Comparison of carbon footprints of steel versus concrete pipelines for water transmission

ABSTRACT

The global demand for water transmission and service pipelines is expected to more than double between 2012 and 2022. This study compared the carbon footprint of the two most common materials used for large-diameter water transmission pipelines, steel pipe (SP) and prestressed concrete cylinder pipe (PCCP). A planned water transmission pipeline in Texas was used as a case study. Four life-cycle phases for each material were considered: material production and pipeline fabrication, pipe transportation to the job site, pipe installation in the trench, and operation of the pipeline. In each phase, the energy consumed and the CO₂-equivalent emissions were quantified. It was found that pipe manufacturing consumed a large amount of energy, and thus contributed more than 90% of life cycle carbon emissions for both kinds of pipe. Steel pipe had 64% larger CO₂-eq emissions from manufacturing compared to PCCP. For the transportation phase, PCCP consumed more fuel due to its heavy weight, and therefore had larger CO₂-eq emissions. Fuel consumption by construction equipment for installation of pipe was found to be similar for steel pipe and PCCP. Overall, steel had a 32% larger footprint due to greater energy used during manufacturing.

Implications: This study compared the carbon footprint of two large-diameter water transmission pipeline materials, steel and prestressed concrete cylinder, considering four life-cycle phases for each. The study provides information that project managers can incorporate into their decision-making process concerning pipeline materials. It also provides information concerning the most important phases of the pipeline life cycle to target for emission reductions.

Introduction

Pipelines provide substantial benefits for modern society, as they are the main source for transport of water to serve our communities. Water pipelines range from just a few inches to more than 20 feet in diameter and supply water over short and long distances via water distribution and water transmission, respectively. In the United States, increasing population and enormous industrial growth are putting increased pressure on aging water transmission pipelines, which will need to be rehabilitated or replaced in the near future. Globally, installation of new water pipelines will be required to serve the 750 million people on earth who currently do not have access to a safe and reliable water supply (Water Organization, 2014). The global demand for water transmission and service pipelines, which was 979 million meters in 2012, is expected to more than double to 2000 million meters in 2022. China will account for a large portion of this increase, along with other industrializing countries in Asia, including India and Indonesia, and in Africa and the Middle East (Freedonia Group, 2013).

Pipelines impact the environment and carbon emissions in particular, due to various energy-intensive activities, including pipe manufacturing, pipe transportation to the installation site, use of construction equipment during pipe installation, and pipeline operation and maintenance (Filion, 2004; Venkatesh and Brattebo, 2011). Installation of new water transmission and distribution pipelines presents an opportunity to choose materials, installation techniques, and operation techniques that reduce carbon emissions.

Few studies, however, have been conducted of pipeline carbon emissions. Piratla et al. (2011) estimated life-cycle CO₂ emissions from an underground 8-inch-diameter potable water distribution pipeline. The RSK Group quantified the climate impacts associated with the materials acquisition, construction, operation, and final disposal phases of a proposed 48-inch-diameter, 316-km natural gas steel pipeline between Felindre and Tirley, United Kingdom (National Grid, 2011). Several studies (Sihabuddin and Ariaratnam, 2009; McPherson, 2009) in the past have discussed the life-cycle analysis of steel, ductile iron, and plastic pipes; however, they typically did not address energy consumption and CO₂ emissions related to large-diameter water transmission pipelines, which are commonly made from steel or prestressed concrete. Plastic pipes are typically only used up to a diameter of 2 feet, and ductile iron up to 4 feet. Steel pipe (SP) and prestressed concrete cylinder pipe (PCCP) diameters can be as large as 12 feet for water arterial mains (Nathanson, 2014). Figure 1 shows the pie chart distribution for 21 U.S. water utilities, serving a population of approximately 14 million (CUIRE, 2013).

Figure 1. U.S. pipe materials usage (CURIE, 2013).

The objective of the study presented here is thus to compare greenhouse gas emissions associated with the two primary pipe materials used for large-diameter water transmission pipelines, steel pipe and prestressed concrete cylinder pipe. World demand for steel water pipe is expected to more than double from 231 to 503 million meters from 2012 to 2022; similarly, world demand for concrete water pipe is expected to double from 80 to 161 million meters from 2012 to 2022 (Freedonia Group, 2013). The carbon footprint comparison of steel pipe and PCCP will provide information that project managers can incorporate into their decision-making process concerning pipeline materials. It will also provide information concerning the most important phases of the pipeline life cycle to target for emission reductions.

Methodology

An ongoing 150-mile water transmission pipeline project, called the Integrated Pipeline (IPL) Project, located in North Central Texas, was used to compare greenhouse gas emissions. The pipeline consists of eight segments A–I of various diameters ranging from 60 inches to 108 inches, as shown later in Table 2.

Table 1. Embodied energy coefficients for pipe component materials (Hammond and Jones, 2011).

Pipe component material	Embodied energy coefficient
	MJ/kg	kWh/lb
Steel pipe (world average recycled content 39%)	24.90	3.14
Concrete (5000 psi)	0.95	0.12
Cement mortar (1:3 cement:sand mix)	1.33	0.17
Prestressed steel wires	36.00	4.54

Table 2. Embodied energy and CO₂ emissions for pipe segments, steel pipe vs. PCCP.

Parameter	Subparameter	Unit	Pipeline segment								Total
			A	B	C	D	E	F	G	I
Segment length		ft	220,394	26,159	329,388	114,131	8,517	57,768	7,120	14,765	778,242
Pipe section diameter		inches	84	108	108	84	72	66	108	84	N/A
Steel pipe	Embodied energy per foot	kWh/ft	899	1607	1607	899	770	706	1607	899	N/A
	Embodied energy—total segment	10^6 kWh	198.3	42.1	529.6	102.7	6.6	40.8	11.4	13.3	944.8
	Wsteel	lb/ft	280	503	503	280	240	220	503	280	N/A
	Wcement	lb/ft	129	165.70	165.70	129	110.72	101.56	165.70	129	N/A
	CO2 emissions	short tons	113,725	24,144	303,726	58,898	3,785	23,399	6,538	7,628	541,843
		lb/ft	1032	1846	1844	1032	889	810	1836	1033	1392
PCCP	Embodied energy per foot	kWh/ft	794	1058	1058	794	528	478	1058	794	N/A
	Embodied energy—total segment	10^6 kWh	175.0	27.7	348.8	90.7	4.5	27.7	7.5	11.7	693.6
	Wsteel	lb/ft	56.11	71.66	71.66	56.11	48.45	44.3	71.66	56.11	N/A
	Wcement	lb/ft	341.73	442.49	442.49	341.73	309.25	285.26	442.49	341.73	N/A
	Wconcrete	lb/ft	1516.16	2612.38	2612.38	1516.16	1312.54	1148.15	2612.38	1516.16	N/A
	Wsteel wire	lb/ft	83.26	98.05	98.05	83.26	36.67	33.78	98.05	83.26	N/A
	CO2 emissions	short tons	100,363	15,886	200,037	52,016	2,581	15,886	4,301	6,710	397,780
		lb/ft	911	1215	1215	912	606	550	1208	909	1022

The pipe diameters, based on hydraulic analysis, were required to be in the range of 66–108 inches. Steel and concrete are the only options for such large-diameter pipes. American Water Works Association (AWWA) pipe design guidelines were used in this study. The AWWA M11 and C200 manual outlined the standard design and manufacturing guidelines for steel pipe (AWWA, 2004, 2005). AWWA M9, C301, and C304 standards were used for PCCP (AWWA, 2007, 2008).

Four life-cycle phases for each material were considered in this study: material production and pipeline fabrication, pipe transportation to the job site, pipe installation in the trench, and operation of the pipeline. In each phase, the energy consumed and the CO₂ emissions were quantified, for the entire 150 miles of pipeline, and per foot of pipeline (functional unit being foot of pipeline). Each phase is discussed in more detail in the following.

Material production and pipeline fabrication phase

The database entitled Inventory of Carbon and Energy (ICE) version 2.0 was used for this phase (Hammond and Jones, 2011). This database provides an embodied energy (MJ/kg) for each construction material. Although ICE contains both embodied energy and carbon data, the embodied energy coefficients are more accurate (Hammond and Jones, 2008). The system boundary in ICE version 2.0 is generally cradle-to-gate, or from resource extraction to the factory gate, before the product is transported to the consumer. The next step in the LCA is transport of the pipeline segments to the installation location. Table 1 presents the embodied energy coefficients for steel pipe and PCCP components from ICE.

The embodied energy per foot of steel pipe, EE _{Steel Pipe}, was calculated as per eq 1. Pipe cross sections have been provided in Figure 2 to show the presence of various materials.

(1)

Figure 2. Pipe cross sections: (a) steel and (b) PCCP.

where EE_Steel and EE_{Cement Mortar} are embodied energy coefficients for steel and cement mortar, kWh/lb; and W_Steel and W_{Cement Mortar} are weight of steel and cement mortar used per foot of pipeline, lb/ft.

Weight of steel and cement mortar per foot of pipeline was provided by Hanson, a PCCP manufacturer. Table 2 shows EE_{Steel Pipe} values for the various pipeline diameters. EE_{Steel Pipe} was then multiplied by length of each pipe segment to find the total embodied energy for that segment in kilowatt-hours, as summarized in Table 2. For ERCT, the eGRID subregion that includes most of Texas, 2012 emission rates were 1143 lb CO₂/MWh, 16.70 lb methane/GWh, and 12.33 lb nitrous oxide (N₂O)/GWh (EPA, 2015a). The global warming potential values of methane and nitrous oxide are 28 and 265 times that of carbon dioxide, respectively, meaning that their emission rates must be multiplied by 28 and 265 to estimate CO₂ equivalents (EPA, 2015b). Thus, emissions of CO₂ equivalents for Texas are 1.147 lb/kWh, including contributions from CO₂, methane, and nitrous oxide. The embodied energy was therefore multiplied by 1.147 lb CO₂-eq/kWh of electricity usage to find CO₂-eq emissions for each segment.

Similarly, the embodied energy per foot of PCCP was calculated using eq 2:

(2)

where EE_Concrete, EE_Steel, EE_{Cement Mortar}, and EE_{Steel Wire} are the embodied energy coefficients for concrete, steel, cement mortar, and prestressed steel wires, and W_Concrete, W_Steel, W_{Cement Mortar}, and W_{Steel Wire} are the unit weight per foot length for concrete, steel, cement mortar, and prestressed steel wires, respectively.

Table 2 summarizes embodied energy and CO₂-eq emission values for each pipeline segment for steel pipe and PCCP.

Pipe transportation phase

This phase includes calculation of total fuel consumption during transport of the pipe to the actual job site, based on total distance traveled. Table 3 presents the length and weight of each pipe section for different segments. A 48- to 53-foot flat-bed truck–trailer combination was selected on the basis of pipe section lengths and weights (16 feet and 24,000–52,000 lb for PCCP; 50 feet and 11,000 to 25,000 lb for steel pipe). Each truck–trailer unit could carry one steel pipe because of the length restriction of 50 feet and one PCCP due to weight restriction of maximum 52,000 lb. The truck–trailer, with fuel consumption of 5.4 miles per gallon (mpg; Davis and Diegel, 2010), will be able to carry one pipe section per trip; the number of trips required is thus equal to the number of pipe sections, calculated by dividing the segment length by the pipe section length.

Table 3. Diesel fuel required and CO₂ emissions for pipe transportation, steel pipe vs. PCCP.

Parameter	Subparameter	Unit	Pipeline segments								Total
			A	B	C	D	E	F	G	I
Segment length		ft	220,394	26,159	329,388	114,131	8,517	57,768	7,120	14,765	778,242
Pipe section diameter		inches	84	108	108	84	72	66	108	84	N/A
Steel pipe	Pipe section weight	lb.	13,979	25,153	25,153	13,979	11,974	10,972	25,153	13,979	140,342
	Pipe section length	ft	50	50	50	50	50	50	50	50	N/A
	Number of pipe sections = number of truck trips	number	4,408	523	6,588	2,283	170	1,155	142	295	15,564
	Distance from production plant to segment center	miles	240	228	162	54	222	234	72	58	N/A
	Diesel required for truck–trailer round trips	gallon	182,404	20,604	185,189	21,278	6,546	46,645	1,801	3,014	467,481
	CO2 emissions	short tons	2,025	229	2,056	236	73	518	20	33	5,190
		lb/ft	18.4	17.5	12.5	4.1	17.1	17.9	5.6	4.5	13.3
PCCP	Pipe section weight	lb	39,945	51,593	51,593	39,945	40,961	24,184	51,593	39,945	339,759
	Pipe section length	ft	16	20	20	16	24	16	16	20	N/A
	Number of pipe sections = number of truck trips	number	13,775	1,308	16,470	7,134	355	3,611	445	739	43,837
	Distance from production plant to segment center	miles	150.2	182.2	116.6	73.8	173.8	169.2	57.6	56.4	N/A
	Diesel required for all truck–trailer round trips	gallon	286,961	51,601	415,446	72,822	10,650	105,875	4,450	7,265	955,070
	CO2 emissions	short tons	3,186	573	4,612	808	118	1,175	49	81	10,602
		lb/ft	28.9	43.8	28.0	14.2	27.7	40.7	13.8	11.0	27.2

The steel pipe was assumed to be supplied by the Northwest Pipe Company, Saginaw, TX, and PCCP was assumed to be supplied by Hanson Pipe & Precast, Grand Prairie, TX. Locations of these facilities were used to determine the distance between the production plant and the segment center or midpoint, as given in Table 3. Total diesel consumption was calculated according to eq 3, which includes a factor of 2 since the truck must travel to and from the production plant each time it delivers a pipe section:

(3)

The diesel consumption was converted into CO₂ emissions using an emission factor of 22.2 lb CO₂ per gallon of diesel fuel (EPA, 2005). The carbon dioxide emission factors for diesel fuel consumption assume that 99% of the fuel is oxidized to form CO₂, so any methane emissions of methane or nitrous oxide would be minimal.

Pipe installation phase

For this study, only major construction activities were considered, such as excavation, loading, backfilling, hauling, and compaction. The production rate of construction equipment depends on several factors listed below, related to the pipeline itself and trench, surrounding soil, and equipment used listed next.

Factors related to pipeline:

• Weight of each pipe section;

• Length of each pipe section;

• Trench cross-section details;

• Volume of earthwork (excavation and backfill).

Factors related to soil:

• Topography of right-of-way;

• Type of soil;

• Soil compaction level.

Factors related to equipment:

• Age;

• Capacity;

• Power requirement;

• Efficiency.

An increase in equipment production efficiency, or decrease in operational hours, will proportionally decrease emissions, and thus decrease the carbon footprint. Production efficiency of equipment will vary by weight and length of a pipe section. Smaller pipe sections will consume less time for installation than larger sections. The right-of-way should allow uninterrupted installation activities for greater productivity (Dey, 2006). Soft soil is more easily excavated compared to hard soil or rock; moreover, soil that is lightly compacted will require less effort to excavate, compared with soil that is highly compacted. For a given soil type and density, the greater the volume of earthwork required, the greater is the number of hours of construction equipment use. Newer equipment with greater capacity, power, and efficiency will have a faster production rate.

Additional factors impacting production rate include hauling distance (shorter distance increases production rate), water table (higher water table decreases production rate), weather conditions (good weather increases production rate), and operator efficiency (the experienced person will take less time and be more efficient than a less experienced and inefficient person).

All of the factors impacting construction equipment productivity mentioned already would be the same for steel and PCCP pipeline, except for factors related to the pipes themselves, such as length and weight. Another example of a difference related to the pipes themselves is that joining two segments in the field is required for steel pipes but only a push for connecting concrete pipe segments via a rubber gasket. Such differences in installation were accounted for by using the different lengths of time for construction equipment operation of various types for the two kinds of pipe (Tables 4 and 5).

Table 4. Hours of construction equipment needed and CO₂ emissions for installation of steel pipe.

			Duration	Fuel consumption	CO2 emissions
Activity	Quantity needed for the project	Productivity per hour	(hr)	(1000 gal)	Short tons	lb/ft	%
Excavation, bcy	6,484,293	68.03	95,315	797.8	8,855.4	22.4	33.0
Trench support, feet	778,242	30.5	25,516	213.6	2,370.6	6.0	8.8
Pipe laying, feet	778,242	138.37	5,624	47.1	522.5	1.3	1.9
Backfilling, lcy	6,762,888	96.47	70,104	586.8	6,513.2	16.4	24.2
Compaction, lcy	6,762,888	3,024.19	2,236	18.7	207.7	0.5	0.7
Material handling, lcy	6,762,888	310.70	21,767	182.2	2,022.3	5.1	7.5
Material screening, lcy	6,762,888	321.59	21,030	176.0	1,953.8	4.9	7.2
Water sprinkling, lcy	6,082,086	228.69	26,595	222.6	2,470.9	6.2	9.1
Sweeping, ft^2	116,736,300	5,604.09	20,831	174.4	1,935.3	4.9	7.1
Total			289,018	2419.1	26,852	67.8	100

Note: bcy = bank cubic yard and lcy = loose cubic yard.

Table 5. Hours of construction equipment needed and CO₂ emissions for installation of prestressed concrete cylinder pipe (PCCP).

Activity	Amount needed for the project	Productivity per hour	Duration	Fuel consumption	CO2 emissions
			(hr)	(gal)	Short tons	lb/ft	%
Excavation, bcy	6,159,634	68.03	90,543	757.8	8,412.1	21.2	30.9
Trench support, feet	778,242	16.00	48,640	407.1	4,519	11.4	16.6
Pipe laying, feet	778,242	70.00	11,118	93.1	1,032.9	2.6	3.8
Backfilling, lcy	5,813,336	96.47	60,261	504.4	5,598.7	14.1	20.6
Compaction, lcy	5,813,336	3,024.19	1,922	16.1	178.6	0.5	0.7
Material handling, lcy	5,813,336	310.70	18,710	156.6	1,738.3	4.4	6.4
Material screening, lcy	5,813,336	321.59	18,077	151.3	1,679.5	4.2	6.1
Water sprinkling, lcy	5,074,435	228.69	22,189	185.7	2,061.5	5.2	7.6
Sweeping, ft^2	116,736,300	5,604.09	20,831	174.4	1,935.3	4.9	7.1
Total			292,291	2446.5	27,156	68.6	100

Note: bcy = bank cubic yard and lcy = loose cubic yard.

Tables 4 and 5 show hours of construction equipment operation needed for installation of steel pipe and PCCP, respectively. The hours were estimated by first estimating various quantities, such as excavation and backfill material, needed for the project for each kind of pipe. These quantities were then multiplied by productivities obtained from previous projects executed by Garney Construction for the Lake Eagle Mountain Pipeline Project conducted by Tarrant Regional Water District in 2006–2007 to determine the total time spent by various types of construction equipment at the job site (see Supplemental Information). Finally, an average fuel consumption per hour value (8.37 gal/hr) provided by the contractor was multiplied by hours at the job site to calculate the total fuel consumption for each activity. CO₂ emissions were estimated using an emission factor of 22.2 lb CO₂ per gallon of diesel fuel (EPA, 2005).

The following equation, eq 4, can be used to calculate CO₂ emissions for each activity for both pipes:

(4)

Pipeline operation phase

Energy consumption for operating the pipeline depends on various parameters independent of pipe material, such as water demand, season, terrain of pipeline, and pump’s efficiency, as well as parameters that depend on pipe material, including Manning’s coefficient, corrosion, and breaks and leakage in pipe (AWWA, 2011; McPherson et al., 2009). It was assumed that both pipes will be lined with cement mortar lining material inside, as this is considered the best lining material for water transmission pipelines (AWWA, 2012). Hence, energy consumption for both pipes will be same due to the same Manning’s coefficient of the lining material. Average annual energy consumption, as shown in Table 6, was estimated on the basis of energy consumption available from existing pipelines (Tarrant Regional Water District, 2010). The average annual energy consumption varies by decade, as water demand varies, causing changes in pumping energy requirements. The pipeline commissioning and operation are scheduled for the year 2020. Hence, the energy consumption was calculated from year 2020 to 2060, based on the project lifetime recommended by Tarrant Regional Water District. Finally, the energy consumption was multiplied by the Texas emission rate of 1.147 lb CO₂/kWh of electricity usage (EPA, 2015a) to find CO₂ emissions.

Table 6. Energy consumption and CO₂ emissions for pipeline operation, both steel and PCCP.

Value	Year 2020–2030	Year 2030–2040	Year 2040–2050	Year 2050–2060	Total 2020–2060
Average annual energy consumption (kWh)	160,753	124,103	165,534	165,544	N/A
Total energy consumption (kWh)	1,607,528	1,241,028	1,655,342	1,655,442	6,159,340
Total CO2 emissions (short tons)	922	712	949	949	3532
Total CO2 emissions (lb/ft)	2.3	1.8	2.4	2.4	8.9

The following equation, eq 5, can be used to calculate quantities in Table 6:

(5)

Results and discussion

CO₂ emissions for total four life cycle phases for both pipe materials are provided as Figure 3.

Figure 3. CO₂ emissions for different phases for steel pipe and PCCP.

Figure 4. Energy consumption for material production and pipeline fabrication for steel and PCCP.

For the material production and pipeline fabrication phase, it was found that total energy consumption for steel pipe was greater than for PCCP, as shown in Figure 4, for total length of the pipeline. The calculations for energy consumption and CO₂ emissions for each pipe segment are provided in Table 2. For the total pipeline, steel pipe CO₂ emissions were 36% greater than for PCCP (1392 vs. 1022 lb/ft, respectively). For each pipeline segment, the weight of materials used to make PCCP (steel, cement, concrete, and steel wire) was greater than the weight of materials used to make steel pipe (steel and cement). However, the embodied energy coefficient for steel is about 30 times higher than for concrete and cement (24.9 MJ/kg for steel vs. 0.95 MJ/kg for concrete and 1.33 for cement, as shown in Table 1). Steel production from virgin materials is energy intensive, due primarily to chemical energy required to reduce iron ore to iron using reducing agents (Worldsteel Association, 2014). Since steel pipe contained a higher fraction of steel, this meant that the embodied energy for each steel pipeline segment was greater than for PCCP.

As mentioned previously, the embodied energy data from the ICE database for concrete and steel pipe represent global averages. In particular, the recycled content used for the embodied energy coefficient of steel pipe in the database was 39% (world average). However, some of the U.S. steel manufacturers, such as Nucor Corporation, use higher recycled content in manufacturing their steel. Producing a product from recycled materials takes less energy than producing a product from virgin materials; hence, a higher content of recycled steel would reduce the embodied energy coefficient.

For the transportation phase, shown in Table 3, the distance that steel pipe would have to be transported from the production plant to the installation location would be greater for seven of the eight pipeline segments than the distance for PCCP. However, due to the heavy weight of PCCP, a larger number of truck–trailer trips would be required (43,837 trips for PCCP, compared to 15,565 trips for steel). Although the PCCP sections are only around one-third as long as the steel sections (16 feet for PCCP compared to 50 feet for steel), the PCCP sections weigh almost 3 times as much (39,945 pounds for each PCCP section, compared to 13,979 pounds for each steel section). The greater number of truck–trailer trips required for PCCP would lead to greater fuel consumption and CO₂ emissions that are over twice as high for PCCP (27.2 lb/ft) compared to steel pipe (13.3 lb/ft).

For the pipeline installation phase, as indicated in Tables 4 and 5 for steel pipe and PCCP, respectively, excavation and backfilling were the primary contributors to the fuel consumption and thus CO₂ emissions. Excavation contributed 33% and 31% of CO₂ emissions, and backfilling contributed 24% and 21%, for steel pipe and PCCP, respectively. Large amounts of excavation and backfilling were required, and the productivity of equipment used in excavation and backfilling was low, resulting in a large number of hours of operation being required. The large number of hours of operation resulted in high fuel consumption and high CO₂ emissions.

For the pipe installation phase, overall CO₂-eq emissions were 1.1% higher for PCCP (68.6 lb/ft) compared to steel pipe (67.8 lb/ft). This is due to the lower productivity of certain activities such as trench support and pipe laying for PCCP compared to steel pipe. However, pipe segment C for steel pipe had slightly higher values of fuel consumption and CO₂-eq emissions than PCCP due to longer durations for various activities such as excavation, pipe laying, compaction, and material handling and screening. This study assumed open cut installation for both kinds of pipe. Using alternate pipe installation methods, such as trenchless, would make a significant difference in the carbon footprint analysis.

Table 6 shows emissions from pipeline operation and maintenance over a 40-year time period. As explained in the Methodology section, it was assumed that both pipes will be lined with the same lining material inside, and thus that energy consumption for both pipes will be same due to the same Manning’s coefficient of the lining material. The average annual energy consumption varies by decade based on water demand. The annual energy consumption drops during the 2030–2040 decade, presumably due to demand decreasing due to additional sources of water becoming available. During 2040–2060, energy consumption again increases, presumably due to continued population growth in the North Texas region.

Table 7 and Figure 3 compare total CO₂-eq emissions for steel pipe and PCCP, for all four phases. The manufacturing and fabrication of the pipe contributed more than 90% of CO₂-eq emissions for both kinds of pipe. As mentioned earlier, the contribution of this phase was overestimated somewhat due to underestimation of percent recycled material used in steel pipe. However, even for concrete pipe, for which a higher recycling percent was assumed, more than 90% of emissions was due to the manufacturing and fabrication phase. This is primarily due to the embodied energy in the concrete. From a practical perspective, the only way to reduce embodied energy is to increase recycling percent (which is already high for concrete), or switch to another material with a lower embodied energy.

Table 7. Life-cycle CO₂ emission summary for steel pipe and PCCP.

	Steel pipe CO2 emissions			Prestressed concrete cylinder pipe (PCCP) CO2 emissions
Life-cycle phase	Short tons	lb/ft	%	Short tons	lb/ft	%
Manufacturing and fabrication	541,843	1392	93.9	397,780	1022	90.7
Transportation	5,190	13	0.9	10,602	27	2.4
Installation	26,531	67	4.6	26,763	68	6.1
Operation	3532	8.9	0.6	3532	8.9	0.8
Total	577,096	1,481	100	438,677	1,126	100

The phase with the second largest contribution to CO₂-eq emissions, for both kinds of pipe, was installation. This contribution could be reduced by manufacturers increasing the productivity and/or fuel economy of the construction equipment. Alternative pipe installation techniques, such as trenchless technology, could also potentially reduce this contribution.

Total emissions for steel pipe exceeded that for PCCP by 32%, primarily due to the material production and pipeline fabrication phase, and the greater energy required to obtain raw materials for steel pipe and manufacture it. Even if the percent use of recycled steel were increased, total emissions for steel pipe would likely still exceed that for PCCP. Thus, to reduce CO₂-eq emissions, this case study suggests that PCCP should be chosen instead of steel pipe.

Study assumptions and recommendations for future work

This study examined CO₂-eq emissions from one particular case of a water transmission pipeline. Pipe lengths and weights, which impact emissions for several of the life cycle phases, were thus particular to this project.

CO₂-eq emissions from the material production and pipeline fabrication phase predominated in the total emissions. CO₂-eq emissions for this phase came from global averages, including global averages for percent recycled content. CO₂-eq emissions for this phase could thus vary considerably if emissions specific to a particular pipeline manufacturer were used, rather than global averages. A particular manufacturer may use different material extraction methods and manufacturing technologies. Carbon emissions from electricity used to power the manufacturing process were specific to the ERCT U.S. subregion; a pipeline located in a different region of the United States, or outside the United States, would produce different emissions due to a different mix of fuels used to generate electricity.

For the transportation phase, distances that the pipe segments had to be transported were specific to this project. If pipe segments had to be transported longer distances on a different project, the transportation phase would be more significant in the overall life-cycle emissions. The transportation phase also assumed use of diesel fuel, and current fuel efficiencies. Use of alternate fuels, in the future in the United States, or currently in other parts of the world, would impact transport emissions, as would future increases in vehicle fuel efficiencies.

For the pipeline installation phase, hours of operation for construction equipment were based on quantities specific to the project, and equipment productivity specific to one company. These values would be expected to vary for other projects. The installation phase also assumed use of diesel fuel, and current fuel efficiencies. Use of alternate fuels or increases in vehicle fuel efficiencies would impact installation phase emissions. Finally, this study assumed open cut installation for both kinds of pipe. Alternate pipe installation methods would impact emissions from the installation phase.

Emissions from pipeline operation were specific to the pipe diameters, lengths, lining materials (cement mortar), water demand levels, and electricity generation subregion used in this study. Variations in these factors would impact emissions from the operation phase.

The pipeline operation phase for both kinds of pipe was assumed to last 40 years, which was the project lifetime specified by Tarrant Regional Water District (2020–2060). In actuality, the two kinds of pipe would likely have different life spans. Each time a pipeline has to be replaced, another large amount of emissions would occur from material production and fabrication. The pipeline that has a longer life span would need replacement less frequently, which would reduce its emissions due to material production and fabrication over the long term. Unfortunately, for multiple reasons, we were not able to viably compare PCCP and steel pipeline lifetimes in order to be able to include their life span in the analysis. Most PCCP water pipelines, installed in the 1940s in the United States, are still in service and has not yet reached the end of its lifetime; therefore, few data are available on PCCP life spans. For steel pipe, industries have shifted design requirements from wall thickness to the soil rather than pipe stiffness (Najafi and Gokhale, 2005). Consequently, older steel pipe with thicker walls would last longer than today’s steel pipe, due to sufficient steel to resist corrosion and extend its life span. Few data are available on the lifetime of steel pipe used today, because it has not yet reached the end of its lifetime. Furthermore, advances in condition assessment technologies for PCCP over the last 15 years have allowed owners to conduct PCCP surveys to determine PCCP pipeline conditions (Najafi, 2003). No similar technologies exist for steel pipe. Thus, it is difficult to provide a meaningful comparison of the anticipated lifetimes of newly installed PCCP and steel pipe.

A sensitivity analysis is recommended as future work, to examine how varying assumed factors already mentioned, for the four life-cycle phases, impact overall emissions. In addition, pipeline inspection and maintenance and the end-of-life phase were not included in this study, and could be examined in future work.

Regarding inspection and maintenance, steel pipe and concrete have differing requirements because of the way they are designed. Steel pipe is considered as flexible pipe, which can deflect under loads; concrete pipe is considered as rigid pipe. Therefore, the pipe–soil interaction and load transferring mechanism are different for the two kinds of pipe. Furthermore, steel pipes are joined in the field using welding, which can lead to failures due to poor quality or degradation, while concrete pipes are joined using bell and spigot, which eliminates the need for welding but has its own limitations. Finally, steel and precast concrete pipes are both subject to corrosion; therefore, cathodic protection along with frequent field inspection and maintenance is required. An in-depth study of inspection and maintenance frequency and material requirements is recommended in order to be able to quantify environmental impacts from the inspection and maintenance phase.

Conclusions

Among the four phases of the pipeline life cycle, the manufacturing and fabrication proved to be the most influential, contributing more than 90% of life cycle carbon emissions for both kinds of pipe. Steel pipe had 36% larger CO₂-eq emissions from manufacturing compared to PCCP. For the transportation phase, PCCP consumed more fuel due to its heavy weight, and therefore had larger CO₂-eq emissions. Fuel consumption by construction equipment for installation of pipe was found to be similar for steel pipe and PCCP. Overall, steel had a 32% larger footprint due primarily to greater energy used during manufacturing. Even if the percent use of recycled steel were increased, total emissions for steel pipe would likely still exceed that for PCCP. Thus, to reduce CO₂-eq emissions, this case study suggests that PCCP should be chosen instead of steel pipe.

Funding

The authors thank Tarrant Regional Water District (TRWD) and Dallas Water Utilities (DWU) for the invaluable support and funding provided for this study. The authors would also like to thank Northwest Pipe Company, Hanson Pipe & Precast, and Garney Construction for providing information throughout this study. The authors thank the China Section of the Air & Waste Management Association for the generous scholarship they received to cover the cost of page charges and make the publication of this paper possible.

Supplemental data

Supplemental data for this paper can be accessed on the publisher’s Web site.

Additional information

Notes on contributors

Lalit Chilana

Lalit Chilana is a senior pipeline engineer at Gulf Interstate Engineering.

Arpita H. Bhatt

Arpita H. Bhatt is a postdoctoral fellow in the Civil Engineering Department at the University of Texas at Arlington, TX.

Mohammad Najafi

Mohammad Najafi is a professor and director of construction management program in the Civil Engineering Department at the University of Texas at Arlington, TX.

Melanie Sattler

Melanie Sattler is an associate professor in the Civil Engineering Department at the University of Texas at Arlington, TX.