Development and application of tests for microplastic detection in soil

Plastics serve as versatile and malleable matrices that manufactures can easily manipulate in an effort to develop new technologies to benefit society. These new technologies allow manufacturers to produce products that customers can use either a single time or for more than 50 years. Due to its great versatility, industries across all sectors use plastic at some point during the production process or include plastic in the final products. This great versatility comes with a high demand and use of plastic. This produces a complex scenario with regards to plastic disposal. According to the best educated guess, nearly 24 million tonnes of plastics that were improperly disposed of ended up in terrestrial ecosystems in 2010 alone. Since then, human demand for plastics and plastic consumption have steadily risen and increased to roughly 4 times what they were in 2010 (they reached 51.2 million tonnes in 2018). The implication reads: pollution and the transfer of plastics to ecosystems have worsened over the last 10 years.

Large plastic chunks threaten the environment mainly through direct ingestion by wildlife. Smaller plastics measuring less than 0.5cm threaten the environment not only by their inherent toxicity but also by transporting other pollutants as they themselves move through the soil to other sites. Scientists named these smaller plastic pieces microplastics. Because microplastics constitute a potential threat to soil biota and water or wind might spread them through the environment by transporting them off-site, researchers have finally started to study the occurrence of microplastics in soils. Although researchers have gathered evidence on the occurrence of microplastics in soils as well as the effects that high concentrations might have on soil biota, a few methodological problems have hampered the proposal of new mitigation measures.

To date, researchers have failed to articulate and standardize a reliable soil test to identify different plastic polymers and quantify particles in soil samples. In general, if we cannot effectively measure something, we cannot study it. For this reason, comprehensive data on the occurrence of microplastics in different soil environments or under different land uses is lacking. The ramifications of this lack of data can be seen in the unrealistic laboratory tests where researchers tested the effects of exaggerated concentrations of microplastics on soil biota. Thus, this thesis aims to contribute to the growing body of evidence that identifies end clarifies the sources and dynamics of microplastics in terrestrial ecosystems. It intends to shed light on the occurence of microplastics across different land uses and to reveal major pollution sources. To do so, i proposes new approaches to detect and quantify plastic polymers in soil samples and then uses these new approaches in a case study form Central Chile.

In Chapter 2, we evaluated a handheld spectroradiometer working in the near infrared range (350-2500nm) as an instrument to directly measure microplastic concentrations in soil samples. The Chapter reads as a proof of concept. The results suggest that vis-NIR techniques are able to identify and quantify LDPE, PET, and PVC microplastics in soil samples, with a 10 g kg-1 accuracy and a detection limit ≈ 15 g kg-1. The method stands out since it allows researchers to process samples fast (2 min), avoids extraction steps, and can directly quantify microplastic quantities. As a proof of concept, the proposed approach has motivated the development of other similar methods intended to measure soil and water samples.

The approach proposed in Chapter 2 worked only for pollution hotspots. We wanted to add a method to the toolbox that scientists could use to detect and classify even small amounts of microplastic particles made of different polymers in soil samples. In the original research plan, this ‘detailed’ method was to be used in the environmental assessment of Chapter 5. μFTIR analysis would have allowed this. However, the current available software lacks the functionalities needed to process large amounts of data and lacks the ability to take the most out of μFTIR spectroscopy images. Therefore, in Chapter 3, we present a new software released by The Comprehensive R Archive Network that was designed to be used within the R environment and optimizes current procedures by deploying a novel algorithm to process spectroscopy images (https://CRAN.R-project.org/package=uFTIR).

In Chapter 4, we studied the effect of long-term sludge applications on the accumulation of microplastics in soils. To do so, we sampled soils in Chile’s central valley. Like many scientists studying microplastics in soils, we suffered methodological limitations when we analysed the samples. At the time we carried out the analysis, the μFTIR spectrometer was not available yet and the approach proposed in Chapter 2 did not detect microplastics in the concentrations we expected to find. Therefore, in Chapter 4, we validated an extraction method that uses flotation to isolate microplastic particles from bulk soil samples and then sorted the plastics by their morphology. The study results indicated that the number of microplastic particles increased over time in the soils that received long-term applications of sludge (from 0 to 3.500 microplastic particles per kilo of soil). The study stresses the relevance of sludge as a driver of soil microplastic pollution. It was the first evidence of the role of sludge disposal as a pathway for microplastics to enter into soils.

In chapter 5, we look at the occurrence of microplastics in soils under different land uses. The motivation behind this chapter is in line with the current idea of a global microplastic cycle. The study aimed to assess the presence of microplastics in the topsoil of land exposed to different land management systems at a regional level in Chile’s central valley. To do so, we used the sorting method validated in Chapter 4 and classified microplastics using μFTIR, processing FTIR images with the software described in Chapter 3. Results showed that croplands and pastures were exposed to microplastic pollution, while this type of pollution seldom occurred in rangelands and natural grasslands (both not managed). The study emphasized the role of agriculture in spreading microplastics through the environment. As the first study that has reported the occurrence of microplastics in soils on a broad geographical scale, it underscores the need for more studies that offer actual monitoring data concerning microplastics in soils.

The combination of chapters in this thesis contributes to the growing body of evidence on microplastics in terrestrial ecosystems. The results rank human activities, such as agriculture and waste management, as the first factor that contributes to the direct pollution of soils. It provides new insights that will help to bridge some of the knowledge gaps related to analytical procedures. This thesis also reflects on methodological limitations, stressing the need of proper soil tests that can help quantify and classify microplastics in soils. Taken together, the chapters of this thesis support the idea that an analyst’s toolbox should comprise versatile but standardized soil tests in order to study microplastics in the environment. All in all, this thesis describes three methods that can be used to quantify and qualify microplastics and uses them to report temporal (Chapter 4) and spatial (Chapter 5) variations of microplastics in soils. It is my honest opinion that the data gathered using these methods will support the concept of the global plastic cycle. A concept that will help scientists to communicate their concerns to a broader audience. It will also help policymakers to craft mitigation strategies that buffer the impacts of human activities on the environment.