Characterizing weed communities, management practices, environmental factors, and their association with snap bean (Phaseolus vulgaris) production across the United States

Pavlovic, Pavle

Permalink

https://hdl.handle.net/2142/124572 Copy

Description

Title

Characterizing weed communities, management practices, environmental factors, and their association with snap bean (Phaseolus vulgaris) production across the United States

Author(s)

Pavlovic, Pavle

Issue Date

2024-05-01

Director of Research (if dissertation) or Advisor (if thesis)

Williams, II, Martin M

Committee Member(s)

• Hager, Aaron G
• Martin, Nicolas F

Department of Study

Crop Sciences

Discipline

Crop Sciences

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

M.S.

Degree Level

Thesis

Keyword(s)

• snap bean
• Phaseolus vulgaris
• weed survey
• United States
• machine learning

Abstract

Snap bean has been a very popular vegetable in many dishes in the United States for well over a century. The most popular is casserole, which became a staple at family gatherings on Thanksgiving Day. Snap bean is a popular vegetable in many household gardens across the United States. There is also a substantial large-scale production which usually goes to the processing industry. There have been numerous studies conducted on snap bean both in the agronomic and biological sense. Even from the aspect of weed science snap bean is not lacking and plenty of information has been gathered over the decades. However, when it comes to looking at snap bean from a weed science perspective on a large production scale, the information is severely lacking. To fill this knowledge gap, a massive survey was conducted across the major U.S. production regions (Northeast, Midwest, Northwest) where snap bean is grown for the processing industry. Within five years, between 2019 and 2023, a total of 358 production fields have been surveyed. The data obtained from the surveys was used in compiling the chapters of the thesis, which are all first reports of this kind, providing valuable insight into snap bean production from a weed science perspective. Chapter 1 of the thesis is a literature review divided into three major sections. The first section is an overview of snap bean as both a plant and a crop, the second section highlights the important weed science research conducted in the crop, while the third section gives a brief overview of weed surveys and a nod to many surveys conducted in North America and other parts of the world. Chapter 2 focuses on the data collected regarding residual weed community structure in snap bean production fields. The results demonstrated that snap bean production fields were not particularly “weedy” and mostly constituted of plants that were not that big in stature and therefore were not particularly competitive. Nevertheless, there were fields with very substantial residual weed populations, that were very large in stature. In addition, the three weed species/groups that have been commonly observed across the surveys were common lambsquarters (Chenopodium album L.), amaranth species (Amaranthus spp. L.), and large crabgrass [Digitaria sanguinalis (L.) Scop.] are all problematic weeds in many other crops and are very adaptable, which is a cause for concern in snap bean production going forward. Chapter 3 focuses on the extent of not only weed management but also agronomic practices employed in snap bean across the production regions. This data was obtained at the discretion of collaborating growers and demonstrated that most of the practices employed followed recommended guidelines when it came to both agronomy and weed management. However, there was also novel insight when it comes to the distribution of varieties, planting dates, and rotating crops, as well as differences noted in management practices between regions, especially in the Northwest. While the results of the weed surveys from Chapter 2 demonstrate that the residual weed population wasn’t that abundant, implying that weed control measures were utilized effectively in the early growing season, there is a concerning lack of diversity when it comes to herbicides used which could be a problem in future snap bean production. Chapter 4 focuses on synthesizing all the different weed, management, and environmental data with machine learning algorithms to better understand which factors have the most influence on snap bean production and the associated weed community. The results of the analysis demonstrated that environmental factors (including weed competition) have a key role in influencing snap bean yield. Most management factors don’t have a significant effect on yield except for planting date and row cultivation. Utilization of row cultivation can cause a decrease in yield, possibly because of crop injury in early season growth. However, at the same time row cultivation was the most important factor in influencing weed population density, as the fields where it was utilized had on average lower weed densities compared to those where it wasn’t.

Graduation Semester

2024-05

Type of Resource

Thesis

Copyright and License Information

Copyright 2024 Pavle Pavlovic

Owning Collections