulk Density and Aeration Porosity Samples were collected using 7.5 cm height by 7.5 cm diameter cores and a double cylinder drop-hammer sampler. One sample per depth (0–7.5 cm, 7.5–15 cm, and 15–30 cm) was collected at each sub-plot. These core samples were used to determine both aeration porosity and bulk density. Sampling took place in spring 2018 (pre-tillage in ACR fields), fall 2018 (post ACR field harvest but before cultivation), and spring 2019 (pre-tillage).  For determination of aeration porosity, soil cores were weighed, then progressively saturated in a container with tap water for 24 h (Danielson and Sutherland 1986). After saturation, the cores were weighed and placed on a tension table, which contained a tension medium of silicon carbide sand (grit 400). Aeration porosity (i.e., proportion of soil pores with diameter >50 µm) was determined on a tension table set at 6 kPa of matric suction. The same samples were then used to determine bulk density as oven-dried mass of soil per volume of soil at field moisture (Blake and Hartage 1986).   2.2.4 Total Soil Carbon and Nitrogen In spring of 2019, soil samples for total carbon and nitrogen were collected at 0–15 cm, 15–30 cm, 30–45 cm, and 45–60 cm depths using an auger. At each field, one sample per depth was collected. These samples were air-dried, sieved to < 2mm, and sent to the Analytical Chemistry Services Laboratory of the BC Ministry of Environment and Climate Change Strategy for analysis, where total carbon and nitrogen were determined by dry combustion using a Flash 2000 elemental analyzer (Nelson and Sommers 1982). Total soil carbon and nitrogen on the baseline samples collected in the spring of 2015 as part of the larger study were determined using 21  diffuse Fourier transform mid-infrared spectroscopy method (Reeves et al. 2001) run on a Tensor 37 HTS-XT spectrometer (Bruker Optics, Ettlingen, Germany).  2.2.5 Vegetation Biomass Aboveground biomass of GLSA vegetation was sampled in April 2018 and September 2018, by cutting the biomass (to about 2 cm above ground) within 50 cm × 50 cm quadrats. The samples were oven-dried at 60°C for five to seven days, sorted into categories - grass, clover, and weeds/other (i.e., all plant species that were not included in the GLSA mix), and weighed. This was done to provide information that could help understand and explain results pertaining to the study objectives.   2.3 Statistical Analyses Linear mixed-effects models were used to examine the treatment effect on MWD, aggregate size fractions, bulk density, and aeration porosity. These models had treatment as a fixed effect with two levels (ACR and GLSA), site type as a fixed effect with two levels (productive and unproductive), and subplot within treatment within site as a random effect. Data from each sampling time and each depth were analyzed in separate models. The models for MWD in spring 2019, the fraction of total soil in the 2–6 mm aggregate size class in spring 2018 and spring 2019, the fraction of total soil in the 1–2 mm aggregate size class in spring 2019, and the fraction of total soil in the 0.25–1 mm aggregate size class in summer 2018 included a variance function to account for unequal variance  22  In order to examine changes in soil properties relative to the 2015 baseline conditions, linear mixed-effects models with treatment as a fixed effect with two levels (ACR and GLSA), site type as a fixed effect with two levels, year as a fixed effect with four levels (2015, 2016, 2018 and 2019), and subplot within field within site as a random effect. Data from spring 2015 and spring 2016 were included in this analysis to address Objective 2. Data from each depth were analyzed in separate models. Models for total soil carbon, total nitrogen, and C:N ratio, did not include site type as a fixed effect as data for unproductive sites from 2015 and 2016 was not available. For aeration porosity, productive and unproductive sites were analysed in separate models, without site type as a fixed effect. Data for MWD and proportion of total soil in the 2–6 mm aggregate size class was log transformed. The models for aeration porosity at 7.5–15 cm and 15–30 cm depth at productive sites and 7.5–15 cm depth at unproductive sites included a variance function to account for unequal variance.  To compare above-ground vegetation biomass across GLSA fields, separate linear models with site as the fixed effect were used for each sampling time. In addition, I used a linear model with field type as a fixed effect with two levels (productive and unproductive), time as a fixed effect with three levels (spring 2018, fall 2018, and spring 2019), and subplot within site as a random effect to examine the effect of field type and sampling time on above-ground biomass.  Assumptions were tested for all models using Shapiro-Wilk normality tests and diagnostic plots. Type III analysis of variance (ANOVA) was to evaluate significance of interaction terms and factors in each model. Pairwise contrasts with Bonferroni adjustments were used to compare estimated marginal means where applicable. Significance was generally set at an α of 0.05, and in some cases 0.10. Statistical computation was performed using R software version 3.6.2. (R Core Team 2019).23   Figure 2.1 Overview of management history and soil characteristics of seven study sites located at operational farms in the lower Fraser River Delta (LFRD), British Columbia.Laser leveledYesOrganicManure application < 3 years prior to baselineYesExchangeable Na > 0.64 and < 2.08 cmolc kg-1Site 5ProductiveACR 5Humic Luvic Gleysol/ Orthic Humic GleysolSoil Series: LadnerDeltaSilt Loam% Sand: 12% Silt: 64% Clay: 24Drainage Class:moderately poor to poorComposted chicken manure applied2016, 2017, 2018 GLSA 5Rego-Humic GleysolSoil Series: BensonSpetiforeSilt Loam% Sand: 30% Silt: 55 % Clay: 15Drainage Class:poor to very poorBaseline exchange-able Na: 0.85 cmolc kg-1NoExchangeable Na < 0.64 cmolc kg-1Site 9ProductiveACR 9Humic Luvic GleysolSoil Series: LadnerSilt Loam% Sand: 25% Silt: 57% Clay: 18Drainage Class: moderately poor to poorComposted chicken manure applied 2015, 2016, 2017 GLSA 9Orthic Humic GleysolSoil Series: DeltaSilt Loam% Sand: 21% Silt: 60% Clay: 19Drainage Class:poorBaseline exchange-able Na:0.13cmolc kg-1NoOrganicManure application < 3 years prior to baselineYesExchangeable Na < 0.64 cmolc kg-1Site 8ProductiveACR 8Rego Humic Gleysol/ Orthic Humic Gleysol/ Orthic GleysolSoil Series: Westham Delta CrescentSilty Clay Loam% Sand: 8% Silt: 64% Clay: 28Drainage Class: moderately poor to poorComposted chicken manure applied2015, 2016, 2017GLSA 8Rego Humic Gleysol/ Orthic Gleysol/  Rego GleysolSoil Series: Westham Delta BlundellSilt Loam% Sand: 6% Silt: 67% Clay: 27Drainage Class: poor to very poorBaseline exchange-able Na: 0.20 cmolc kg-1Conventional Manure application < 3 years prior to baselineYesExchangeable Na < 0.64 cmolc kg-1Site 2ProductiveACR 2Humic Luvic GleysolSoil Series: LadnerSilt Loam% Sand: 7% Silt: 69% Clay: 24Drainage Class: moderately poor to poorChicken and steer maure applied 2016, 2017, 2018GLSA 2Rego Gleysol/Orthic Gleysol/ Humic Luvic GleysolSoil Series: Blundell  Crescent  LadnerSilt Loam% Sand: 5% Silt: 69% Clay: 26Drainage Class: moderately poor to very poorBaseline exchange-able Na:0.33 cmolc kg-1NoExchangeable Na < 0.64 cmolc kg-1Site 1ProductiveACR 1Rego Humic Gleysol/ Orthic GleysolSoil Series: Westham  CrescentSilty Clay Loam% Sand: 9% Silt: 63% Clay: 27Drainage Class: moderately poor to poorNo manure applicationsGLSA 1Orthic Gleysol/ Rego Humic Gleysol/ Rego GleysolSoil Series:CrescentWesthamBlundell Silt Loam% Sand: 14% Silt: 60% Clay: 26Drainage Class: moderately poor to very poorBaseline exchange-able Na: 0.07 cmolc kg-1Exchangeable Na > 0.64 and < 2.08 cmolc kg-1 Site 3UnproductiveACR 3Humic Luvic Gleysol/ Orthic GleysolSoil Series: LadnerCrescentSilt Loam% Sand: 12% Silt: 66% Clay: 22Drainage Class: moderately poor to poorNo manure applicationsGLSA 3Humic Luvic GleysolSoil Series:  LadnerSilt Loam% Sand: 15% Silt:  64% Clay: 21Drainage Class: moderately poor to poorBaseline exchange-able Na: 1.78 cmolc kg-1Exchangeable Na > 2.08 cmolc kg-1Site 4UnproductiveACR 4Humic Luvic Gleysol/ Orthic Humic GleysolSoil Series:  Ladner DeltaSilt Loam% Sand: 12% Silt: 64% Clay: 24Drainage Class:moderately poor to poor No manure applicationsGLSA 4Orthic Humic GleysolSoil Series: DeltaSilt Loam% Sand: 12% Silt: 64% Clay: 24Drainage Class:poorBaseline exchange-able Na:2.59 cmolc kg-124  Chapter 3: Results and Discussion Unlike many other long-term GLSAs found globally, GLSAs in the LFRD are short-term (one to four years) due to various land tenure issues. While the long-term GLSA programs mainly aim to improve soil quality on degraded sites, farmers in the LFRD mostly use the short-term GLSAs to transition fields to organic production and to diversify crop rotations, and in some cases to try to improve degraded soil. As a result, the fields entering the GLSA program in the LFRD vary from productive to highly degraded or unproductive, characterized by high soil salinity, poor structure, and low soil organic matter (Yates et al. 2017; Lussier et al. 2019). Degradation in the region is a result of the pairing of natural factors such as poor drainage and heavy precipitation with intensive cultivation due to market shifts and a history of precarious land tenure, which did not affect tenant and land-owner farmers equally. Thus, soil properties at fields entering GLSA vary as a result of differing management practices that may span decades.  Both productive and unproductive sites were included in my study. This distinction, meant to differentiate highly degraded sites, was determined by Lussier et al. (2019), who evaluated effects of GLSA on soil quality during the initial two years of GLSA establishment. In the first two GLSA growing seasons (i.e., 2015 and 2016), GLSA vegetation did not establish well at the unproductive sites and there was also a higher presence of weeds at unproductive sites relative to the productive sites (Lussier et al. 2019). Despite the initial delay in GLSA establishment and growth at the unproductive sites, by 2018 and 2019, there was no significant difference in the GLSA aboveground biomass between productive and unproductive sites (Fig. 3.1). Total mean aboveground biomass was higher in 2019 than 2018, and this was driven by significantly higher biomass at unproductive sites in spring 2019 than spring 2018 (p<0.10). Spring biomass at productive sites did not differ from 2018 to 2019. The proportion of weeds at 25  unproductive sites was 2% in spring 2018 and 0% in spring 2019. At productive sites, there were almost no weeds (i.e., 0.7% in 2018 and 0% in 2019) (see Appendix C). The observation that GLSA vegetation was equally well-established at unproductive and productive sites by the third growing season shows promise for improvements to soil quality at unproductive sites after three years.  26   Figure 3.1 Total aboveground dry grassland set-aside (GLSA) vegetation biomass in productive and unproductive study sites. Error bars represent standard error of the mean; n=4.  27  3.1 Comparing GLSA to ACR 3.1.1 Aggregate Stability  Average MWD and proportions of total soil sample in the four aggregate size classes (2–6, 1–2, 0.25–1, and <0.25 mm) are summarized in Table 3.1. Across productive and unproductive sites, MWD and the fraction of total soil in the 2–6 mm aggregate size class were significantly higher at GLSA than ACR in spring 2018 (p<0.01), fall 2018 (p<0.10), and spring 2019 (p<0.05). Meanwhile, the fraction of total soil in the <0.25 mm size class, which includes non-aggregated soil particles, was significantly lower at GLSA than ACR in spring 2018, fall 2018, and spring 2019 (p<0.10). These results illustrate seasonal effects on aggregate stability, likely driven by seasonal fluctuations in soil water content, which has been shown to have an inverse relationship to structural stability (Caron and Kay 1992; Angers and Caron 1998). It is to be expected that in dry, hot months, aggregates are stronger than they are in spring after the rainy winter of the LFRD. Due to these seasonal cycles of soil structural stability, MWD was lowest in spring and highest in summer at all fields. Temporal changes in soil structural stability were more pronounced at ACR, which included various cultivation practices, than at GLSA, which had no cultivation. These findings are similar to findings of another study carried out in the LFRD by Hermawan and Bomke (1996). Since seasonal variations in aggregate stability parameters were more pronounced at ACR, the difference between aggregate stability at GLSA and ACR was smaller in summer than at other times, and therefore no differences in aggregate stability parameters were found between treatments at this time of determination.   Summer notwithstanding, overall results for MWD and the proportion of soil in the 2–6 mm and <0.25 mm aggregate size classes show that after three and four years of GLSA 28  establishment, aggregate stability was greater at GLSA than ACR at both productive and unproductive sites. These results were expected as GLSA is thought to improve soil structure by providing a reprieve from cultivation and cover from raindrop impact over the winter months, supplying root exudates and rhizosphere habitat for organisms that promote aggregation, and enmeshing aggregates (Tisdall and Oades 1982; Degens 1997; Angers and Caron 1998; Lehmann et al. 2017). For the two middle size fractions, there were no differences due to treatment, with the exception of fall 2018 in the fraction of total soil in the 0.25–1 mm aggregate size class. In this case, there was a significant difference between treatments across productive and unproductive sites, showing that the fraction of total soil in the 0.25–1 mm aggregate size class was higher in ACR than GLSA (p<0.10). Other studies in the LFRD have similarly found that the effects of GLSA on aggregate stability are mainly evident in MWD and the proportion of soil in the 2–6 mm and <0.25 mm aggregate size classes (Hermawan and Bomke 1996; Yates 2014).  The interaction between treatment and site type was not significant in most cases, indicating that aggregate stability parameters responded similarly at productive and unproductive sites, where GLSA had greater aggregate stability than ACR (Table 3.1). Productive sites nevertheless had higher aggregate stability than unproductive sites. In spring 2018, MWD was significantly higher at productive than at unproductive sites (p<0.10) and the fraction of total soil in the <0.25 mm size class was significantly lower at productive than at unproductive sites (p<0.05). Some other studies carried out in the LFRD have found that productive and unproductive sites responded differently to GLSA. For instance, Lussier et al. (2019) found no differences between MWD at GLSA and ACR after two years of GLSA establishment only when unproductive sites were excluded from the analysis. Yates et al. (2017) found that MWD differed 29  significantly between ACR and GLSA only at a six-year GLSA site but not at sites where GLSA was two, three, or four years old, citing differences in management practices and site history as likely causes. Nevertheless, the values for MWD in Table 3.1 are largely comparable to those reported in other regional studies, values ranging from 0.57 mm (post-tillage) to 3.2 mm (Hermawan and Bomke 1996; Principe 2001; Liu et al. 2005; Yates et al. 2017; Fausak 2019; Lussier et al. 2019).    30  Table 3.1 Mean weight diameter (MWD) and proportion of total soil in four aggregate size classes in spring, summer and fall of 2018 and spring 2019 for productive and unproductive ACR and GLSA fields. The standard error of the mean is shown in brackets. Different lowercase letters indicate a significant difference due to treatment (i.e., ACR vs. GLSA at each sampling time). Values followed by different letters are significantly different at p<0.10.  Fraction of total soil in each aggregate size class (kg kg-1) Sampling Time Site Type Treatment n MWD (mm) 2-6 mm 1-2 mm 0.25-1 mm <0.25 mm 2018 Spring Productive ACR 5 1.40 (0.232) a 0.27 (0.055) a 0.13 (0.015) a 0.10 (0.003) a 0.50 (0.069) a GLSA 5 2.07 (0.134) b 0.44 (0.037) b 0.15 (0.005) a 0.08 (0.011) a 0.33 (0.023) b Unproductive ACR 2 0.74 (0.095) a 0.13 (0.022) a 0.06 (0.003) a 0.07 (0.003) a 0.74 (0.024) a GLSA 2 1.20 (0.223) b 0.24 (0.058) b 0.09 (0.002) a 0.08 (0.003) a 0.59 (0.055) b 2018 Summer Productive ACR 3 2.92 (0.489) a 0.68 (0.151) a 0.10 (0.049) a 0.07 (0.052) a 0.16 (0.051) a GLSA 3 3.02 (0.362) a 0.70 (0.110) a 0.11 (0.034) a 0.05 (0.027) a 0.14 (0.048) a Unproductive ACR 2 1.87 (0.215) a 0.38 (0.058) a 0.15 (0.002) a 0.14 (0.023) a 0.33 (0.035) a GLSA 2 1.62 (0.227) a 0.31 (0.062) a 0.16 (0.000) a 0.14 (0.010) a 0.40 (0.048) a 2018 Fall Productive ACR 3 2.31 (0.663) a 0.53 (0.185) a 0.10 (0.033) a 0.05 (0.023) a 0.33 (0.132) a GLSA 3 2.89 (0.409) b 0.67 (0.118) b 0.10 (0.027) a 0.04 (0.026) b 0.19 (0.065) b Unproductive ACR 2 1.09 (0.166) a 0.18 (0.039) a 0.14 (0.008) a 0.13 (0.009) a 0.54 (0.055) a GLSA 2 1.60 (0.194) b 0.32 (0.054) b 0.13 (0.008) a 0.07 (0.008) b 0.48 (0.038) b 2019 Spring Productive ACR 3 1.37 (0.482) a 0.26 (0.115) a 0.12 (0.024) a 0.14 (0.010) a 0.48 (0.138) a GLSA 3 1.92 (0.441) b 0.39 (0.120) b 0.16 (0.020) a 0.12 (0.033) a 0.33 (0.084) b Unproductive ACR 2 0.47 (0.002) a 0.05 (0.008) a 0.06 (0.023) a 0.15 (0.006) a 0.74 (0.010) a GLSA 2 1.00 (0.239) b 0.18 (0.057) b 0.10 (0.007) a 0.12 (0.004) a 0.61 (0.062) b Sampling Time Source of Variation df p-value 2018 Spring Treatment (T) 5 0.0028 0.0027 0.12 0.39 0.0075  Site Type (ST) 5 0.056 0.11 0.0062 0.20 0.023  T  ´ ST 5 0.46 0.25 0.80 0.22 0.77 2018 Summer Treatment (T) 3 0.86 0.87 0.52 0.80 0.70  Site Type (ST) 3 0.11 0.13 0.43 0.20 0.043  T  ´ ST 3 0.43 0.47 0.83 0.93 0.29 2018 Fall  Treatment (T) 3 0.066 0.07 0.76 0.067 0.085  Site Type (ST) 3 0.17 0.18 0.36 0.16 0.16  T  ´ ST 3 0.88 0.96 0.72 0.065 0.45 2019 Spring Treatment (T) 3 0.014 0.019 0.18 0.31 0.087  Site Type (ST) 3 0.23 0.25 0.065 0.79 0.15  T  ´ ST 3 0.63 0.82 0.92 0.59 0.88 31  3.1.2  Soil Bulk Density Results for bulk density are summarized in Table 3.2. Soil bulk density was generally similar on GLSA and ACR treatments across both site types and at all three depths of sampling. The only exceptions were observed in spring 2018, at 0–7.5 cm depth when bulk density was higher in ACR than GLSA, averaged across site types (p<0.01) and in spring 2019 at the 15–30 cm depth where bulk density was lower in ACR than GLSA, averaged across site types (p<0.10).  The values obtained for bulk density at depth 0–7.5 cm were within a similar range to those reported in other studies of GLSA in the LFRD. For instance, for depth 0–7.5 cm, bulk density values between 1.07–1.19 Mg m-3 were reported for productive GLSA (Lussier et al. 2019) and bulk density values between 1.13–1.37 Mg m-3 were observed at GLSA sites with mixed management (Yates et al. 2017). Krzic et al. (2000) suggested that a bulk density of 1.33 Mg m-3 could be potentially root-limiting in the LFRD based on its correspondence to aeration porosity of 0.10 m3 m-3, which is generally considered root-limiting (Greenland 1981). Some of my results for bulk density at unproductive ACR at depth 0–7.5 cm, unproductive ACR and GLSA at depth 7.5–15 cm, and unproductive ACR and GLSA at depth 15–30 cm fall within this potentially root-limiting range for the LFRD, above 1.33 Mg m-3 (Krzic et al. 2000).  As the high bulk density values at unproductive sites suggest, site type had an effect on bulk density. Averaged across treatment, bulk density was higher at unproductive sites than at productive sites at depth 0–7.5 cm in spring 2018 and spring 2019 (p<0.10 for both contrasts), at depth 7.5–15 cm in spring 2018 (p<0.05) and spring 2019 (p<0.10), and at depth 15–30 cm in fall 2018 (p<0.05). However, the interaction between treatment and site type was not significant at any time, indicating that the trends in response to treatment were the same at productive and unproductive sites. 32   The differences in bulk density observed at depth 0–7.5 cm, revealed lower bulk density at GLSA than ACR across productive and unproductive sites. These differences were inconsistent; however, as they were only significantly in the spring of 2018, and not 2019. My results contradict somewhat with findings of some other studies that after four years of GLSA fields had lower bulk density compared to adjacent ACR. For example, a study by Pranagal et al. (2007) carried out in Poland did report lower soil bulk density after four years of GLSA but the soils in that study were sandier than the soils in the LFRD. In the LFRD, mixed results have been found when comparing bulk density at GLSA and ACR. A four-year GLSA was found to have significantly lower bulk density than its ACR pair, at 1.13 and 1.34 Mg m-3 respectively (Yates et al. 2017). However, the same study reported no differences in bulk density comparing two-, three-, or six-year GLSA to adjacent ACR and bulk density at these sites, and bulk density ranged from 1.20–1.37 Mg m-3 (Yates et al. 2017). The study by Lussier et al. (2019), which was carried out on the same sites as my study, reported that bulk density was lower at GLSA than ACR, at 1.08 and 1.17 Mg m-3, respectively. While these values were close to what I found in spring 2019 and higher than what I observed in fall 2018, I found no differences in bulk density between GLSA and ACR in spring 2019 or fall 2018.  These results indicate that GLSA did not reduce compaction relative to ACR and suggest that compaction may have been worse at GLSA fields before seeding. The differences between bulk density at GLSA and ACR, however minimal and inconclusive, were only observed at the 0–7.5 cm depth. Improvements to bulk density by GLSA most likely occur through the loosening of soil caused by root penetration (Angers and Caron 1998). GLSA may also prevent further reduction of bulk density through the cessation of activities that cause compaction, such as farm 33  traffic (Paul and de Vries 1979) and tillage (Krzic et al. 2000). Grass roots, while capable of deep rooting, mainly populate the upper soil profile (Bolinder et al. 2002; Brown et al. 2010; Jackson et al. 1996; Lecain et al. 2006). It is possible that this rooting pattern would develop in compacted subsoils with a high water-table and may explain why no differences were observed below 7.5 cm depth. Perhaps if kept under GLSA beyond four years, we would see differences in soil structure at greater depths and more pronounced and consistent differences at 0–7.5 cm depth.   3.1.3 Aeration Porosity Results for aeration porosity are summarized in Table 3.2. At depth 0–7.5 cm, in spring 2018, aeration porosity was significantly higher in GLSA than ACR at productive sites (p<0.10), but not at unproductive sites (Table 3.2). At depth 7.5–15 cm, at all sampling times, aeration porosity was lower at GLSA than at ACR averaged across site type (interaction between treatment and site type was not significant). Aeration porosity was significantly higher at productive sites than unproductive sites at depth 0–7.5 cm in spring 2019 (p<0.05) and at depth 7.5–15 cm in spring 2018 (p<0.05) and spring 2019 (p<0.10). These results partially support the hypothesis that after three and four years, aeration porosity would be higher at GLSA relative to ACR and are consistent with bulk density results. GLSA only had significantly higher aeration porosity relative to ACR in spring 2018, and this was only the case at productive sites. While not statistically different, the value for aeration porosity at productive GLSA was lower than at productive ACR in spring 2019. Aeration porosity was found to be significantly higher at GLSA than ACR at productive sites after the second GLSA growing season, in fall 2016 (Lussier et al. 2019), but in the fall of 2018 (the 34  fourth GLSA season) there was no difference in aeration porosity between GLSA and ACR. My observations indicate that four years of GLSA is not sufficient to result in consistently higher aeration porosity at GLSA relative to ACR at unproductive and productive sites.  These results are in accordance with mixed findings for aeration porosity reported in a previous GLSA study in the LFRD. Differences in aeration porosity between GLSA and ACR were observed at a six-year GLSA, but not at sites where the GLSA was two or three years old (Yates et al. 2017). Management practices and history differed among the four sites in that study, including differences in terms of compost applications, organic management, laser-levelling and drainage installation, and it may be that aeration porosity responded to factors other than GLSA duration (Yates et al 2017). By contrast, increased total aeration after four years of GLSA was observed compared to adjacent cropped fields in Poland, indicating that improvements to aeration porosity may be possible under different soil types or management history outside the LFRD (Pranagal et al. 2007).  As with bulk density, any improvements I observed in aeration porosity due to GLSA were only observed at depth 0–7.5 cm. Improvements to aeration porosity due to GLSA most likely occur through the enlargement of existing pores and creation of biopores, both effects of root penetration (Dexter 1991; Angers and Caron 1998). As with bulk density, it may be that the majority of GLSA roots populated the upper soil profile, thus concentrating improvements in this area. Yates (2014) found that a six-year GLSA had higher aeration porosity than a paired ACR field at 15–30 cm depth, suggesting that perhaps a longer duration under GLSA could improve subsoils. However, they did not report improvements in aeration porosity at 7.5–15 cm depth (Yates 2014). 35  The values I obtained were similar to those reported by Yates et al. (2017) for two- to six-year GLSA and adjacent ACR in the LFRD, ranging from 0.06–0.11 m3 m-3. By contrast, Lussier et al. (2019) reported aeration porosity of 0.16 m3 m-3 after one year of GLSA in spring 2016 and 0.15 m3 m-3 at adjacent ACR. Aeration porosity of 0.10 m3 m-3 or below is generally considered root-limiting (Greenland 1981). In many cases my findings for aeration porosity fell near or close to this value, including at productive GLSAs in 2019. At depth 0–7.5, aeration porosity at unproductive ACR fields in fall 2018 was strikingly high. Since fall samples were taken after crop harvest, it is possible that increased cultivation throughout the season and the disruptive nature of root and tuber crop harvest loosened the soil, contributing to a temporary boost in aeration porosity. Out of the two unproductive ACR fields, one was planted with turnips and the other with potatoes in the 2018 growing season. Turnips were grown in multiple consecutive plantings, requiring cultivation throughout the growing season. Meanwhile, potato crops transform fields into a landscape of furrows and hills (Chow and Rees 1994; BC Ministry of Agriculture 2012), and potato planting, cultivation, hilling, and harvest have been found to cause significant soil disruption (Tiessen et al. 2007).         36  Table 3.2 Soil aeration porosity and bulk density for productive and unproductive ACR and GLSA fields. The standard error of the mean is shown in brackets. Different lowercase letters indicate a significant difference due to treatment (i.e., ACR vs. GLSA at each sampling time) (α = 0.10), while uppercase letters indicate significant differences between treatment × site type (α = 0.10) where the interaction was significant. Letters cannot be compared between the different times or depths.  Aeration Porosity  Bulk Density Depth (cm) Time Site Type Treatment n (m3 m-3)  n (Mg m−3) 0-7.5 2018 Spring Productive ACR 5 0.11 (0.010) A  5 1.18 (0.037) a GLSA 5 0.15 (0.006) B  5 1.06 (0.054) b Unproductive ACR 2 0.11 (0.005)   2 1.33 (0.016) a GLSA 2 0.11 (0.009)   2 1.21 (0.040) b 2018 Fall Productive ACR 3 0.19 (0.026)   3 1.02 (0.074)  GLSA 3 0.20 (0.025)   3 1.00 (0.082)  Unproductive ACR 1 0.25   2 1.13 (0.085)  GLSA 2 0.14 (0.019)   2 1.22 (0.045)  2019 Spring Productive ACR 3 0.12 (0.017)   3 1.12 (0.076)  GLSA 3 0.14 (0.010)   3 1.07 (0.046)  Unproductive ACR 2 0.09 (0.000)   2 1.34 (0.029)  GLSA 2 0.09 (0.004)   2 1.27 (0.031)  7.5-15 2018 Spring Productive ACR 5 0.13 (0.009) a  5 1.18 (0.041)  GLSA 5 0.10 (0.007) b  5 1.21 (0.018)  Unproductive ACR 2 0.09 (0.004) a  2 1.35 (0.024)  GLSA 2 0.07 (0.007) b  2 1.36 (0.040)  2018 Fall Productive ACR 3 0.19 (0.016) a  3 1.08 (0.071)  GLSA 3 0.14 (0.008) b  3 1.20 (0.033)  Unproductive ACR 1 0.22              a  2 1.26 (0.100)  GLSA 2 0.13 (0.003) b  2 1.31 (0.027)  2019 Spring Productive ACR 3 0.14 (0.003) a  3 1.13 (0.061)  GLSA 3 0.10 (0.010) b  3 1.22 (0.048)  Unproductive ACR 2 0.09 (0.002) a  2 1.36 (0.009)  GLSA 2 0.08 (0.021) b  2 1.31 (0.024)  15-30 2018 Spring Productive ACR 5 0.11 (0.014)   5 1.23 (0.050)  GLSA 5 0.09 (0.011)   5 1.25 (0.024)  Unproductive ACR 2 0.09 (0.008)   2 1.38 (0.029)  GLSA 2 0.09 (0.006)   2 1.36 (0.025)  2018 Fall Productive ACR 3 0.17 (0.006)   3 1.14 (0.044)  GLSA 3 0.13 (0.010)   3 1.21 (0.021)  Unproductive ACR 1 0.11       2 1.35 (0.017)  GLSA 2 0.13 (0.018)   2 1.33 (0.029)  2019 Spring Productive ACR 3 0.14 (0.019)   3 1.20 (0.073) a GLSA 3 0.08 (0.004)   3 1.31 (0.039) b Unproductive ACR 2 0.10 (0.021)   2 1.34 (0.068) a GLSA 2 0.09 (0.003)   2 1.38 (0.003) b  37  3.1.4 Total Soil Carbon and Nitrogen  After four years of GLSA establishment, there were no differences in total soil carbon and nitrogen by treatment or site type at 0–15, 15–30, and 45–60 cm depths (Table 3.3). At depth 30–45 cm, both total carbon (p<0.10) and total nitrogen (p<0.05) were higher under GLSA than ACR fields, averaged across productive and unproductive sites. In addition, total carbon and nitrogen at depth 30–45 cm were higher at unproductive sites than productive sites (p<0.10). For total carbon at depth 45–60 cm, the interaction between treatment and site type was significant (p<0.10), but no differences were found when pairwise comparisons were performed using Bonferonni’s adjustment. Without Bonferroni’s adjustment, there was a significant difference between ACR and GLSA at unproductive sites (p=0.10). It is likely that these differences were not evident using Bonferroni’s adjustment due to a lack of statistical power. The interaction between treatment and site type was also significant (p=0.10) for total nitrogen at depth 45–60 cm, but no differences were found when pairwise comparisons were performed.  C:N ratio was significantly higher at GLSA than ACR at depth 15–30 cm (p<0.05). At depth 30–45 cm, there was a significant interaction between treatment and site type (p<0.10). No differences were found when pairwise comparisons were performed using Bonferroni’s adjustment. Without this adjustment, C:N ratio was found to be higher at productive ACR than at unproductive ACR (p< 0.05). Similarly, the interaction at depth 45–60 cm was also significant (p<0.05). Again, significant differences between pairwise comparisons were only found without Bonferroni’s adjustment. At this depth, C:N ratio at unproductive sites was found to be higher at GLSA than ACR (p<0.05).  These results do not support the hypothesis that after four years of GLSA establishment, total carbon and nitrogen would be higher in GLSA relative to ACR, nor are these results 38  surprising. Most studies reporting increases in total soil carbon due to GLSA examined longer-term GLSA, ranging from 8–35 years in duration (Post and Kwon 2000, Rosenzweig et al. 2016), or GLSA in markedly different climatic regions and soil types (Gerbhart et al. 1995, Bowman and Anderson 2002). Meanwhile, studies focused on the short-term GLSA have reported increases in total soil carbon. For instance, no differences in total organic carbon were found between four- to seven-year GLSAs in the CRP and adjacent cropped fields in Washington, USA (Staben et al. 1997). In Nebraska, no differences in total carbon or nitrogen were found when recently established CRP sites (with zero seasons under GLSA) were compared to CRP sites that had spent ten seasons under GLSA (Baer et al. 2000).  Within the LFRD, Yates et al. (2017) did not observe differences in total carbon at 0–7.5 cm depth when comparing two-, three-, four, and six-year GLSAs to adjacent ACR. They found no differences in total nitrogen comparing two- and three-year GLSAs to adjacent bare ACR in the spring and found that in four- and six-year GLSAs, total nitrogen at 0–7.5 cm depth was higher in ACR pairs (Yates et al. 2017). However, GLSA duration was not the only factor in this study, as the sites were under variable management practices. For instance, the six-year GLSA had been highly degraded while the site with the four-year GLSA was transitioning to organic management and received poultry manure applications (Yates et al. 2017).  Similarly, the carbon and nitrogen results of my study are best examined through the lens of management practices used at different sites. Despite a lack of significant results overall, GLSA at productive sites tended to have lower carbon and nitrogen than ACR pairs, while the opposite was true at unproductive sites. All three ACR fields at productive sites that remained in the study until 2019 received additional carbon inputs in the form of compost, whereas the unproductive sites did not (Figure 2.1). Meanwhile, at unproductive sites, where ACR did not 39  receive additional carbon inputs, increases to carbon and nitrogen due to GLSA are becoming perceptible.   Table 3.3 Mean total carbon, total nitrogen, and C:N ratio at productive (n=3) and unproductive (n=2) ACR and GLSA fields in spring 2019. The standard error of the mean is shown in brackets. Lowercase letters indicate significant differences due to treatment when there was no interaction between treatment and site type while uppercase letters indicate significant differences between treatment × site type (α = 0.10). Depth (cm) Site Type Treatment Total C (%) Total N (%) C:N 0-15 Productive ACR 2.97 (0.353) 0.26 (0.030) 11.2 (0.16) GLSA 2.70 (0.700) 0.24 (0.059) 11.0 (0.82) Unproductive ACR 1.45 (0.050) 0.14 (0.005) 10.7 (0.03) GLSA 2.05 (0.150) 0.17 (0.010) 12.2 (0.17) 15-30 Productive ACR 2.53 (0.484) 0.22 (0.044) 11.4 (0.21) a GLSA 2.17 (0.549) 0.20 (0.038) 10.8 (1.05) b Unproductive ACR 1.25 (0.150) 0.12 (0.015) 10.9 (0.12) a GLSA 1.95 (0.050) 0.16 (0.005) 12.6 (0.09) b 30-45 Productive ACR 0.69 (0.091) a 0.07 (0.006) a   9.7 (0.51) GLSA 0.66 (0.084) b 0.06 (0.005) b   9.7 (0.79) Unproductive ACR 0.40 (0.040) a 0.05 (0.002) a   8.3 (0.49) GLSA 1.18 (0.420) b 0.10 (0.031) b 11.7 (0.65) 45-60 Productive ACR 0.52 (0.045) 0.06 (0.004) A   9.2 (0.39) GLSA 0.42 (0.104) 0.05 (0.006)   8.9 (1.19) Unproductive ACR 0.27 (0.010) A 0.04 (0.001) B   7.6 (0.18) A GLSA 0.64 (0.145) B 0.06 (0.012) 11.4 (0.26) B  3.1.5 Conclusions of GLSA and ACR comparison My results show that at both productive and unproductive sites in the LFRD, three years of GLSA is sufficient to observe differences in aggregate stability relative to ACR. While differences in bulk density, aeration porosity, total carbon, and total nitrogen were observed in particular years or depths, these were inconsistent, indicating perhaps a weak effect of GLSA. It is likely that more pronounced differences in aeration porosity, bulk density, and total carbon and nitrogen would either require long-term GLSA or additional management practices besides GLSA. For instance, none of the sites included in this study had sub-surface drainage installed, 40  and only some have been laser-levelled or receive compost or manure applications (Figure 2.1). At productive sites, GLSA plays a role either as a part of crop rotation or in transitioning fields to organic production. These fields often receive manure and compost applications and some have been laser-levelled (Figure 2.1). Neither of the sites classified as unproductive received compost or manure applications since at least as far back as 2012, neither have been laser-levelled, and both have high exchangeable sodium (Figure 2.1). These short-term GLSAs show promise in improving and maintaining aggregate stability of soils in the LFRD. Still, additional management practices, beyond GLSA, must be implemented to address issues of soil physical quality, especially at unproductive sites.   3.2 Changes in Soil Properties Over Four Years of GLSA Relative to Baseline Soil Conditions 3.2.1 Aggregate Stability MWD and the fraction of total soil in the 2–6 mm aggregate size class from spring 2015 to spring 2019 are presented in Figures 3.2 and 3.3. At both productive and unproductive GLSA, MWD and the proportion of total soil in the 2–6 mm aggregate size fraction were significantly higher in 2016, 2018, and 2019 compared to 2015 - the year in which GLSA was established (p<0.05 for MWD, unproductive 2–6 mm size fraction 2015-2019 comparison; p<0.01 for other 2–6 mm size fraction comparisons). By contrast, across both productive and unproductive GLSA, the proportion of total soil in the <0.25 mm size fraction was significantly higher in 2015 compared to 2016, 2018, and 2019 (p<0.05) (Fig. 3.4).  At productive ACR, MWD and the proportion of total soil in the 2–6 mm aggregate size class were significantly higher in 2016 than 2015 (p<0.05 for MWD; p<0.01 for 2–6 mm size 41  fraction) but differences were not observed in other years. Meanwhile, at unproductive ACR, MWD decreased over time as it was significantly higher in 2015 and 2016 than in 2019 (p<0.05). The proportion of total soil in the 2–6 mm aggregate size fraction at unproductive ACR was significantly lower in 2019 than 2015 (p<0.10, data not shown), 2016 (p<0.01), and 2018 (p<0.10, data not shown).  Results from the middle aggregate size classes were not as distinct. The proportion of total soil in the 1–2 mm aggregate size class was higher in 2015 and 2016 (0.13 kg kg-1) than in 2018 and 2019 (0.10 kg kg-1) across productive and unproductive ACR fields (p<0.05). This size fraction did not change significantly at GLSA, ranging from 0.13–0.15 kg kg-1 (data not shown). While there were some significant fluctuations in the proportion of total soil in the 0.25–1 mm aggregate size class, there were no significant differences between baseline (i.e., 2015) and 2018 or 2019, indicating that ACR and GLSA had no discernible effects on this size fraction (data not shown).  The changes in MWD and the proportion of total soil in the 2–6 mm and <0.25 mm aggregate size fractions indicate that the improvements to aggregate stability took place in the first year of GLSA and were maintained thereafter. Given the absence of further changes beyond the first year of GLSA, it can be deduced that further improvements to MWD would require additional interventions (such as installation of subsurface drainage system and/or organic matter additions). At ACR fields, continued cropping had a deleterious effect on aggregate stability at unproductive sites. Since larger size fractions did not increase at ACR, the decrease in 1–2 mm size fraction likely represents a degradation of aggregates in this fraction in the third and fourth years of this study.  42  MWD is known to be a very responsive soil parameter to management practices, and previous studies in the LFRD have reported improvements to MWD under GLSA and vegetative cover in short time periods. For instance, Liu et al. (2005) reported significantly higher MWD after only 8 months (including winter months) under cover crops compared to bare soil, while Lussier et al. (2019) found GLSA had higher MWD than ACR at productive sites after only one year. Hermawan and Bomke (1996) observed higher MWD and proportion of total soil in the 2–6 mm aggregate size class after two years of GLSA relative to ACR with winter cover cropping at a degraded site in the LFRD, with and without sub-surface drainage. Despite this, some regional studies have suggested that improvements to aggregate stability at unproductive sites may take longer than one year under GLSA (Lussier et al. 2019) and even longer than four years (Yates et al. 2017). These aforementioned studies compared GLSA to paired ACR fields, not to baseline data from before the establishment of GLSA. Some GLSA studies attribute mixed results to different management practices at different sites (Karlen et al. 1999; Yates et al. 2017). My results underscore the need to not only compare GLSA to ACR, but to also track changes over time, as gains provided by GLSA may be obscured by differences in soil quality among sites at the time of GLSA seeding. In the LFRD, this comparison may be particularly necessary if farmers are choosing to enroll in the GLSA program their most degraded fields. 43   Figure 3.2 Mean weight diameter of water stable aggregates (MWD) at productive and unproductive annual crop rotation (ACR) and grassland set-aside (GLSA) fields in spring 2015, 2016, 2018, and 2019. Different letters denote significant differences based on pairwise contrasts of estimated marginal means using Bonferroni’s adjustment at p <0.05.     44   Figure 3.3 Fraction of total soil in the 2–6 mm aggregate size class at productive and unproductive annual crop rotation (ACR) and grassland set-aside (GLSA) fields in spring 2015, 2016, 2018, and 2019. Different letters denote significant differences based on pairwise contrasts of estimated marginal means using Bonferroni’s adjustment at p <0.05.   45   Figure 3.4 Fraction of total soil in the <0.25 mm aggregate size class annual crop rotation (ACR) and grassland set-aside (GLSA) fields in spring 2015, 2016, 2018, and 2019. Different letters denote significant differences by year (averaged across site type) based on pairwise contrasts of estimated marginal means using Bonferroni’s adjustment at p <0.05.  3.2.2 Soil Bulk Density At depth 0–7.5 cm, across productive and unproductive GLSA, bulk density was significantly higher in 2015 and 2016 compared to 2018 (p<0.01, p<0.05, respectively), but not to 2019 (Fig. 3.5). No changes were observed over time at ACR at this depth. At depth 7.5–15 cm, averaged across site type, ACR fields had lower bulk density in 2015 than in 2018 and 2019 (p<0.05). Bulk density at GLSA did not change significantly at this depth, indicating a mitigating effect of GLSA on compaction. At depth 15–30 cm, the effects of treatments and site type were not significant and bulk density was higher in 2015 and 2016 than in 2018 and 2019 across site types and treatments (p<0.01).  46   The results for bulk density were varied, suggesting that three years of GLSA improve bulk density at depth 0–7.5, but that this improvement may not necessarily be maintained over time. At productive GLSA, mean bulk density in 2018 and 2019 were almost identical at depth 0–7.5 cm (Table 3.2).  My results were similar to findings of a study in Norway, where after 15 years of crop rotation that included two- or three-year GLSA, bulk density was similar to baseline or slightly lower, but not significantly different (Riley et al. 2008). In my study, carried out on operating farms, the data was spread widely, and therefore it is difficult to ascertain whether GLSA improves bulk density after four years. At depth 7.5–15, it appears that GLSA had a mitigating effect on compaction at unproductive sites, as bulk density increased at ACR over time (p<0.05), but not at GLSA. The cause of mixed bulk density results is unclear. Bulk density means at productive GLSA in 2018 (1.06 Mg m-3) and 2019 (1.07 Mg m-3) were almost identical; however, the estimated marginal means are different enough that there was no significant difference in bulk density in 2019 relative to the beginning of the study in 2015. Two productive sites were removed from the study after spring 2018 sampling, as some farmers chose to return GLSA to agricultural production after three years. Only three productive sites remained in the study by 2019 instead of five, which could have affected findings. The variability in bulk density may also be partially attributed to rainfall patterns (See Appendix D for summary of weather conditions during years relevant to this study). Rainfall from May through August of 2018 was unusually low, at 43.2 mm, compared to 109.6 mm for the same period in 2017 and the historical average of 154 mm (Environment and Climate Change Canada no date a). This exceptionally dry growing season was followed by unusually high rainfall in September 2018, at 116.6 mm compared to the historical average of 39.8 mm, and rainfall in September 2017 which was 37.8 47  mm (Environment and Climate Change Canada no date a). Heavy rainfall has a deleterious effect on soil structure (Krzic et al. 2000), and rapid wetting can cause microcracks and slaking (Dexter 1991). Thus, the combination of a dry growing season followed by intense rainfall may have caused more severe damage to soil structure than normal and may explain why bulk density was higher in spring 2019 than spring 2018 at GLSA. Aggregate stability, which is more directly affected by rainfall impact than bulk density, also tended to be lower at GLSA in 2019 than in 2018. Total rainfall from September through April was higher in the 2017-2018 season than 2018-2019, driven by a drier period from December through April in the 2018-2019 season compared to the 2017-2018 season, but occurred more gradually and was not preceded by as dry a growing season. Cultivation and vegetation growth could also impact bulk density. While changes in crop type from year to year could have affected bulk density at ACR, crop type does not affect GLSA. At GLSA, aboveground biomass was higher in spring 2019 than spring 2018. Therefore, GLSA growth does not explain the differences in bulk density. It may be that bulk density fluctuates in response to factors not captured by this study. Ultimately, the effect of GLSA on bulk density appears to be positive but since the improvements due to GLSA are inconsistent, short-term GLSA alone cannot be relied upon to alleviate soil compaction in the LFRD. 48   Figure 3.5 Mean bulk density at annual crop rotation (ACR) and grassland set-aside (GLSA) fields at depth 0–7.5 cm in spring 2015, 2016, 2018, and 2019. Different letters denote significant differences by year (averaged across site type) based on pairwise contrasts of estimated marginal means using Bonferroni’s adjustment at p <0.05.  3.2.3 Aeration Porosity Results for aeration porosity are presented in Figures 3.6 to 3.10. At productive sites, aeration porosity at depth 0–7.5 cm was significantly higher in 2016 than in 2018 and 2019 (p<0.05) at ACR but did not change at GLSA. At the same sites at depth 7.5–15 cm, however, aeration porosity was higher in 2016 than in 2018 and 2019 (p<0.05) at GLSA and did not change at ACR. At depth 15–30 cm, aeration porosity was higher in 2016 than in 2018 and 2019 at GLSA (p<0.05) and lower in 2016 at ACR. At unproductive sites, aeration porosity was lower in 2018 and 2019 than in 2016 (p<0.01) at both ACR and GLSA at all depths. There was no data available for aeration porosity from 2015.  49  These results indicate that at depth 0–7.5 cm at productive sites, GLSA prevented a further reduction of aeration porosity. GLSA did not improve aeration porosity at lower depths or at unproductive sites. This result was unexpected, given the hypothesis that GLSA aeration porosity would increase at GLSA over time. A study in Norway found that after 15 years, rotations that included two- to three-year GLSA tended to have a higher or similar aeration porosity than they did at baseline (Riley et al. 2008). While we lack baseline data, the lack of change from 2016 to 2019 at productive GLSA indicates that GLSA may have a similar effect in the LFRD, one of maintaining aeration porosity. 50   Figure 3.6 Mean aeration porosity at productive and unproductive annual crop rotation (ACR) and grassland set-aside (GLSA) fields at depth 0–7.5 cm in spring 2016, 2018, and 2019. Different letters denote significant differences based on pairwise contrasts of estimated marginal means using Bonferroni’s adjustment at p <0.05.   51   Figure 3.7 Mean aeration porosity at productive annual crop rotation (ACR) and grassland set-aside (GLSA) fields at depth 7.5–15 cm in spring 2016, 2018, and 2019. Different letters denote significant differences based on pairwise contrasts of estimated marginal means using Bonferroni’s adjustment at p <0.05.   Figure 3.8 Mean aeration porosity at unproductive annual crop rotation (ACR) and grassland set-aside (GLSA) fields at depth 7.5–15 cm in spring 2016, 2018, and 2019. Different letters denote significant differences by year (averaged across treatment) based on pairwise contrasts of estimated marginal means using Bonferroni’s adjustment at p <0.05. 52   Figure 3.9 Mean aeration porosity at productive annual crop rotation (ACR) and grassland set-aside (GLSA) fields at depth 15–30 cm in spring 2016, 2018, and 2019. Different letters denote significant differences based on pairwise contrasts of estimated marginal means using Bonferroni’s adjustment at p <0.05.   Figure 3.10  Mean aeration porosity at unproductive annual crop rotation (ACR) and grassland set-aside (GLSA) fields at depth 15–30 cm in spring 2016, 2018, and 2019. Different letters denote significant differences by year (averaged across treatment) based on pairwise contrasts of estimated marginal means using Bonferroni’s adjustment at p <0.05. 53   3.2.4 Total Soil Carbon and Nitrogen There were no significant differences in total soil carbon, total nitrogen, or C:N ratio between spring 2015 and spring 2019 at any depth (Appendices E–G). As noted in section 1.1.4, studies that have found increases in soil carbon due to GLSA have usually assessed long-term GLSA and/or taken place in different climatic regions (Gerbhart et al. 1995; Post and Kwon 2000; Bowman and Anderson 2002; Rosenzweig et al. 2016). It is therefore unsurprising to have found no significant change in carbon after four years. It may be that total nitrogen is more influenced by other management practices than by GLSA, such as additions of fertilizer, manure, and/or compost.  3.2.5 Conclusions of Changes in Soil Properties Over Four Years of GLSA Relative to Baseline Soil Conditions Changes observed during the four years of GLSA establishment relative to the soil conditions just before GLSA seeding show that baseline measures are required to fully understand the effect of GLSA. When LFRD farmers enroll fields into the GLSA Stewardship Program to improve degraded soils, they often choose their worst fields, and as a result, changes may not be apparent if the fields are only compared to adjacent ACR. The importance of the comparisons to the baseline soil conditions is best demonstrated by the results for aggregate stability in this study. At productive sites, the improvements to aggregate stability caused by GLSA in the first year were apparent through comparisons of ACR and GLSA (Lussier et al. 2019), likely because baseline aggregate stability was similar at ACR and GLSA. Lussier et al. (2019) found that GLSA did not have improved aggregate stability relative to ACR when unproductive and productive sites were analysed together but did find differences when 54  unproductive sites were excluded from the analysis. In a site-by-site analysis of the same data set after two seasons of GLSA (i.e., in September 2016), no differences were found between MWD at ACR and GLSA at either of the two unproductive sites, but MWD was higher at GLSA at four out of six productive sites (Lussier 2018). However, my comparison to baseline data suggests that MWD at these unproductive sites improved in the first season of GLSA. Because unproductive GLSA had lower MWD than ACR at the beginning of the study in 2015, this improvement was not clear when ACR and GLSA were compared after the first year. As the soil at the ACRs continued to degrade, the difference between aggregate stability at GLSA and ACR at unproductive sites became perceptible and was significant by the third year of GLSA establishment (2018).  This result indicates that current management practices at unproductive ACR are not suitable for improvement of soil physical quality in the LFRD. Furthermore, while aggregate stability improved significantly at unproductive sites due to GLSA, aggregate stability was still lower than at productive sites, indicating that additional management practices used in conjunction with GLSA may be required to further improve aggregate stability at these sites. At productive sites, the improvements to aggregate stability caused by GLSA in the first year were apparent through comparisons of ACR and GLSA (Lussier et al. 2019), likely because baseline aggregate stability was similar at ACR and GLSA. At productive sites, current management practices paired with the inclusion of GLSA in rotation could likely maintain aggregate stability at acceptable levels. Given the state of land-tenure in the region, which limits management for long-term soil quality (Fraser 2004), these findings alone are compelling enough to support the continued use of GLSA in the LFRD. 55  Short-term GLSA can be an effective and practical tool for maintaining and improving soil aggregate stability at both productive and unproductive sites in the LFRD, even if only kept for one year. However, even after three and four years of GLSA, no improvements to carbon and nitrogen were found and improvements to bulk density and aeration porosity were not pronounced. After three and four years, GLSA could mitigate a reduction of aeration porosity and may play a role in reducing bulk density at 0–7.5 cm depth. These are positive findings; yet, it is likely that any improvements to bulk density due to GLSA would be reversed upon returning the field to crop production. I observed that bulk density and aeration porosity were within the potentially root-limiting zone in many cases, and therefore alternate management practices beyond GLSA (namely, the installation of sub-surface drainage systems) should be adopted to alleviate poor physical soil quality and associated drainage issues in the LFRD.  56  Chapter 4: General Conclusions and Recommendations for Future Research Grassland set-asides have been used in different contexts around the world and have typically led to improvements in soil quality parameters, including total carbon and soil physical properties. Farmers in the LFRD have been using short-term (i.e., 1–4 years) GLSA to address a variety of management needs, including diversifying crop rotation, transitioning fields to organic production, and improving degraded soils. My study sought to evaluate the effect of these short-term GLSAs on soil physical quality in the LFRD, both in relation to paired ACR fields and compared to the soil conditions just before GLSA establishment. The findings of my study will be of value to the DF&WT and provide information for farmers interested in or using the GLSA Stewardship Program in the LFRD.   4.1 General Conclusions After three and four years of GLSA, MWD and the proportion of total soil in the largest (i.e., 2–6 mm) aggregate size class were higher, while the proportion of aggregates in the smallest (<0.25 mm) class was lower in GLSA than ACR. This was true for both productive and unproductive sites. The analysis of aggregate stability over the four years of GLSA duration demonstrated that the improvements at productive and unproductive GLSA fields took place in the first year of GLSA and that aggregate stability remained the same during the additional three years. Despite the observed improvements of aggregate stability at unproductive sites, MWD did not reach the levels observed at productive sites, even after four years of GLSA. This finding furthers our understanding of the response to GLSA, particularly at unproductive sites, as previous research suggested one year of GLSA might not be long enough to improve aggregate stability at unproductive sites.  57  Other soil parameters did not show as clear a response to GLSA. After four years of GLSA, total soil carbon and nitrogen were higher under GLSA than ACR at productive and unproductive sites only at 30–45 cm depth. In addition, total soil carbon and nitrogen remained unchanged after four years of GLSA relative to the beginning of the study in 2015. Soil bulk density at 0–7.5 cm depth was lower in GLSA than ACR at both productive and unproductive sites after three years, but not after four years. Compared to soil bulk density in 2015 and 2016, bulk density at GLSA was significantly lower in 2018, but not 2019. In sum, the benefits to bulk density due to GLSA were not definitive, and it is likely that reductions to bulk density did not persist after the fields were returned to crop production.  Similarly, aeration porosity at 0–7.5 cm depth was higher in GLSA relative to ACR after three years but not after four years. These changes were observed only at productive sites. In addition, at productive sites, aeration porosity did not change from 2016 to 2019 at GLSA but was significantly lower in 2018 and 2019 compared to 2016 ACR. It is possible that aeration porosity improved relative to the conditions before this study. Unfortunately, aeration porosity was not determined in 2015. Nevertheless, a further decline in aeration porosity at productive sites was stalled due to GLSA, which is a positive outcome. Even so, GLSA alone is not enough to address issues of low aeration porosity.  In spring 2018 and spring 2019 at productive sites, aeration porosity was at or near potentially root-limiting levels at 7.5–15 and 15–30 cm depths. Meanwhile, at unproductive sites, aeration porosity and bulk density were at or near potentially root-limiting levels at all depths. Furthermore, I observed a decline in aeration porosity at unproductive ACR. Poor drainage and ponding in the spring lead farmers to cultivate when soils are wet, causing further damage to soil physical quality and pushing harvest later into the fall. In a potato field adjacent to an 58  unproductive ACR in my study, I observed a tractor get stuck in the wet soil as the farmer tried to harvest potatoes in the fall. He later informed me that they were unable to harvest that portion of the field and lost some of their potato crop. More must be done to address issues of compaction and drainage in the LFRD. My results indicate that three- and four-year GLSAs did not improve total soil carbon, total nitrogen, bulk density, or aeration porosity, but did result in improvements in aggregate stability. Therefore, improvement of soil physical quality, particularly at unproductive sites requires either longer-term GLSA and/or other changes to management practices. In the LFRD, the implementation of long-term GLSA is not a viable option to address soil quality concerns. Without substantial financial incentives, it is unlikely that long-term GLSA would be adopted in the region due to socioeconomic pressures such as insecure land tenure and the need for farmers to earn a living. Other options include management practices at the farm level such as drainage installation, laser-levelling, reduction of tillage, and organic matter additions. Policy changes should also be made to address land tenure issues or to further incentivize or compensate farmers for implementing practices that benefit long-term soil quality. Soil quality is an important component of ecosystem functioning and resilience in the face of climate change. Soil degradation is not an issue that exclusively impacts farmers, but one that affects society at large.  My study reveals that short-term GLSA alone cannot address all issues of soil quality. Nevertheless, my results support the inclusion of one-year GLSA in crop rotations in order to improve aggregate stability. Already, 35% of GLSA in the LFRD are kept for only one year, which represents the largest fraction by GLSA length (D. Bondar, Personal communication, Sept. 2020). While other goals and soil parameters not addressed in this study should be considered, from the lens of soil structure, it is likely preferable to keep GLSA for one year and 59  increase the frequency of GLSA in rotation, instead of maintaining GLSA for additional years. If one-year GLSAs are to be promoted over multi-year GLSAs, perhaps the funding that would have paid for the additional years of GLSA could be used instead to support supplementary practices that could enhance GLSA benefits, such as compost additions. Crop rotations in the region are short and not typically diverse, often dominated by potato production, which can be taxing on soil physical quality, including soil structure. Hence, GLSAs do play an important role in diversifying these short-term rotations and help improve and maintain soil structure in the LFRD.  4.2 Recommendations for Future Research My study demonstrates that short-term GLSA can improve aggregate stability in the LFRD. Aggregate stability at unproductive GLSAs did not reach the same levels observed at productive sites, indicating that additional interventions at unproductive sites were needed to improve aggregate stability. Studies in the region, including mine, have reported MWD values ranging from 0.57 mm to 3.2 mm; yet, the optimum MWD for soils in the LFRD remains unclear. This topic represents an area for future research. It is likely that improvements to aggregate stability, aeration porosity, and bulk density were limited by salinity issues, as sodium causes dispersion of soil particles and prevents flocculation. Preliminary thresholds for exchangeable sodium associated with reduced GLSA biomass after two growing seasons were developed for the LFRD (Lussier et al. 2019). Since high sodium may be preventing improvements to soil physical quality even after three and four years of GLSA, it is necessary to further refine these thresholds and determine whether sites with 60  high sodium would benefit from a more salt-tolerant GLSA mix, laser levelling, and/or drainage installation. While my analysis of above-ground GLSA biomass did not take into account different weed species, it could be useful to determine if specific weed species are associated with unproductive sites, as these may be used as indicators of soil quality, given that different weed species are tolerant to different conditions (such as water-logging or salinity).  My findings support the use of one-year GLSA to improve aggregate stability, but more research should look into the effects of GLSA on other soil quality parameters that were not addressed by this study, such as biological indicators of soil quality and active carbon pools, as these may also be affected by GLSA duration. Frequency of GLSA in crop rotations is another area that warrants future research, as is the effect of returning GLSA to ACR on all parameters included in this study. Wildlife goals should also be considered and effects of GLSA length on habitat use should be examined. It may be that other DF&WT programs, such as the Hedgerow and Grass Margin Stewardship Programs, can be used to provide habitat connectivity from year to year, and the effects of pairing one-year GLSA with these programs could also be evaluated.  The data set for total soil carbon and nitrogen from 2019 was limited. It is possible that a more robust data set could reveal significant improvements in total soil carbon and nitrogen after three and four years of GLSA. Active carbon pools are more responsive than total soil carbon and could also be examined.  My findings underscore the need to include comparisons to baseline in future studies of soil quality in response to management practices in the LFRD. Finally, further research should be conducted to determine a combination of management practices that addresses compaction and drainage issues in conjunction with GLSA in the LFRD.  61  References  Albrecht, M., Duelli, P., Schmid, B., and Müller, C. B. 2007. Interaction diversity within quantified insect food webs in restored and adjacent intensively managed meadows. J. Anim. Ecol. 76(5): 1015–1025.   Angers, D.A., and J. Caron. 1998. Plant-induced change in soil structure: processes and feedbacks. Biochemistry. 42: 55–72.   Aviron, S., Nitsch, H., Jeanneret, P., Buholzer, S., Luka, H., Pfiffner, L., and Herzog, F. 2009. Ecological cross compliance promotes farmland biodiversity in Switzerland. Front. Ecol. Environ. 7(5): 247–252.   Baer, S.G., Rice, C.W., and Blair, J.M. 2000. Assessment of soil quality in fields with short and long term enrollment in the CRP. J. Soil Water Conserv. 55: 142–146.  Bertrand, R.A., Hughes-Games, G., and Nikkel, D.C. 1991. Soil management handbook for the Lower Fraser Valley. 129 pp.  Blake, G.R., and Hartge, K.H. 1986. Bulk density. Pages 363–375 in C.A. Black, ed. Methods of soil analysis, Part 1: physical and mineralogical methods. Soil Science Society of America, Madison, WI, USA  Bolinder M.A., Angers D.A., Bélanger, G. and Michaud, R. 2002. Root biomass and shoot to root ratios of perennial forage crops in eastern Canada. Can. J. Plant Sci. 82(4): 731–737. doi:10.4141/P01-139.  Boundary Bay Conservation Committee (BBCC). 2008. History. [Online]. Available: http://www.sunburyneighbourhood.ca/BBCC/History.html [19 Jan. 2020].  Bowman, R.A., and Anderson, R.L. 2002. Conservation Reserve Program: effects on soil organic carbon and preservation when converting back to cropland in northeastern Colorado. J. Soil Water Conserv. 57: 121–126.  Bregha & Moffet. 1995. Replenishing the prairies: the Canadian permanent cover programme. Pages 105-112 in R. Gale, S. Bargs, and A Gillies, eds. Green budget reform. Earthscan Publications Ltd., London.  British Columbia Ministry of Agriculture. 2012. British Columbia vegetable production guide. [Online]. Available: http://productionguide.agrifoodbc.ca/guides/17 [2020 Aug. 30].  Brown, R., Percivalle, C., Narkiewicz, S., and DeCuollo, S. 2010. Relative rooting depths of native grasses and amenity grasses with potential for use on roadsides in New England. HortScience. 45(3): 393–400.  62   Caron, J., and Kay, B.D. 1992. Rate of response of structural stability to a change in water content: influence of cropping history. Soil Tillage Res. 25(2), 167–185. doi:10.1016/0167-1987(92)90109-O.  Chow T.L., and Rees, H.W. 1994. Effects of potato hilling on water runoff and soil erosion under simulated rainfall. Can. J. Soil Sci. 74: 453–460.  Clarke, J. 1992. Set-aside. British Crop Production Council, Farnham, Surrey, UK.  Danielson, R.E., and Sutherland, P.L. 1986. Porosity. Pages 443–462 in A. Klute, ed. Methods of soil analysis, Part 1: physical and mineralogical methods. Soil Science Society of America, Madison, WI, USA.  Degens, B.P. 1997. Macro-aggregation of soils by biological bonding and binding mechanisms and the factors affecting these: a review. Soil Res. 35(3): 431. doi:10.1071/S96016.  Delta Farmland & Wildlife Trust. 2017a. History of farming in Delta. [Online]. Available: https://deltafarmland.ca/resources/history-of-farming-in-delta/ [25 Sept. 2020].  Delta Farmland & Wildlife Trust. 2017b. Supporting local farms that feed families and the birds! [Online]. Available: https://deltafarmland.ca/ [30 Sept. 2020].  Delta Farmland & Wildlife Trust. 2019. Grassland set-asides: program overview & guidelines. [Online]. Available: https://deltafarmland.ca/wp-content/uploads/2019/03/Fact-Sheet-GLSA-2019.pdf [25 Sept. 2020].  Dexter, A.R. 1991. Amelioration of soil by natural processes. Soil Tillage Res. 20(1): 87–100. doi:10.1016/0167-1987(91)90127-J.  Dunn, C.P., Stearns, F., Guntenspergen, G. R., & Sharpe, D. M. 1993. Ecological benefits of the conservation reserve program. Conserv. Biol. 7(1): 132–139.   Environment and Climate Change Canada (no date a). Canadian climate normals 1981-2010 station data. [Online]. Available: https://climate.weather.gc.ca/climate_normals/results_1981_2010_e.html?searchType=stnName&txtStationName=delta&searchMethod=contains&txtCentralLatMin=0&txtCentralLatSec=0&txtCentralLongMin=0&txtCentralLongSec=0&stnID=766&dispBack=0 [2018 Sept. 30].  Environment and Climate Change Canada (no date b). Canadian climate normals 1981-2010 station data. [Online]. Available: https://climate.weather.gc.ca/climate_normals/results_1981_2010_e.html?searchType=stnProv&lstProvince=BC&txtCentralLatMin=0&txtCentralLatSec=0&txtCentralLongMin=0&txtCentralLongSec=0&stnID=889&dispBack=0 [2018 Sept. 30]. 63    Environment Canada and Canadian Wildlife Services. 1998. [Online]. Available: http://www.sccp.ca/sites/default/files/species-habitat/documents/Fraser%20Historical%20Vegetation%20-%20Lower%20Mainland%20fraser%20serp%20nick%20%20-%2028Oct1998_0.pdf [24 Nov. 2020].  European Commission. 2004. Summary: set-aside. [Online]. Available: http://europa.eu/rapid/press-release_MEMO-93-43_en.htm [18 Jan. 2020].  Fairbrother C, Middler, J. and Waldie, L. 2006. Habitat use and abundance trends of wintering raptors in the Fraser delta 2005-2006. Diploma report, British Columbia Institute of Technology, Burnaby, BC. 29 pp.  Fausak, L.K. 2019. The effects of 2- and 3-year grassland set-asides on plant available nitrogen and greenhouse gas emissions in Delta, British Columbia. MSc. thesis, University of British Columbia, Vancouver, BC. 90 pp.  Fraser, E.D.G. 2004. Land tenure and agricultural management: soil conservation on rented and owned fields in southwest British Columbia. Agric. Hum. Values, 21:73–79. doi:10.1023/B:AHUM.0000014020.96820.a1.  Gebhart, D.L., Johnson, H.B. Mayeux, H.S., and Polleym, H.W. 1994. The CRP increases soil organic carbon. J. Soil Water Conserv. 49: 488–492.   Greenland, D.J. 1981. Soil management and soil degradation. J. Soil Sci. 32: 301-322.  Guo, L.B., and Gifford, R.M. 2002. Soil carbon stocks and land use change: a meta analysis. Glob. Chang. Biol. 8: 345–360. doi:10.1046/j.1354-1013.2002.00486.x.  Hermawan, B., and Bomke, A.A. 1996. Aggregation of a degraded lowland soil during restoration with different cropping and drainage regimes. Soil Technol. 9(4): 239–250.   Hermawan, B., and Bomke, A.A. 1997. Effects of winter cover crops and successive spring tillage on soil aggregation. Soil Tillage Res., 44: 109–120. doi:10.1016/S0167-1987(97)00043-3.  Jackson, R.B., Canadell, J., Ehleringer, J.R., Mooney, H.A., Sala, O.E., and Schulze, E.D. 1996. A global analysis of root distributions for terrestrial biomes. Oecologia 108:389–411.  Karlen, D.L., Rosek, M.J., Gardner, J.C., and Allan, D.L. 1999. Conservation Reserve Program effects on soil quality indicators. J. Soil Water Conserv. 54(1): 439–444.    64  Kennedy, G. 2017. Canada 150: Delta farmers institute changing with the times. North Delta reporter. [Online]. Available: https://www.northdeltareporter.com/community/canada-150-delta-farmers-institute-changing-with-the-times/ [23 Sept. 2020].  Kleijn, D., and Báldi, A. 2005. Effects of set-aside land on farmland biodiversity: comments on Van Buskirk and Willi. Conserv. Biol. 19(3): 963–966.   Krzic, M., Bondar, D., Avery, E., and Smukler, S.M. 2020. Evaluating the benefits of short-term grassland set-asides on Delta farmland. Final report to Investment Agriculture Foundation of BC for the project: demonstrating long-term improvements in soil productivity on Delta farmland grassland set-asides. 67 pp.  Krzic, M., Fortin, M.-C., and Bomke, A.A. 2000. Short-term responses of soil physical properties to corn tillage-planting systems in a humid maritime climate. Soil Tillage Res. 54: 171–178. doi:10.1016/S0167-1987(00)00092-1.  Lecain, D.R., Morgan, J.A., Milchunas, D.G., Mosier, A.R., Nelson, J.A., and Smith, D.P. 2006. Root biomass of individual species, and root size characteristics after five years of CO2 enrichment on native shortgrass steppe. Plant Soil. 279: 219–228. doi:10.1007/s11104-005-2301-9.  Lehmann A., Zheng W., and Rillig, M. C. 2017. Soil biota contributions to soil aggregation. Nat. Ecol. Evol. 11: 1828–1835. doi:10.1038/s41559-017-0344-y.  Litalien, A., and Zeeb, B. 2020. Curing the earth: a review of anthropogenic soil salinization and plant-based strategies for sustainable mitigation. Sci. Total Environ. 698: 134235. doi:10.1016/j.scitotenv.2019.134235  Liu, A., Ma, B.L., and Bomke, A.A. 2005. Effects of cover crops on soil aggregate stability, total organic carbon, and polysaccharides. Soil Sci. Soc. Am. J. 69(6): 2041–2048. doi:10.2136/sssaj2005.0032.  Lussier, J.M. 2018. The effects of short-term grassland set-asides on soil properties in the Fraser River delta of British Columbia. MSc. thesis, University of British Columbia, Vancouver, BC. 101 pp.   Lussier, J.M., Krzic, M., Smukler, S.M., Bomke, A.A., and Bondar, D. 2019. Short-term effects of grassland set-asides on soil properties in the Fraser River delta of British Columbia. Can. J. Soil Sci. 99: 136–145. doi:10.1139/cjss-2018-0097.   Lussier, J.M., M. Krzic, S.M. Smukler, K. Neufeld, C. Chizen, and A.A. Bomke. 2020. Soil organic carbon and aggregate stability in two seasons of grassland set-aside and annual crop rotations. Soil Res. 58(4): 364–370. doi:10.1071/SR19180.    65  Luttmerding, H.A. 1981. Soils of the Langley-Vancouver map area. RAB Bulletin 18. Vol. 3. BC Ministry of Environment, Kelowna, BC, Canada. 227 pp.  Merkens, M. 2004. Partners in stewardship: an example of using incentives for agricultural wildlife habitat enhancement. Pages 1–12 in T. D. Hooper, ed. Species at risk 2004 pathways to recovery conference. Victoria, BC. Available: https://deltafarmland.ca/wp-content/uploads/2020/02/Partners-in-Stewardship-An-example-of-using-incentives-for-agricultural-wildlife-habitat-enhancement.pdf [26 Sept. 2020].  McNeil, R. 2013. Conversion of cultivated lands to native perennialsin the parkland. Alberta Agriculture and Rural Development. [Online]. Available: http://www.albertapcf.org/rsu_docs/parkland-native-conversion-landwise-28feb-2013.pdf [20 Jan. 2020].  National Soil Conservation Program. (no date). National Soil Conservation Program in Ontario: final report. [Online]. Available: https://atrium.lib.uoguelph.ca/xmlui/bitstream/handle/10214/14378/nscp_final_rep.pdf?sequence=1&isAllowed=y [20 Jan. 2020].  Nelson, D. W., and Sommers, L. E. 1982. Total carbon, organic carbon and organic matter. Pages 961-1010 in A. L. Page, ed. Methods of soil analysis. Part 3: chemical methods. Soil Science Society of America, Madison, WI.  Nimmo, J.R., and Perkins, K.S. 2002. Aggregate stability and size distribution. Pages 317–328 in J.H. Dane and G.C. Topp, eds. Methods of soil analysis. Part 4: physical methods. Soil Science Society of America, Madison, WI, USA.  Osborn T. 1993. Soil and Water Conservation Society. J. Soil Water Conserv. 48(4):271-279.  Paul, C.L., and de Vries, J. 1979. Effect of soil water status and strength on trafficability. Can. J. Soil Sci. 59:313–324. doi:10.4141/cjss79-035.  Post, W.M., and Kwon, K.C. 2000. Soil carbon sequestration and land-use change: processes and potential. Global Change Biol. 6(3): 317–327. doi: 10.1046/j.1365- 2486.2000.00308.x.  Pranagal, J., Podstawka-Chmielewska, E., and Slowinska-Jurkiewicz, A. 2007. Influence on selected physical properties of a haplic podzol during a ten-year fallow period. Pol. J. of Environ. Stud. 16:875–880.  Principe, L. L. 2001. Plant species abundance, diversity and soil quality of grassland set-asides on the Fraser River delta. MSc. thesis, University of British Columbia, Vancouver, BC. 119 pp. doi:10.14288/1.0090271  R Core Team 2019. R: A language and environment for statistical computing. Vienna, Austria. [Online] Available: https://www.r-project.org/ [20 Apr. 2020]. 66   Reeves, J.B., G.W. McCarthy, and V.B. Reeves. 2001. Mid-infrared diffuse reflectance spectroscopy for the quantitative analysis of agricultural soils. J. Agric. Food Chem. 49: 766–772. http://doi.org/10.1021/jf001123.   Riley, H., Pommeresche, R., Eltun, R., Hansen, S., and Korsaeth, A. 2008. Soil structure, organic matter and earthworm activity in a comparison of cropping systems with contrasting tillage, rotations, fertilizer levels and manure use. Agric. Ecosyst. Environ. 124: 275–284. doi: 10.1016/j.agee.2007.11.002.  Rosenzweig, S.T., Carson, M.A., Baer, S.G., and Blair, J.M. 2016. Changes in soil properties, microbial biomass, and fluxes of C and N in soil following post-agricultural grassland restoration. Appl. Soil Ecol. 100: 186-194. doi:10.1016/j.apsoil.2016.01.001  Species at Risk Partnership on Agricultural Lands. 2018. Grassland stewardship program. [Online]. Available: https://www.ontariosoilcrop.org/wp-content/uploads/2017/12/GSP-Brochure-2018-Digital-1.pdf [16 Jan. 2020].  Staben, M.L., Bezdicek, D.F., Fauci, M.F., and Smith, J.L. 1997. Assessment of soil quality in Conservation Reserve Program and wheat-fallow soils. Soil Sci. Soc. Am. J. 61(1): 124–130. doi: 10.2136/sssaj1997.03615995006100010019x.  Statistics Canada (2016a). Table 32-10-0436-01 Farms classified by total gross farm receipts in the year prior to the census. doi: 10.25318/3210043601-eng.  Statistics Canada (2016b). Table 32-10-0406-01 Land Use. doi: 10.25318/3210040601-eng.  Statistics Canada (2016c). Table 32-10-0418-01 Vegetables (excluding greenhouse vegetables). doi: 10.25318/3210041801-eng.  Statistics Canada (2016d). Table 32-10-0416-01 Hay and field crops. doi:10.25318/3210041601-eng.  Statistics Canada (2016e). Table 32-10-0407-01 Tenure of land owned, leased, rented, crop-shared, used through other arrangements or used by others. doi:10.25318/3210040701-eng.  Steves, H. 2007. A commitment to the future — 2007: a proposal for the protection & management of Richmond and Delta farmland and wetlands. [Online]. Available:  http://www.richmond.ca/__shared/assets/Steves18601.pdf [25 Sept. 2020].  Thiel, B., Oka, G., Radley, M., and Smukler, S. 2015. Climate change adaptation and on-farm drainage management in Delta, British Columbia: current knowledge and practices. [Online] Available: https://www.bcagclimateaction.ca/wp/wp-content/media/DL09-Delta-Drainage-Sub-irrigation-full.pdf [23 Jan. 2020].  67  Tiessen, K.H.D., Lobb, D.A., Mehuys, G.R., and Rees, H.W. 2007. Tillage translocation and tillage erosivity by planting, hilling and harvesting operations common to potato production in Atlantic Canada. Soil Tillage Res. 97(2): 123–139. doi: 10.1016/j.still.2007.09.005  Tisdall J.M., and Oades J.M. 1982. Organic matter and water‐stable aggregates in soils. J. Soil Sci., 33: 141-163. doi:10.1111/j.1365-2389.1982.tb01755.x  Tscharntke, T., Batáry, P., and Dormann, C. F. 2011. Set-aside management: how do succession, sowing patterns and landscape context affect biodiversity? Agric. Ecosyst. Envir. 143(1): 37–44. doi:10.1016/j.agee.2010.11.025  United States Department of Agriculture. 2019. Conservation reserve program fact sheet. [Online]. Available: https://www.fsa.usda.gov/Assets/USDA-FSA-Public/usdafiles/FactSheets/2019/conservation-reserve_program-fact_sheet.pdf [18 Jan. 2020].  Vos, J., Van Loon, C.D., and Bollen, G.J. (eds.). 1989. Effects of crop rotation on potato production in temperate zones. Kluwer Academic Publishers, Dordrecht, Netherlands.  Vancouver Fraser Port Authority. 2018. Roberts Bank terminal 2 project: about the project. [Online]. Available: http://www.robertsbankterminal2.com/about-the-project/project-overview/  [23 Sept. 2020].  Walji, K., Krzic, M., Bondar, D. and Smukler S.M. 2020. Nitrogen dynamics following incorporation of 3-year old grassland set-asides in the Fraser River delta of British Columbia. Agronomy. 10(9): 1382.  Wallace, B.M., Krzic, M., Forge, T.A., Broersma, K., and Newman, R.F. 2009. Biosolids increase soil aggregation and protection of soil carbon five years after application on a crested wheatgrass pasture. J. Environ. Qual. 38(1): 291–298. doi:10.2134/jeq2007.0608.  Western Producer. 1994. 29 Sept. Future of Delta farmland under review. Delta, B.C. [Online]. Available: https://www.producer.com/1994/09/future-of-delta-farmland-under-review/ [5 July 2020].  Yates, D.E. 2014. Effects of grassland set-asides on selected soil properties in the Fraser River Delta of British Columbia. MSc. thesis, University of British Columbia, Vancouver, BC. 97 pp.  Yates, D.E., Krzic, M., Smukler, S.M., Bradfield, G., Bomke, A.A., and Terpsma, C. 2017. Comparison of selected soil properties following grassland set-aside and annual rotations in the Fraser River delta of British Columbia. Can. J. Soil Sci. 97(4): 783-788. 68  Appendices  Appendix A  Crop history on eight study sites. Site 7 was not part of my study but was included in the larger study initiated in 2015 and is included here. PT=Potatoes; P= Peas; GLSA=Grassland set-aside; B= Barley; W=Wheat; BN= Beans; C= Corn; T= Turnips, F=Fallow.   Site 1 Site 2 Site 3 Site 4 Site 5 Site 7 Site 8 Site 9 Season GLSA Crop GLSA Crop GLSA Crop GLSA Crop GLSA Crop GLSA Crop GLSA Crop GLSA Crop 2018 - - GLSA C GLSA PT GLSA T GLSA C - - GLSA P/(B) - - 2017 GLSA PT GLSA C GLSA PT GLSA PT GLSA BN - - GLSA PT GLSA C 2016 GLSA PT GLSA C GLSA B GLSA P GLSA PT GLSA BN GLSA B GLSA PT 2015* GLSA B GLSA B GLSA B GLSA PT GLSA F GLSA BN GLSA BN GLSA P 2014 PT PT B P PT PT B B P P PT PT BN BN P P 2013 PT PT PT W P P BN BN C C BN BN C C P PT 2012 B B B PT PT PT BN BN C C PT PT P P PT P 2011 GLSA PT P BN PT PT BN BN C C BN BN PT PT PT PT *Larger GLSA study was initiated in the spring of 201569  Appendix B  Soil baseline properties of fields entering the larger study in spring 2015. All properties were determined at the 0-15 cm depth with the exception of mean weight diameter (MWD) and bulk density (determined at 0-7.5 cm depth), and Na+ (determined at 0–30 cm depth). Source: Lussier 2018.   Site Treatment Bulk density (Mg m-3) MWD (mm) Total C (%) Total N (%) EC (S/m) Na+  (cmolc kg-1) CEC  (cmolc kg-1) 1 GLSA 1.23 1.18 1.96 0.17 133.60 0.07 12.57  ACR 1.19 1.28 2.14 0.19 166.63 0.08 15.82 2 GLSA 1.21 1.00 2.08 0.18 213.18 0.33 13.23  ACR 1.24 0.46 1.32 0.11 137.48 0.10 15.99 3 GLSA 1.32 0.45 1.92 - 677.00 1.78 15.82  ACR 1.28 0.69 1.70 - 201.03 0.27 14.52 4 GLSA 1.21 0.55 2.02 - 784.75 2.59 18.07  ACR 1.28 0.92 1.74 - 166.60 0.40 17.74 5 GLSA 1.18 1.71 2.73 0.26 273.60 0.85 13.78  ACR 1.01 1.69 3.02 0.27 301.70 1.14 11.34 7 GLSA 1.25 1.06 1.63 0.15 154.98 0.06 14.18  ACR 1.33 0.96 1.49 0.13 357.28 0.66 14.28 8 GLSA 1.14 0.78 2.67 0.22 246.25 0.20 13.25  ACR 1.07 1.32 3.15 0.25 280.38 0.17 11.96 9 GLSA 1.20 0.81 3.07 0.29 143.60 0.13 16.64  ACR 1.20 1.18 2.93 0.27 131.10 0.14 12.37 70  Appendix C  Percentage of total dry biomass by vegetation group (grass, clover, and weeds/other) in productive and unproductive grassland set-asides (GLSA) at each study site. 71  Appendix D  Average temperature, rainfall, snowfall, and precipitation in the LFRD from September 2014 to April 2019. Data is from the Delta - Tsawwassen Beach weather station (Environment and Climate Change Canada no date a).     Average Temperature (°C)                    Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec historical average (1981-2010) 5.1 5.8 7.7 10.4 13.4 16 17.9 17.9 15.3 11.2 7.5 5.1 2014         16.5 13.2 6.7 5.7 2015 6.3 7.9 8.4 10.4 15.1 18.3 19.1 18.4 14.2 12.1 5.9 5.3 2016 5.1 7.2 8.5 11.5 13.5 15.6 17.7 17.9 13.9 10.4 8.8 1.5 2017 2.1 3.1 6.4 9 12.7 14.9 17.4 17.7 15 9.5 6.5 2.4 2018 4.6 3.3 5.8 8.5 13.9 14.7 17.8 17.2 13.3 9.1 7 4.4 2019 5.1 2.5 6.6 12.3 15.5                             Rainfall (mm)                      Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec historical average (1981-2010) 124.3 77.1 77 67.9 52.2 42.6 30.5 28.7 39.8 101 142.7 116.3 2014           186.7 126.2 130.7 2015 133 70.6 66.6 40.4 4.2 7.4 12.6 48.2 27.6 101.3 106 194.7 2016 117 138.3 98.2 13.4 11.2 35.4 18 9.6 55.2 194.9 181 113.7 2017 65 76.5 181.7 111.3 66.6 37.4 0.4 5.2 37.8 83.1 133.8 173.1 2018 184.5 86.5 81.6 101.5 0.8 29.6 7.4 5.4 116.6 90.4 165.4 135.6 2019 137.3 44.8 23.4 95.4 39          72    Snowfall (cm)             Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec historical average (1981-2010) 10.4 3.3 1.6 0 0 0 0 0 0 0.3 2.4 9.6 2014         0 0 1 0 2015 0 0 0 0 0 0 0 0 0 0 0 0 2016 0 0 0 0 0 0 0 0 0 0 0 12.4 2017 0 26 4 0 0 0 0 0 0 0 0 0 2018 0 19.4 0 0 0 0 0 0 0 0 0 0 2019 0 20 0 0 0                        Precipitation (mm)            Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec historical average (1981-2010) 134.6 80.4 78.5 67.9 52.2 42.6 30.5 28.7 39.8 101.3 145.1 125.9 2014         55.2 186.7 127.2 130.7 2015 133 70.6 66.6 40.4 4.2 7.4 12.6 48.2 27.6 101.3 106 194.7 2016 117 138.3 98.2 13.4 11.2 35.4 18 9.6 55.2 194.9 181 126.1 2017 65 102.5 185.7 111.3 66.6 37.4 0.4 5.2 37.8 83.1 133.8 173.1 2018 184.5 105.9 81.6 101.5 0.8 29.6 7.4 5.4 116.6 90.4 165.4 135.6 2019 137.3 64.8 23.4 95.4 39         73  Appendix E  Total carbon at productive annual crop rotation (ACR) and grassland set-aside (GLSA) fields in spring 2015 and 2019 at 0–15, 15–30, 30–45, and 45–60 cm depths. No significant differences were found (α = 0.10).  74  Appendix F  Total nitrogen at productive annual crop rotation (ACR) and grassland set-aside (GLSA) fields in spring 2015 and 2019 at 0–15, 15–30, 30–45, and 45–60 cm depths. No significant differences were found (α = 0.10).  75  Appendix G  C:N ratio at productive annual crop rotation (ACR) and grassland set-aside (GLSA) fields in spring 2015 and 2019 at 0–15, 15–30, 30–45, and 45–60 cm depths. No significant differences were found (α = 0.10).  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0395492/manifest

Comment

Related Items