Life Cycle Assessment of Integrated Nutrient,Energy,and Water Recovery on Large-Scale Dairy Farms

Abstract

Manure management at large-scale dairy farms poses environmental risks to air and water resources through greenhouse gas emissions and nutrient pollution. Current practices do not fully utilize the organic matter and nutrients contained in manure. This work evaluated the environmental performance of novel nutrients, energy, and water innovations for resource recovery (NEWIR) system intended to improve the sustainability of dairy manure management by reducing environmental impacts and recovering valuable resources from the manure. NEWIR consists of hydrothermal carbonization (HTC) for energy recovery, algae cultivation for nutrient recovery, and membrane distillation for water recovery. Life cycle assessment was used to compare the NEWIR system with conventional dairy manure management practices and to identify the benefits and unintended consequences of resource recovery. Results show vastly improved nutrient management performance by NEWIR, with marine and freshwater eutrophication potentials reduced 0.148 kg N-equivalent (eq.) and 0.168 kg P-eq. per functional unit, respectively, compared with conventional practices. NEWIR also reduces global warming potential (GWP) by 15% when recovered energy and nutrients are considered. However, NEWIR has higher water consumption than conventional practices, primarily due to the dilution water needed to support algae growth in HTC aqueous products. The dominant contributor to environmental impacts in NEWIR is algae cultivation, primarily due to buffering chemicals required to maintain high pH growth conditions for Spirulina maxima. Uncertainty analysis shows that NEWIR GWP may exceed that of the conventional farm, highlighting the need to reduce the environmental impacts of algae cultivation. Although additional work is needed to optimize algae cultivation and reduce GWP, NEWIR shows considerable potential to improve nutrient management and energy recovery from dairy manure.

Introduction

Increasing global population coupled with a rising demand for animal products has led to expanding livestock production (Mateo-Sagasta et al., 2017). Despite growth in livestock production, the number of farms is decreasing, representing a shift from small farms to large-scale concentrated animal feeding operations (CAFOs) (U.S. EPA, 2013; CDFA, 2018). In the United States, almost 50% of dairy cattle are housed on farms with at least 1,000 animals; however, in 1992, <10% of dairy cattle were housed on farms of that size (MacDonald et al., 2016). The CAFOs not only provide economic benefits to producers, but they also contribute to environmental concerns, particularly through manure management (MacDonald and McBride, 2009).

Manure management practices at CAFOs lead to greenhouse gas (GHGs) emissions that impact climate change and nutrient pollution that contributes to eutrophication and other water quality concerns (Mateo-Sagasta et al., 2017; Naranjo et al., 2020). In 2017, 9.4% of U.S. methane emissions and 5% of U.S. nitrous oxide emissions came from manure management (U.S. EPA, 2017). Conventionally, manure is used as a soil amendment to produce feed for livestock; however, CAFOs typically produce more manure than can be managed on nearby land (U.S. EPA, 2013), leading to overapplication of nutrients and subsequent surface or groundwater contamination (Kleinman et al., 2012). Based on these challenges, innovative solutions for manure management are needed to address the wide range of environmental concerns with CAFOs.

Recent research has focused on incorporating resource recovery technologies into manure management systems, indicating a paradigm shift from treating manure as a waste stream to viewing manure as a valuable resource (Guillen et al., 2018). A variety of technologies have been explored to recover bioenergy, concentrated nutrients, or clean water from manure to address environmental concerns such as GHGs or nutrient contamination (Higgins and Kendall, 2012; Zhang et al., 2013; Aguirre-Villegas et al., 2014; Joshi and Wang, 2018; Wu et al., 2020).

Life cycle assessment (LCA) is widely used to compare the environmental impacts of resource recovery technologies with conventional farm practices (Rajendran and Murthy, 2019). However, there have been limited studies that incorporate recovery of multiple resources from manure, despite additional environmental benefits possible in a multifaceted approach (Higgins and Kendall, 2012; Zhang et al., 2013; Joshi and Wang, 2018).

In this work, a novel resource recovery system is proposed. The nutrients, energy, and water innovations for resource recovery (NEWIR) system is intended to improve the sustainability of dairy manure management by reducing harmful environmental impacts while simultaneously recovering valuable products. The NEWIR consists of three integrated technologies: (1) hydrothermal carbonization (HTC) for energy recovery, (2) algae cultivation for nutrient recovery, and (3) membrane distillation (MD) for water recovery.

The HTC is an emerging thermochemical treatment for wet organic waste that produces a carbon-rich solid fuel called hydrochar and a nutrient-rich aqueous stream (Reza et al., 2016). Previous work has identified HTC as a promising conversion technology to produce hydrochar from cow manure (Reza et al., 2016) and recover bioenergy (Berge et al., 2015; Marin-Batista et al., 2020), but LCA studies on HTC of manure are limited. Previous work has identified benefits in GHGs when hydrochar replaces coal for electricity production (Berge et al., 2015). Other studies have noted that the environmental performance of HTC depends on the waste management system it is replacing (Owsianiak et al., 2016). To date, an LCA on HTC of dairy manure in the context of a large-scale CAFO has not been performed.

Algae cultivation has been widely explored as a wastewater treatment method for livestock manure to reduce nutrient contamination and subsequent eutrophication impacts (Higgins and Kendall, 2012; Wu et al., 2020), but its application to HTC aqueous products (HAP) is limited. Previous work has shown that the treatment of HAP is essential to reduce eutrophication impacts from HTC (Berge et al., 2015). The HAP contains most of the nitrogen and phosphorus found in dairy manure in the forms of ammonia, nitrate, organic nitrogen, and phosphate (Qaramaleki et al., 2020), presenting an opportunity for nutrient recovery via microalgae.

Some preliminary work has evaluated algae cultivation in HAP (Du et al., 2012; Belete et al., 2019), with more than 90% nutrient removal achieved. However, to date, no published studies address algae cultivation in HAP derived from dairy manure, nor do they use LCA to evaluate performance. Further, microalgae biomass can be used as a protein supplement for dairy cattle (Madeira et al., 2017), presenting an opportunity to couple nutrient recovery with energy recovery through microalgae cultivation and HTC and recycle the manure nutrients on-farm as cattle feed.

The MD is an emerging membrane treatment technology that uses low thermal gradients to produce distillate-quality water (Salls et al., 2018). The MD can be integrated with HTC to utilize waste heat and produce clean water with no additional energy inputs (Gustafson et al., 2018). In addition, MD provides a stream of distillate-quality water for reuse on farm, an added benefit over wastewater treatment with microalgae alone. Previous work has evaluated MD as a treatment for HAP (Silva and Hiibel, 2021), but to date, an LCA has not been performed.

The objective of this study is to perform a comparative LCA to evaluate the environmental impacts of NEWIR relative to conventional manure management practices. The conventional system selected for comparison is based on a large-scale dairy CAFO in California (CA). The specific aims of this work are to (1) develop and analyze life cycle inventory (LCI) data for the NEWIR system and a conventional CA dairy CAFO, (2) identify the impacts and benefits of resource recovery technologies, and (3) identify areas of high impact or unintended consequences within NEWIR to propose mitigation strategies that can improve the novel NEWIR system.

Materials and Methods

Case study location

The LCA presented in this study compares the environmental impacts of NEWIR relative to a large-scale dairy CAFO in the Central Valley of CA. The CA produces 18% of the US milk supply (CDFA, 2018), with the majority of dairy cattle housed on CAFOs (Meyer et al., 2019). In CA, 60% of agricultural GHGs come from dairies, with manure management as a primary contributor (CARB, 2019). Impending regulations on the dairy industry mandate reductions in methane emissions from manure (State of California, 2016), encouraging the development of energy recovery technologies (Lee and Sumner, 2018).

Nitrate contamination of groundwater is a widespread concern (Rosenstock et al., 2014; Ransom et al., 2018), prompting N application limits in the form of nutrient management plans. Dairies must limit the application of N to crop fields receiving manure to 1.4 times the N harvested in the crop (R5-2013-0122, 2013). Resource recovery technologies have the potential to address these GHG and nutrient management concerns while simultaneously creating useful products from manure.

Goal and scope

The goal of this study was to compare the environmental impacts of the NEWIR resource recovery system with conventional manure management practices. The conventional CA dairy farm is referred to as the “baseline” scenario to represent typical operations. The LCA was performed in compliance with ISO 14040 standards (ISO, 2006) by using a functional unit of 1,000 kg wet manure produced on a large-scale (1,000 milking cows) CAFO. The findings of the study are intended to help dairy farmers, producers, researchers, and other stakeholders understand the benefits and challenges of innovative manure management solutions relative to conventional farm practices.

System boundaries

The system boundaries include relevant processes for handling dairy manure from its generation in the dairy barns to its end use as a soil amendment or recovered product (Fig. 1). Other farm practices such as animal husbandry, milk processing, and most feed production were assumed the same with both manure treatment systems and thus were not included in the LCA.

FIG. 1.

Process flow diagrams with system boundaries of the two scenarios analyzed by LCA: a conventional CA dairy CAFO management practice (baseline) and the NEWIR system with resource recovery. CAFO, concentrated animal feeding operation; LCA, life cycle assessment; NEWIR, Nutrients, Energy, and Water Innovations in Resource Recovery.

Baseline scenario description

A freestall dairy with a predominantly flush-based manure collection system was modeled for the baseline scenario (Fig. 1), representative of 50% of CA dairies (Meyer et al., 2019). Flushed manure is collected from the barn and milking parlor areas and stored in a lagoon; solid manure is scraped from transfer lanes and open corral areas and stored in a solid pile. Liquid from the lagoon is applied to crop fields as a fertilizer throughout the growing season, whereas manure solids are applied twice per year between crop cycles (Miller et al., 2020). The dairy grows wheat in the winter and corn in the summer to provide a portion of the herd's feed (Miller et al., 2020).

NEWIR scenario description

The NEWIR treatment system consists of three processes for integrated resource recovery from dairy manure: (1) HTC for energy recovery, (2) algae cultivation for nutrient recovery, and (3) MD for high-quality water recovery (Fig. 1). The HTC converts manure slurry into carbon-rich hydrochar and nutrient-rich HAP (Reza et al., 2016). Hydrochar is dried, pelletized, and burned in a conventional coal power plant for electricity production, replacing a fossil fuel with bio-energy (Reza et al., 2014).

The HAP serves as a nutrient source for microalgae cultivated in raceway ponds. Microalgae are harvested by using a filter system, and they are then blended with crop residue for use as a protein-rich cattle feed in place of soybean meal. After separation, half of the HAP stream is recycled back to the pond for continued nutrient removal. Finally, MD uses waste heat from HTC to recover high-quality water from the algae harvest effluent that can be reused on-farm for cleaning and sterilization, replacing conventional reverse osmosis (RO) systems for water purification.

Life cycle inventory

The LCI contains all inputs and outputs in each unit process for both scenarios (Table 1 and Supplementary Table S1 and S2). Inputs include manure, water, energy, and materials for all manure management processes. Outputs include environmental emissions to air, water, and soil, along with recovered products. Calculation methods for manure production and air emissions are detailed in Supplementary Tables S3 and S4. The LCI was modeled by using Ecoinvent version 3.6 and US LCI databases for upstream processes, including energy from the CA electricity grid and materials production ( U.S. Life Cycle Inventory Database, 2012; Wernet et al., 2016).

Table 1.

Overview of Life Cycle Inventory Data Based on Unit Processes in Each Scenario

Unit process	Item (units)	Baseline inventory	NEWIR inventory	Description
Manure collection	Diesel (kg)	✓	✓	Scraping manure
	Water (L)	✓	—	Sanitation water (RO treatment)
	Water (L)	✓	✓	Flushing manure
	Electricity (kWh)	✓	✓	Pumping
	CH₄, N₂O, NH₃ (kg)	✓	✓	Emission to air
Manure storage	CH₄, N₂O, NH₃ (kg)	✓	✓	Emission to air
HTC	Steel (kg)	—	✓	Reactor and infrastructure
	Water (L)	—	✓	Water input for 10% solids content
	CO₂ (kg)	—	✓	Biogenic CO₂ emissions
	Electricity (kWh)	—	✓	Pumping, filter press operation
	Heat from natural gas (kWh)	—	✓	Reactor
Algae cultivation	HDPE (kg)	—	✓	Pond liner
	Buffer chemicals (kg)	—	✓	Bicarbonate, carbonate for pH adjustment
	Water (L)	—	✓	HAP dilution
	Electricity (kWh)	—	✓	Pumping, paddle wheel mixing, filter press operation
Membrane distillation	PTFE (kg)	—	✓	Membrane material
Cattle Feed	Corn stover (kg)	—	✓	Mix with algae for feed
Cattle Feed	Avoided soybean (kg)	—	✓	Soybean replaced by algae feed
Power generation	Electricity (kWh)	—	✓	Drying, pelletizing hydrochar
	CO₂ (kg)	—	✓	Biogenic CO₂ emissions
	Avoided coal (kg)	—	✓	Avoided mining operations and fossil CO₂ emissions
Land application	Diesel (kg)	✓	—	Equipment use
	N₂O, NH₃ (kg)	✓	—	Emission to air
	N, P (kg)	✓	—	Crop uptake (corn and wheat)
	NO₃⁻, PO₄³⁻ (kg)	✓	—	Emitted to soil and water

The “✓” symbol signifies that an inventory component is included in the system. The complete LCA inventory for both systems can be found in Supplementary Tables S1 and S2.

Italicized entries represent avoided products associated with resource recovery.

HAP, HTC aqueous products; HTC, hydrothermal carbonization; LCA, life cycle assessment; NEWIR, Nutrients, Energy, and Water Innovations in Resource Recovery; RO, reverse osmosis.

Inventory data were collected by using a combination of published farm surveys in CA, emissions calculation methods (IPCC, 2006), and a literature review of existing HTC and algae cultivation systems. Experimental data for cultivation of Spirulina maxima in HAP were collected specifically for this study and are presented in Supplementary Table S5. Experimental results for HTC and MD were taken from Reza et al. (2016) and Silva and Hiibel (2021), respectively. Process modeling of the full-scale NEWIR system was performed in ChemCAD software.

The baseline includes all processes associated with manure management on a large-scale freestall dairy farm in CA; NEWIR includes all inventory data for HTC, algae cultivation, and MD (Fig. 1).

The NEWIR also accounts for avoided products offset by recovered resources. Avoided products are based on the on-farm and upstream processes that are avoided when recovered energy or nutrients are used in place of a conventional product. With HTC, hydrochar is a substitute for low-grade coal in conventional coal-fired power plants (Reza et al., 2014). Energy recovery from hydrochar avoids upstream processes associated with coal mining as well as the fossil CO₂ emissions from coal combustion. The quantity of avoided coal is based on equivalent amounts of energy through the higher heating values of coal and hydrochar.

With nutrient recovery, algae-based cattle feed replaces soybean feed as a protein supplement in the cow diet. The avoided product is represented with soybean feed production, and the quantity of avoided feed is based on equivalent protein amounts in the soybean and algae feeds. Water recovery is accounted for in the baseline inventory as conventional farms use RO systems for clean water production, whereas NEWIR produces high-quality water from MD.

Impact assessment method

The impact assessment was performed in SimaPro (PhD version 9.1.1.1) by using the ReCiPe Midpoint (E) impact assessment method (Huijbregts et al., 2017). Four impact categories were selected for focus: global warming potential (GWP), marine eutrophication potential (MEP), freshwater eutrophication potential (FEP), and water consumption (WC). The GWP is a measure of GHGs in kg CO₂-equivalent (eq.) from dairy manure as well as recovered energy; MEP and FEP are measures of nutrient emissions in kg N-eq. and kg P-eq., respectively. The WC is a measure of water use in m³ in each scenario.

Sensitivity and uncertainty analyses

Sensitivity analysis was performed to evaluate the sensitivity of impact category results based on changes in the LCI. Because information regarding the statistical distribution of inventory data is limited, values were varied by ±10%, similar to previous studies (Aguirre-Villegas et al., 2014). Methods and results of the sensitivity analysis are discussed in the Supporting Information. Monte Carlo uncertainty analysis was performed in SimaPro version (PhD version 9.1.1.1) by calculating impact results, as inventory inputs were varied within defined distributions over 1,000 iterations.

Distributions were assumed to be uniform between a maximum and minimum value since the real distribution of each input was not known. Similar to the sensitivity analysis, maximum and minimum values of each input were assumed to be ±10% of the inventory value unless further information (e.g., survey data) was available. The uncertainty range is based on a 95% confidence interval and calculated for both impacts and avoided impacts. The standard deviation obtained from uncertainty analysis is reported in Table 2.

Table 2.

Life Cycle Impact Assessment Results for the Global Warming Potential, Freshwater Eutrophication Potential, Marine Eutrophication Potential, and Water Consumption Impact Categories

Impact category	GWP	FEP	MEP	WC
Per 1,000 kg manure functional unit	kg CO₂-eq.	kg P-eq.	kg N-eq.	m³
Baseline	72.0 ± 3.00	0.208 ± 0.009	0.109 ± 0.005	1.93 ± 0.443
NEWIR (gross, without avoided products)	187 ± 26.5	0.092 ± 0.039	0.006 ± 0.002	19.0 ± 3.12
Avoided products	−126 ± 7.35	−0.05 ± 0.031	−0.046 ± 0.007	−0.065 ± 1.21
NEWIR (net, with avoided products)	61.3 ± 12.3	0.040 ± 0.034	-0.039 ± 0.007	19.0 ± 3.59

The NEWIR gross total contains all of the NEWIR impacts without any benefits from avoided products through resource recovery, whereas the net total includes avoided products as negative impacts. Error represents the standard deviation obtained from uncertainty analysis.

eq., Equivalent; FEP, freshwater eutrophication potential; GWP, global warming potential; MEP, marine eutrophication potential; WC, water consumption.

Results and Discussion

Global warming potential

The GWP for each scenario reflects the total GHGs released during each process. Results in all impact categories are reported for 1,000 kg manure functional unit. In the baseline scenario, the total GWP is 72.0 ± 3.00 kg CO₂-eq., compared with the NEWIR gross total of 187 ± 26.5 kg CO₂-eq., which does not include avoided products (Table 2). When avoided products are considered, the NEWIR net total is reduced to 61.3 kg CO₂-eq., indicating that NEWIR reduces overall GHGs relative to conventional manure management practices when avoided products are included.

The dominant contributor to the baseline GWP is manure storage in the lagoon, contributing 70.4% of the total (Fig. 2), which is consistent with other studies on GHGs from dairy manure (Aguirre-Villegas et al., 2014; Kaffka et al., 2016; Naranjo et al., 2020). Solid manure storage is the smallest contributor with 3.1%., whereas collection and land application contribute 9.4% and 17.2% of GWP, respectively. Methane (69.5%) and nitrous oxide (13.0%) emissions contribute most of the GWP. Other contributors to GWP include diesel fuel use for collection and land application (11.2%), electricity for pumping (2.4%), synthetic N fertilizer production (1.5%), and RO treatment of water for cleaning and sterilization of milking equipment (2.2%).

FIG. 2.

GWP expressed in kg CO₂-eq. per 1,000 kg manure functional unit for the baseline and NEWIR scenarios. Negative GWP values represent environmental benefits in the form of avoided products. eq., Equivalent; GWP, global warming potential.

In NEWIR, the dominant contributors shift from manure storage to algae cultivation (Fig. 2). Manure collection is equivalent to the baseline scenario due to the same collection practices, and land application is completely avoided. Manure is processed quickly by HTC, reducing storage emissions from 52.9 kg CO₂-eq. in the baseline to 2.0 kg CO₂-eq. in NEWIR. However, the resource recovery processes raise the gross total GWP to 187 kg CO₂-eq. Manufacturing sodium bicarbonate and sodium carbonate for pH management in the pond makes up the majority (88.7%) of the NEWIR GWP, whereas other operations including pumping (0.6%), paddlewheel mixing (0.6%), and filter press separation (0.7%) have minimal contribution.

These results are in contrast to other algae-focused LCAs due to the large contribution from buffer chemicals used to maintain alkaline conditions for S. maxima during cultivation in HAP. Microalgae have been widely studied for nutrient removal from manure-based wastewater, showing promising removal of N, P, and chemical oxygen demand (COD) (Higgins and Kendall, 2012). Belete et al. (2019) cultivated algae on HAP derived from sewage sludge supplemented with micronutrients and boric acid; however, an LCA was not performed so the environmental impacts of such chemical additions were not quantified.

In the current study, a carbonate/bicarbonate buffer was used to control pH, but no other salts or nutrients were supplemented. Based on the GWP associated with algae cultivation (Fig. 2), reducing chemical buffer input is essential to further reducing the GWP of NEWIR relative to conventional farms, and algae species more suited for growth at near-neutral pH should be considered. However, this change poses risks because high pH is an effective way to manage contamination by other microbial species in open pond systems (Richmond and Grobbelaar, 1986; Delrue et al., 2017).

Despite the high impacts associated with buffer chemicals, algae cultivation also provides benefits to GWP in the form of the avoided soybean meal when algae-based cattle feed is used in the cow diet. The NEWIR produces 10.9 kg dry weight algae biomass per functional unit, providing a GWP offset of 58.0 kg CO₂-eq. (Fig. 2) or 31% of the gross total GWP through avoided soybean feed production. Although algae cultivation provides considerable benefits to GWP through avoided soybean, the inputs associated with production of S. maxima make algae cultivation a net contributor to GWP.

Future research should evaluate the growth of alternative algae species at neutral pH, removing the need for chemical buffers. In addition, alternate nutrient recovery methods such as chemical precipitation could be considered (Numviyimana et al., 2022); however, this type of nutrient recovery would not provide cattle feed for reuse on farm, and different avoided products would need to be considered.

The HTC has a relatively small impact in terms of GWP (4.4%) and provides the greatest benefit through avoided products. The NEWIR system produces 74.1 kg dried hydrochar per functional unit, which results in a 36% reduction of GWP by replacing lignite coal with hydrochar for electricity generation (Fig. 2). Berge et al. (2015) identified improvements in GWP when HTC was used to recover energy from food waste relative to anthracite, bituminous, and lignite coals, as well as the average US energy grid.

Natural gas combustion to heat the HTC reactor to 200°C is the greatest contributor to GWP, accounting for approximately half of the total HTC impacts. Infrastructure associated with HTC had minimal contribution of 0.6% of the GWP total. Other operational components had minimal contribution to GWP, including pumping (0.2%), filter press separation of hydrochar and HAP (0.05%), and drying and pelletizing hydrochar (0.8%).

The contribution to GWP from water recovery through MD is negligible. Because MD can utilize waste heat from the HTC reactor (Gustafson et al., 2018), no additional energy inputs are needed. In the baseline, high-quality water for sanitization procedures is produced by using RO treatment, accounting for 2.2% of the baseline GWP. This impact is avoided in NEWIR, where high-quality water comes from MD.

Uncertainty analysis was performed to evaluate the possible range of impacts associated with each scenario for a 95% confidence interval. In the baseline, the uncertainty of GWP ranges from 67.0 to 77.1 kg CO₂-eq. In NEWIR, the gross total GWP is significantly higher than the baseline and ranges from 147 to 251 kg CO₂, whereas the avoided products range from −143 to −113 kg CO₂-eq. When combined, the net total GWP for NEWIR ranges from 4 to 138 kg CO₂-eq., effectively bracketing the baseline GWP range and highlighting the need to address uncertainty in the NEWIR inventory by determining realistic maximum and minimum inventory values. Results from the sensitivity analysis suggest that NEWIR GWP results are the most sensitive to changes in bicarbonate, followed by the coal offset, soybean offset, and carbonate (Supplementary Fig. S1A). Consequently, reducing the buffer inputs and increasing the coal and soybean offsets would be the best way to reduce GWP impacts in NEWIR. In the baseline, sensitivity analysis indicates that GWP is most sensitive to methane emissions, primarily from manure storage (Supplementary Fig. S2A).

Eutrophication potential

Eutrophication impacts from N and P are shown in Fig. 3. The MEP is a measure of N emissions to marine water bodies, whereas the FEP considers P emissions to freshwater bodies (Huijbregts et al., 2017).

FIG. 3.

(A) MEP expressed in kg N-eq. and (B) FEP in kg P-eq. per 1,000 kg manure functional unit for the baseline and NEWIR scenarios. Negative values represent environmental benefits in the form of avoided products. FEP, freshwater eutrophication potential; MEP, marine eutrophication potential.

The NEWIR shows considerable improvement in N management, as reflected in MEP (Fig. 3A). In the baseline scenario, MEP is 0.109 kg N-eq., considerably higher than the 0.006 kg N-eq. gross total for NEWIR and the −0.039 kg N-eq. net total (Table 2). The N emissions in NEWIR come from buffer chemical manufacturing for algae cultivation (100% of gross total). However, avoided products offset more N emissions than NEWIR produces, resulting in net negative MEP.

Avoided soybean cultivation accounts for the bulk of MEP offset with 93% of the offset, whereas avoided electricity production from coal contributes 7%; both nutrient and energy recovery provide benefits to MEP.

These results indicate significant improvements by NEWIR over conventional practices. In the baseline, MEP is dominated by nitrate emissions during land application (Fig. 3A), representing a loss of 9.5% of the manure N content. N is also emitted as ammonia and nitrous oxide air emissions during each unit process, accounting for nearly 50% of the manure N content, which is similar to other reports for Central Valley dairies (Chang et al., 2005); however, MEP does not reflect air emissions, only water and soil emissions (Huijbregts et al., 2017).

In the Central Valley, N management is a major concern due to the high concentration of dairies and consistent overapplication of manure nutrients to crop fields (Rosenstock et al., 2014), with nitrate contamination threatening drinking water resources in some regions (Ransom et al., 2018). In CA, dairies are required to limit N application on crop fields to 1.4 times the N removed in the crop (R5-2013-0122, 2013) in an effort to reduce groundwater nitrate contamination (Miller et al., 2017). Nutrient recovery through NEWIR has considerable potential to reduce N emissions and comply with regulations. Other inputs in the baseline LCI have negligible contribution to MEP (<0.01%), including synthetic N fertilizer production, diesel fuel production, and electricity production.

The NEWIR also shows improvement in FEP relative to conventional practices. In the baseline, the total FEP is 0.208 kg P-eq. In NEWIR, the gross total FEP is 0.092 kg P-eq. and the net total is 0.040 kg P-eq. (Table 2). Similar to MEP, the dominant contributor in NEWIR is manufacturing buffer chemicals for algae cultivation (97.8%), whereas the baseline scenario is dominated by phosphate emissions during land application (Fig. 3B). In the baseline scenario, ∼55% of the manure P content is lost to surface water and soil emissions, whereas 45% is recovered by crops.

Although NEWIR does not result in net negative FEP, it shows a considerable decrease in P emissions relative to conventional practices. The NEWIR as a whole substantially reduces nutrient emissions, highlighting the benefits of integrated resource recovery for dairy manure management, particularly in regions such as the Central Valley with nutrient pollution concerns.

With both MEP and FEP, the upper limit of the 95% confidence interval obtained from uncertainty analysis is lower than that of the baseline, indicating improvements in nutrient management regardless of uncertainty. The NEWIR gross total MEP ranges from 0.004 to 0.011 kg N-eq., whereas the avoided products range from −0.063 to −0.034 kg N-eq. The resulting net NEWIR MEP (−0.059 to −0.023 kg N-eq.) is substantially lower than the baseline (0.100–0.118 kg N-eq.). In addition, the sensitivity analysis shows that NEWIR MEP results are the most sensitive to changes in the soybean offset, followed by bicarbonate buffer and the coal offset (Supplementary Fig. S1B). Increasing algae yield will help increase the soybean offset, bringing the most benefit to MEP.

With FEP, the NEWIR gross total ranges from 0.043 to 0.201 kg P-eq., whereas the FEP of the avoided products range from −0.151 to −0.019 kg P-eq. As with MEP, the net total NEWIR FEP (−0.108 to 0.182 kg P-eq.) is lower than the baseline FEP (0.192–0.225 kg P-eq.), with a net negative FEP possible when NEWIR inputs are minimized and avoided products are maximized. Sensitivity analysis shows that NEWIR FEP results are the most sensitive to changes in the bicarbonate buffer followed by the coal offset, with minimal effects from other inventory components (Supplementary Fig. S1C). As a result, reducing buffer input and improving energy recovery will help reduce FEP.

Water consumption

The NEWIR scenario has considerably higher WC than the baseline (Fig. 4). The NEWIR consumes 19.0 m³, nearly 10 times the baseline scenario WC of 1.92 m³ (Table 2). Avoided products provide minimal benefit in WC, with only 0.3% reduction, due to the small volume of MD water produced. In the baseline, the primary water use is flushing manure from the milk parlor and barn areas. Other baseline inputs, such as diesel fuel production (1.1%), synthetic N fertilizer (<0.0%), and electricity production (0.8%), have minimal contribution to WC. In NEWIR, manure is collected in the same manner as the baseline, leading to the same WC value for collection in addition to other water inputs needed for NEWIR.

FIG. 4.

WC in m³ per 1,000 kg manure functional unit for the baseline and NEWIR scenarios. WC, water consumption.

Algae cultivation is responsible for the majority of WC with 89%. Within algae cultivation, 55% of WC is used for diluting the HAP to 12%, a level suitable for algae growth based on experimental data (Supplementary Table S5). Previous work with HAP has highlighted the need for major dilutions to achieve growth (Du et al., 2012; Belete et al., 2019). Du et al. (2012) report using 0.5–2% HAP for successful cultivation, whereas Belete et al. (2019) used 0.3% HAP. In the current work, growth of S. maxima was experimentally validated at higher HAP concentrations (up to 12%) than in previous studies.

Manufacturing of buffer chemicals is also an important contributor, with 39% of the algae cultivation WC. The other components within algae cultivation, including energy generation for pumping, mixing, and filter press separation, have a minimal combined contribution of 0.02%. Because microalgae production does not take place in the baseline, the additional water inputs needed for cultivation represent a considerable increase in WC. As previously discussed, alternative algae species that can be grown in near-neutral pH are an important future consideration for NEWIR, as reducing the chemical buffers needed for S. maxima production will help reduce WC.

The HTC contributes a small fraction of WC with 2.4%. The HTC requires a water-to-biomass ratio ranging from 5:1 to 12:1 (Reza et al., 2016; Qaramaleki et al., 2020), so water is added as needed to maintain the desired ratio. In this model, a water-to-biomass ratio of 10:1 was used (Marin-Batista et al., 2020); thus, additional dilution water was added to bring the manure slurry to 90% water content. Depending on specific farm practices, water may be available from manure flushing to meet the HTC water requirements, removing the need for additional water before the reaction.

Based on the uncertainty analysis, the 95% confidence interval for NEWIR gross total WC ranges from 11.9 to 24.2 m³. The uncertainty associated with WC of the avoided products is substantial, ranging from −2.24 to 3.00 m³. The resulting net NEWIR WC (9.64–27.2 m³) is substantially higher than the baseline WC (0.991–2.72 m³). Sensitivity analysis indicates that NEWIR WC is the most sensitive to changes in the bicarbonate buffer production and pond dilution water (Supplementary Fig. S1D).

Although algae cultivation provides substantial benefits to eutrophication through avoided soybean production, improving WC is an important area of future research, particularly in drought-prone areas such as CA. Algae cultivation could be substantially improved by growing species at neutral pH, rather than the elevated pH required by S. maxima, removing the need for chemical buffers with high GWP and WC contribution.

In addition, due to the high contribution of algae cultivation to these two categories, alternative technologies may be required to improve nutrient recovery from HAP. Current research has evaluated struvite precipitation as a nutrient recovery method from dairy manure-derived HAP (Numviyimana et al., 2022). However, additional LCA studies are needed to compare the environmental impacts and benefits of alternative nutrient removal or recovery methods.

Conclusions

The proposed NEWIR system is a promising manure management technology that uses integrated resource recovery to address sustainability challenges of large-scale dairies and recover valuable products from manure. The LCA was used to compare the environmental impacts of NEWIR relative to conventional manure management practices on large-scale CA dairy farms. The NEWIR shows substantial improvements in nutrient management, reflected in the FEP and MEP categories.

Manure nutrients are recovered in algae biomass for use as cattle feed rather than being land applied and causing eutrophication impacts. The NEWIR also improves GWP when recovered nutrients and energy are taken into account in the form of avoided products. However, NEWIR has higher WC than the conventional farm, with microalgae cultivation identified as the dominant contributor within NEWIR. The major contributors to GWP and WC are buffer chemicals used for pH adjustment in the pond and water used for dilution of HAP.

Consequently, further work is needed to optimize microalgae cultivation in HAP, ideally with a more HAP-tolerant algae species that can grow at near-neutral pH to avoid the use of chemical buffers. On the other hand, energy recovery through HTC shows promising effects with a large reduction in GWP when manure-derived hydrochar replaces coal for electricity production. Nutrient recovery through algae cultivation significantly benefits the MEP and FEP categories.

Although algae cultivation also provides a considerable offset to GWP by replacing a portion of conventional cattle feed, the inputs required for algae cultivation result in a net positive contribution to GWP. Tradeoffs between different impact categories highlight the importance of LCA to identify the benefits and unintended consequences of resource recovery systems, as well as using LCA as a tool to inform process design and policy. The results of the LCA show that the NEWIR system is a viable technology to recover energy and nutrients from dairy manure, highlighting the benefits of resource recovery to improve the sustainability of food, energy, and water systems.

Footnotes

Acknowledgments

The authors would like to acknowledge Nicholas Silva and Carlos Rocha for their contributions to data collection and process modeling.

Authors' Contributions

P.K.C. and S.R.H. designed the work and obtained funding. C.J.G. collected LCI data and performed impact assessment. All authors performed data analysis and interpretation. C.J.G. drafted the article, and P.K.C. and S.R.H. provided critical revision.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

This material is based on work supported by the National Science Foundation under grant number NSF Award #1856009. Any opinions, findings, and conclusions or recommendations expressed in this article are those of the authors and do not necessarily reflect the views of the National Science Foundation.

Supplementary Material

References

Aguirre-Villegas

H.A.

, Larson

, and Reinemann

D.J.

(2014). From waste-to-worth: Energy, emissions, and nutrient implications of manure processing pathways. Biofuels Bioprod. Biorefin. 8, 770.

Belete

Y.Z.

, Leu

, Boussiba

, Zorin

, Posten

, Thomsen

, Wang

, Gross

, and Bernstein

(2019). Characterization and utilization of hydrothermal carbonization aqueous phase as nutrient source for microalgal growth. Bioresour. Technol. 290, 121758.

Berge

N.D.

, Li

, Flora

J.R.V.

, and Ro

K.S.

(2015). Assessing the environmental impact of energy production from hydrochar generated via hydrothermal carbonization of food wastes. Waste Manag. 43, 203.

California Air Resources Board (CARB). (2019). California greenhouse gas emissions for 2000 to 2017: Trends of emissions and other indicators. Available at: https://ww2.arb.ca.gov/ghg-inventory-archive (Accessed March 9, 2020).

California Department of Food and Agriculture (CDFA). (2018). California agricultural statistics review 2017–2018. Available at: https://www.cdfa.ca.gov/statistics (Accessed June 13, 2020).

Chang

, Harter

, Letey

, Meyer

R.D.

, Campbell-Matthews

, Mitloehner

, Pettygrove

, Robinson

, and Zhang

R.D.

(2005). Managing Dairy Manure in the Central Valley of California. University of California Division of Agriculture and Natural Resources Committee of Experts on Dairy Manure Management. Available at: https://groundwater.ucdavis.edu/files/136450.pdf (Accessed January 16, 2020).

Delrue

, Alaux

, Moudjaoui

, Gaignard

, Fleury

, Perilhou

, Richaud

, Petitjean

, and Sassi

J.F.

(2017). Optimization of Arthrospira platensis (Spirulina) growth: From laboratory scale to pilot scale. Fermentation. 3, 59.

, Hu

, Shi

, Ma

, Cheng

, Chen

, Liu

, Lin

, and Ruan

(2012). Cultivation of a microalga Chlorella vulgaris using recycled aqueous phase nutrients from hydrothermal carbonization process. Bioresour. Technol. 126, 354.

Guillen

D.P.

, Coats

E.R.

, McDonald

A.G.

, and Feris

(2018). An eco-friendly system for the production of value-added materials from dairy manure. JOM. 70, 1946.

10.

Gustafson

R.D.

, Hiibel

S.R.

, and Childress

A.E.

(2018). Membrane distillation driven by intermittent and variable-temperature waste heat: System arrangements for water production and heat storage. Desalination, 448, 49.

11.

Higgins

B.T.

, and Kendall

(2012). Life cycle environmental and cost impacts of using an algal turf scrubber to treat dairy wastewater. J. Ind. Ecol. 16, 436.

12.

Huijbregts

M.A.J.

, Steinmann

Z.J.N.

, Elshout

P.M.F.

, Stam

, Verones

, Vierira

M.D.M.

, Hollander

, Zijp

, and van Helm

(2017). ReCiPe 2016 v1.1: A harmonized life cycle inventory assessment method at midpoint and endpoint level. Int. J. Life Cycle Assess. 22, 138.

13.

Intergovernmental Panel on Climate Change (IPCC). (2006) Chapter 10: Emissions from livestock and manure management. In IPCC Guidelines for National Greenhouse Gas Inventories, Prepared by the National Greenhouse Gas Inventories, Kanagawa, Japan, p. 10.1.

14.

International Organization for Standardization (ISO). (2006). ISO 14040: Environmental management: life cycle assessment principles and framework, 2nd edition. Geneva, Switzerland: International Organization for Standardization.

15.

Joshi

, and Wang

(2018). Manure management coupled with bioenergy production: An environmental and economic assessment of large dairies in New Mexico. Energy Econ. 74, 197.

16.

Kaffka

, Barzee

, El-Mashad

, Williams

, Zicari

, and Zhang

(2016). Evaluation of dairy manure management practices for greenhouse gas emissions mitigation in California: Final Technical Report to the State of California Air Resources Board. Contract No. 14-456. Available at https://biomass.ucdavis.edu.wp-content/uploads/ARB-Report-Final-Draft-Transmittal-Feb-26-2016.pdf (Accessed February 20, 2020).

17.

Kleinman

, Blunk

K.S.

, Bryant

, Saporito

, Beegle

, Czymmek

, Ketterings

, Sims

, Shortle

, McGrath

, Coale

, Dubin

, Dostie

, Maguire

, Meinen

, Allen

, O'Neill

, Garber

, Davis

, Clark

, Sellner

, and Smith

(2012). Managing manure for sustainable livestock production in the Chesapeake Bay Watershed. J. Soil Water Conserv. 67, 54A.

18.

Lee

, and Sumner

D.A.

(2018). Dependence on policy revenue poses risks for investments in dairy digesters. Calif. Agric. 72, 226.

19.

MacDonald

J.M.

, Cessna

, and Mosheim

(2016). Changing structure, financial risks, and government policy for the U.S. dairy industry, ERR-205, U.S. Department of Agriculture, Economic Research Service. Available at: https://www.ers.usda.gov/publications/err-economic-research-report/err205 (Accessed August 11, 2021).

20.

MacDonald

J.M.

, and McBride

W.D.

(2009). The Transformation of U.S. Livestock Agriculture: Scale, Efficiency, and Risks, Economic Information Bulletin No. 43. Economic Research Service, U.S. Department of Agriculture. Available at: https://www.ers.usda.gov/publications.pub-details/?pubid=44294 (Accessed August 4, 2021).

21.

Madeira

M.S.

, Cardoso

, Lopes

P.A.

, Coelho

, Afonso

, Bandarra

N.M.

, and Prates

J.A.M.

(2017). Microalgae as feed ingredients for livestock production and meat quality: A review. Livest. Sci. 205, 111.

22.

Marin-Batista

J.D.

, Villamil

J.A.

, Qaramaleki

S.V.

, Coronella

C.J.

, Mohedano

A.F.

, and de la Rubia

M.A.

(2020). Energy valorization of cow manure by hydrothermal carbonization and anaerobic digestion. Renew. Energy, 160, 623.

23.

Mateo-Sagasta

, Zadeh

S.M.

, Turnal

, and Burke

(2017). Water pollution from agriculture: A global review. Executive Summary. Food and Agriculture Organization of the United Nations Led By International Water Management Institute. Available at: https://www.fao.org/3/i7754e/i7754e.pdf (Accessed February 17, 2020).

24.

Meyer

, Heguy

, Karle

, and Robinson

(2019). Characterize physical and chemical properties of manure in california dairy systems to improve greenhouse gas emission estimates. Final Report: Contract No. 16RD002. Prepared for the California Air Resources Board and the California Environmental Protection Agency. Available at: https:ww2.arb.ca.gov/sites/default/files/classic/research/apr/past/16rd002.pdf (Accessed March 12, 2021).

25.

Miller

C.M.F.

, Waterhouse

, Harter

, Fadel

J.G.

, and Meyer

(2020). Quantifying the uncertainty in nitrogen application and groundwater nitrate leaching in manure based cropping systems. Agric. Syst. 184, 102877.

26.

Naranjo

, Johnson

, Rossow

, and Kebreab

(2020). Greenhouse gas, water, and land footprint per unit of production of the California dairy industry over 50 years. J. Dairy Sci. 103, 3760.

27.

Numviyimana

, Warchoł

, Khalaf

, Leahy

J.J.

, and Chojnacka

(2022). Phosphorus recovery as struvite from hydrothermal carbonization liquor of chemically produced dairy sludge by extraction and precipitation. J. Environ. Chem. Eng. 10, 106947.

28.

Owsianiak

, Ryberg

M.W.

, Renz

, Hitzl

, and Hauschild

M.Z.

(2016). Environmental performance of hydrothermal carbonization of four wet biomass waste streams at industry-relevant scales. ACS Sustain. Chem. Eng. 4, 6783.

29.

Qaramaleki

S.V.

, Villamil

J.A.

, Mohedano

A.F.

, and Coronella

C.J.

(2020). Factors affecting solubilization of phosphorus and nitrogen through hydrothermal carbonization of animal manure. ACS Sustain. Chem. Eng. 8, 12462.

30.

R5-2013-0122. (2013). California Regional Water Quality Control Board Central Valley Region Order R5-2013-0122: Reissued waste discharge requirements general order for existing milk cow dairies. Available at: https:waterboards.ca.gov/centralvalley/board_decisions/adopted_orders/general_orders/r5-2013-0122.pdf (Accessed January 15, 2020).

31.

Rajendran

, and Murthy

G.S.

(2019). Techno-economic and life cycle assessments of anaerobic digestion—A review. Biocatal. Agric. Biotechnol. 20, 101207.

32.

Ransom

K.M.

, Bell

A.M.

, Barber

Q.E.

, Kourakos

, and Harter

(2018). A Bayesian approach to infer nitrogen loading rates from crop and land-use types surrounding private wells in the Central Valley, California. Hydrol. Earth Syst. Sci. 22, 2739.

33.

Reza

M.T.

, Andert

, Wirth

, Busch

, Pielert

, Lynam

J.G.

, and Mumme

(2014). Hydrothermal carbonization of biomass for energy and crop production. Appl. Bioenergy, 1, 11.

34.

Reza

T.M.

, Freitas

, Yang

, Hiibel

, Lin

, and Coronella

C.J.

(2016). Hydrothermal carbonization (HTC) of cow manure: Carbon and nitrogen distributions in HTC products. Environ. Prog. Sustain. Energy, 35, 1002.

35.

Richmond

, and Grobbelaar

J.U.

(1986). Factors affecting the output rate of Spirulina platensis with reference to mass cultivation. Biomass, 10, 253.

36.

Rosenstock

T.S.

, Liptzin

, Dzurella

, Fryjoff-Hung

, Hollander

, Jensen

, King

, Kourakos

, McNally

, Pettygrove

G.S.

, Quinn

, Viers

J.H.

, Tomich

T.P.

, and Harter

(2014). Agriculture's contribution to nitrate contamination of Californian groundwater (1945–2005). J. Environ. Qual. 43, 895.

37.

Salls

K.A.

, Won

, Kolodeziej

E.P.

, Childress

A.E.

, and Hiibel

S.R.

(2018). Evaluation of semi-volatile contaminant transport in a novel, gas-tight direct contact membrane distillation system. Desalination, 427, 35.

38.

Silva

N.A.

, and Hiibel

S.R.

(2021). Nutrient recovery of the hydrothermal carbonization aqueous product from dairy manure using membrane distillation. Environ. Technol. [Epub ahead of print]; DOI: 10.1080/09593330.2021.1995785.

39.

State of California (2016). Senate Bill No. 1383. Available at: https:leginfo.legislature.ca.gov/faces/billNAVClient.xhtml?bill_id=201520160SB1383 (Accessed August 5, 2021).

40.

U.S. Environmental Protection Agency (U.S. EPA). (2013). Literature review of contaminants in livestock and poultry manure and implications for water quality. EPA 820-R-13-002. Available at: https:nepis.epa.gov (Accessed May 27, 2020).

41.

U.S. Environmental Protection Agency (U.S. EPA). (2017). Inventory of U.S. greenhouse gas emissions and sinks: 1990–2017. Chapter 5: Agriculture. Available at: https:epa.gov/ghgemissions/inventory-us-greenhouse-gas-emissions-and-sinks-1990-2017 (Accessed February 3, 2020).

42.

U.S. Life Cycle Inventory Database. (2012). National Renewable Energy Laboratory. Available at: https://www.nrel.gov/lci (Accessed August 6, 2021).

43.

Wernet

, Bauer

, Steubing

, Reinhard

, Moreno-Ruiz

, and Weidema

(2016). The ecoinvent database version 3 (part I): Overview and methodology. Int. J. Life Cycle Assess. 21, 1218.

44.

, Cheng

L.C.

, and Chang

J.S.

(2020). Environmental life cycle comparisons of pig farming integrated with anaerobic digestion and algae-based wastewater treatment. J. Environ. Manag. 264, 110512.

45.

Zhang

, White

M.A.

, and Colosi

L.M.

(2013). Environmental and economic assessment of integrated systems for dairy manure treatment coupled with algae bioenergy production. Bioresour. Technol. 130, 486.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.11 MB