The word model is a loaded term that can mean a multitude of things depending on the nature of the problem and the application domain. In this dissertation, model refers to a representation model, i.e., one that is a “representation of a selected part of the world ”, as opposed to models that represent sets of laws and axioms that constitute a theory. Two categories of representational models exist: those that represent the phenomena occurring within complex systems and those that represent the data constituting the states within said systems. A datapoint can be thought of as an observation or measurement at a particular point in time and space of a particular phenomenon. Examples of this data-phenomena relationship include: observing a person’s mood [data] over the course of a day to investigate effects of fatigue [phenomena], measuring temperature [data] to predict the weather [phenomena], or monitoring energy consumption by a computer server [data] to dynamically balance power loads [phenomena]. The real world, however, is not always neatly separable into phenomena and data.The General Modeling Process: Models are often used to perform a particular analysis to produce a result, such as the identification of trends, patterns, or generalizations that can then be used to understand or improve the original system. Ideally, grow room design the results of a modeling exercise lead to the production of knowledge that can help answer related questions; create, improve or understand similar systems; and add to the overall understanding of the real world.
In response to growing concerns regarding the sustainability of human civilization, rising effects of climate change, and worldwide resource inequalities , researchers have developed a variety environmental impact assessment techniques to reduce and manage the environmental footprint of human-made systems. One technique in particular, Life Cycle Assessment , is used to investigate the environmental impacts of industrial products and production systems. Major issues such as water pollution, greenhouse gas emissions, and eutrophication tend to be a result of resource intensive processes such as the extraction of raw materials, consumption of materials and energy, transformation of one material into another, and the manufacture of products. LCA involves the quantification of these resource flows in a system to calculate the environmental impacts that the system incurs. Models of agricultural products and production systems are commonly created using LCA. Farmers, along with environmental analysts, can conduct LCAs to quantify environmental impacts, and subsequently make improvements in the farming processes to reduce undesired impacts. Interest in tailoring LCA for agricultural systems has led to the identification of many substantive problems at the intersection of agricultural modeling and environmental impact assessment. This has led to the development of many new modeling techniques, creation of software tools, and interdisciplinary collaborations. As a result, myriads of models are produced to address these problems. The creation of LCA models is often money and time intensive, requiring input from many stakeholders, large quantities of data to be collected, and analyses to be conducted by environmental experts trained in LCA methods and tools.
The resulting LCA models are instrumental to understanding and improving the environmental performance of the agricultural system of interest. These LCA models are therefore highly place and time specific: they often represent a single product system in a particular geographic region using data collected during a certain period of time. By the time the environmental impacts are calculated and translated into meaningful recommendations for a change in the real world system, the models may be out of sync with the real world. This is not conducive to making recommendations regarding the assessment, design, or optimization of other similar systems.LCA is a technique used to assess the environmental impacts of products and the processes by which they are constructed. The system of interest typically involves the development of a particular product: a farm where tomatoes are grown, a factory where shoes are made, or a wastewater treatment plant where clean water is produced. There exist numerous LCA variants. Each has a different purpose, system modeling technique, domain applicability, and data requirements. Standardization: The generic LCA process is governed by a suite of standards: ISO 14040 details the principles and overall framework, and ISO 14044 specifies the requirements and guidelines for conducting an LCA. These standards are part of a greater family, ISO 14000, regarding Environmental Management, that provide guidance on environmental assessment, performance, and responsibility. ISO standards are reviewed every five years: the most recent versions of ISO 14040 and ISO 14044 were published in 2006 and reviewed in 2010. These reviews aim to maintain the capacity of LCA models to represent systems resulting from advances in the real world. Not all LCA studies are ISO compliant.
However, compliance tends to occur in government funded LCAs, when analysts intend for their data to be incorporated into national LCA databases , some LCA research, and certain LCA studies conducted for companies and organizations. LCA standards are written in a domain agnostic manner, enabling for LCAs conducted in any domain to potentially be ISO compliant. For instance, an LCA of a beef producing agricultural system, plastics and metals recycling, and a printer cartridge can all be ISO compliant, as long as they adhere to the requirements and guidelines laid out in ISO 14040. For the rest of this section, I will briefly overview the generic LCA process. Figure 3.1 outlines the main phases of an LCA: Goal Definition and Scope, Inventory Analysis, Impact Assessment, and Interpretation. There are varying techniques for conducting both the overall LCA, as well as individual phases.Define the purpose of the LCA study: One is describing a problem in the real world: what environmental impacts are incurred by producing specific quantities of a product, how great are the impacts , and what are the causes? Therefore, the goal is to assess the environmental footprint of the system. Alternatively, the goal could be to compare the environmental performance of competing systems. Depending on the problem domain, reasons for the analysis being conducted, and the familiarity of the analyst with the domain, the analyst may, from experience, know which environmental impacts can and should be calculated. There exist environmental impact categories that can be selected as specific areas of interest, but they are not always selected during Phase One, as they are only used in Phase Three when impact calculation occurs. Define system boundaries and scope: The production systems and stages to be included in the system of interest, and therefore the assessment, must be selected. For example, a cradle-to-grave LCA includes everything that went into the manufacturing of the final product, starting from the extraction of raw materials, to the disposal of the product after use. It takes into account the full life cycle of a product. This also requires specifying the time scale of the analysis: for example, the life cycle of a solar panel includes 25 years for the operational/use phase, in addition to the time spent in the manufacture and disposal of the product. In contrast to the solar panel, the use phase of a cup of coffee depends on how it is made and consumed. The undercradle-to-grave LCA stages for which raw materials and energy are accounted for are: raw material acquisition,transportation, grow racks manufacturing processes, the use, reuse, and maintenance of the product, and finally, recycling and waste management. Alternatively, a cradle-to-gate LCA, accounts for everything up until the product is ready to be transported to the customer. This method of scoping an LCA study works particularly well for analyzing production systems for physical products with a well-defined life cycle, clear-cut inputs and outputs, and a limited set of uses. More recently, it has been suggested that life cycle considerations should extend from “cradle-to-cradle”, taking into account both the production, but eventual disposal, and potential reuse of physical products. A decision must also be made as to how many levels of indirection from the core system are to be included in the study. Attributional LCA studies focus on accounting only the immediate and direct impacts of the system of interest. This has been the traditional approach to LCA, where one quantifies the effects of the system as it currently is.The Life Cycle Inventory Analysis phase involves gathering a list of all the processes occurring within the system, along with the material and energy resources utilized, useful products created, and waste produced. In this phase, the system of interest is decomposed into a set of discrete processes, with inputs, outputs, and quantifiable flows of material and energy resources through each process. A comprehensive LCA relies on the ability of the analyst to find the environmental impact of each process within each subsystem during the LCI Analysis phase. This approach is known as process LCA. Alternatively, in an input-output LCA, the system is treated as a whole, with aggregate inputs and outputs into the system. One of the most popular variants is the economic inputoutput LCA: it uses aggregate economic inputs and outputs into the system of interest to estimate environmental impacts.
It is an application of a popular economic modeling approach that is used to assess the economic performance of a system. This phase has two deliverables. The first is a flow diagram, a high level overview of the main processes or components in the system. The second is the Life Cycle Inventory itself. The LCI is an all-encompassing inventory of inputs, outputs, and processes within the system of interest. Process and input-output LCA represent two different approaches to abstracting the system of interest into a model with quantitative and qualitative attributes. The level of detail in an LCA study also varies based on how much time, effort, and money the analyst has to put in, how much data is actually available or collectable, and what the results will be used for. If the level of detail is low, and the system of interest has been simplified in its representation, the study is labeled as a streamline LCA. Otherwise, every process of every stage in the scope of the system of interest must be represented in the LCI. The LCA ISO 14044 standard specifies mandatory elements within this phase: the selection of impact categories, category indicators, and characterization models; classification ; and characterization . This produces a set of Category Indicator Results, which collectively are known as the LCIA Profile, and where each result represents the actual environmental impacts of the system of interest by associating each of the processes in the LCI with the impact categories affected by the process. Impact categories may include: eutrophication, climate change, global warming potential , land use, water use. There has also been growing interest in conducting more holistic “sustainability” LCAs that include assessment with respect to economic, social, and environmental sustainability, as described by Goodland. The functional unit, the processes represented in the LCI, and selected impact categories are brought together in this phase to produce system level calculations of environmental impacts. Conclusions take the form of: X system causes Y quantity of Z impact, due to [1:n] processes.The interpretation phase of the LCA involves analysis of the results after each phase, validation to ensure that the LCA is proceeding in compliance with the originally stated method, and creation of reports regarding progress and findings. According to the ISO 14040 standard, this phase involves the reflection on the results of the Life Cycle Inventor Analysisand the Life Cycle Impact Assessment, as well as appropriate communication about findings. While there are many interpretation activities within this phase, they are usually specific to the types of environmental impact that are being calculated in a particular life cycle assessment case. Interpretation involves the production of a variety of charts and reports communicating intermediate results and checking in with the stakeholders of the study to ensure correctness and completeness. This is not actually the last phase of LCA, but instead, it occurs after each of the other phases. It allows for constant reflection on the relationships between results in different phases as they are produced.The studies analyzed in this section were constrained to those whose primary focus was on the United States, European or Australian agricultural systems, due to two reasons. First, a vast majority of published LCA literature is in English, and comes from these three regions. Second, while LCA is popular in other parts of the world, such publications are not always easily accessible, or data sources traceable, as they may be using local databases or native language resources. Figure 3.2 describes the top ten most produced agricultural commodities globally, as reported by the Food and Agriculture Organization of the United Nations Statistics Division.