Carbon footprint of sugar production in Mexico

Carbon footprint of sugar production in Mexico

Journal of Cleaner Production 112 (2016) 2632e2641 Contents lists available at ScienceDirect Journal of Cleaner Production journal homepage: www.els...

921KB Sizes 0 Downloads 21 Views

Journal of Cleaner Production 112 (2016) 2632e2641

Contents lists available at ScienceDirect

Journal of Cleaner Production journal homepage: www.elsevier.com/locate/jclepro

Carbon footprint of sugar production in Mexico ~ o b, Noe Aguilar-Rivera c, Cynthia Armenda riz a Carlos A. García a, *, Edgar S. García-Trevin n, Mexico noma de M Escuela Nacional de Estudios Superiores Unidad Morelia, Universidad Nacional Auto exico, Michoaca Universidad de las Am ericas Puebla, Puebla, Mexico c n S/N, Colonia Pen gicas y Agropecuarias Universidad Veracruzana, Km. 1 carretera Pen ~ uela Amatla ~ uela, C.P. 94945, Co rdoba Facultad de Ciencias Biolo Veracruz, Mexico a

b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 22 October 2014 Received in revised form 23 September 2015 Accepted 24 September 2015 Available online 9 October 2015

Global warming, caused mainly by increased worldwide emissions of greenhouse gases, is currently one of the greatest threats to the environment and human societies. Mexico has set an ambitious goal of reducing 30% of its greenhouse gases emissions by year 2020. The sugar agroindustry has been identified as one of the opportunities for mitigating emissions in this country. The aim of this work is to contribute towards identifying policy measures and practices for low-carbon sugar production by assessing the carbon footprint of sugar produced in four sugar mills in Mexico using a life cycle assessment method. System boundaries include agricultural practices, sugarcane harvesting, cane milling and sugar conversion. The results show that sugar production has carbon footprint values in the range of 0.45e0.63 kg CO2e/kg sugar. In these four cases, the agricultural stage contributes the majority of carbon emissions (59e74%). Most greenhouse gases emissions in the agricultural stage were from fertilizer production, nitrous oxide (N2O) emissions and biomass burning. The industrial stage contributed with 14e30% of total greenhouse gases emissions, mainly due to fossil fuel and bagasse use. The carbon footprint value is particularly sensitive to nitrogen fertilization, nitrous oxide emissions from the soil and sugarcane yields. Cogeneration in sugar mills could become an important way to reduce the carbon footprint of sugar and to produce electricity with low carbon emissions. We show the impact of different carbon footprint performance of sugar production process in Mexico. Data used on this manuscript came from real field measurements, and our results are accompanied by sensibility and uncertainty analyses. This is the first time that life cycle assessment has been used to estimate the carbon footprint of sugar production in Mexico including agricultural, industrial and transportation boundaries, to identify greenhouse gases mitigation opportunities. Studying techniques for improving sugar cane yield, making fertilizer use more efficient, minimizing cane burning and developing efficient cogeneration in sugar mills with bagasse as fuel is scientifically relevant. Applying concrete public policy measures to these areas of opportunity would allow production of low carbon sugar in Mexico. The results of this study may also be used as reference by other countries with similar sugar production conditions. © 2015 Elsevier Ltd. All rights reserved.

Keywords: Global warming Greenhouse gas emissions Sugarcane cultivation Sugar mill Cogeneration Uncertainty analysis

1. Introduction Reversing the effects of environmental degradation caused by human activities is one of the greatest challenges currently faced by €m et al., 2009). Global warming, which is humanity (Rockstro

* Corresponding author. Escuela Nacional de Estudios Superiores Unidad Morelia, noma de Me xico, Antigua Carretera a Pa tzcuaro No. 8701, Universidad Nacional Auto  de la Huerta, P.C. 58190 Mexico. Tel.: þ52 443 Colonia Ex Hacienda de San Jose 6893500, Tel/fax.: þ52 443 3222719. E-mail address: [email protected] (C.A. García). http://dx.doi.org/10.1016/j.jclepro.2015.09.113 0959-6526/© 2015 Elsevier Ltd. All rights reserved.

mainly caused by the increased use of energy derived from fossil fuels, is one of the most concerning phenomena that could have severe effects on the environment and human societies (IPCC, 2014a, 2013). Mexico's Climate Change Law was enacted as part of this nation's efforts to combat climate change (Diario Oficial de la n, 2012). Here, an ambitious goal of reducing 30% of Federacio greenhouse gas (GHG) emissions by year 2020 was established. This has led to the development of a number of studies that evaluate possibilities for mitigating GHG emission in this economy (Johnson et al., 2010; Octaviano et al., n.d.; Veysey et al., n.d.). Some

C.A. García et al. / Journal of Cleaner Production 112 (2016) 2632e2641

of these studies point towards the sugar agroindustry as an opportunity for GHG emission mitigation (Islas et al., 2007; Johnson et al., 2010; Riegelhapt et al., 2012). Mexico is the world's seventh largest sugarcane (Saccharum officinarum) producer. Spanish settlers brought the first sugarcane plants to Mexico from Cuba in the early 16th century, and the first sugar mill was established in 1536 in the village of San Andres Tuxtla, Veracruz. Nowadays, cultivation of sugarcane is concentrated in six different regions in Mexico, which provide sugarcane to 54 sugar mills and two autonomous distilleries (UNC, 2014). Around 61 million tons per year of sugarcane are cultivated in 780,254 ha, which produce nearly 7 million tons of sugar (3% of worldwide production) and about 16.7 million liters (ML) of ethanol. The sugarcane agroindustry represents 0.5% of Mexico's gross domestic product (GDP), 2.5% of the manufacturing sector and 11.5% of the primary sector. It also provides significant full time and temporary employment for more than 2.2 million people in 227 municipalities (9.2% of all municipalities in Mexico) (SentíesHerrera et al., 2014). With the increasing threat of global warming and climate change, minimizing carbon emissions during product elaboration has become increasingly important in recent years (Fang et al., 2014; Wiedmann and Minx, 2007). Researchers have quantified GHG emissions using an indicator known as the carbon footprint (CF), which is the sum of all direct and indirect GHG emissions generated throughout the life cycle of a product (Wiedmann and Minx, 2007). To evaluate the CF, a life cycle assessment (LCA) method has been applied. LCA is a useful standardized method (with ISO 14040 and 14044 standards) for estimating the environmental impact of processes and products. This tool has been widely employed to identify products with fewer negative impacts on the environment, or to locate the production stages where the greatest environmental impacts occur. Numerous studies have been done in recent years involving the use of LCA for carbon emissions quantification in agroindustry products. Relevant studies have assessed the CF of sugar produced from different feedstocks (De Figueiredo et al., 2010; Klenk et al., 2012; Rein, 2010; Seabra et al., 2011; Yuttitham et al., 2011), calculated carbon emissions of ethanol and electricity produced from sugarcane (Campbell et al., 2009; Dunkelberg et al., 2014; Khatiwada and Silveira, 2011; Nguyen et al., 2010; Ramjeawon, 2008; Seabra et al., 2011; Soam et al., 2015); compared environmental impacts of sugarcane product diversification (Renouf et al., 2013); and analyzed the influence of methodological variations on results of carbon emissions (Cherubini and Strømman, 2011; Cherubini et al., 2009; Plassmann et al., 2010). The aim of this paper is to assess the CF of sugar produced in four sugar mills in Mexico using LCA method. This is the first assessment of how sugar production contributes towards carbon emissions in Mexico that also identifies the main sources of GHG emissions during the production cycle. Additionally, uncertainty and sensitivity analysis were performed using Monte Carlo Simulation, and two scenarios were developed to explore the impact of efficient cogeneration on the CF of sugar production. This information is useful because it contributes towards identifying concrete policy measures and practices for low-carbon sugar production, which in turn brings Mexico closer to its mitigation goals. The results of this study may also be used as reference by other countries with similar sugar production conditions. 2. Materials and methods In this section we present the methods, data and information sources used to calculate the carbon footprint (CF) in four case

2633

studies. We also explain the methods for performing uncertainty and sensibility analysis, through which the changes in CF due to variations in the initial system parameter values were calculated. Finally, we explore the effects that efficient cogeneration in sugar mills can have on the CF. 2.1. Cases The criteria used to select the cases for assessment were: 1) That the cases have similar sugarcane milling capacities; 2) That they cover different geographic regions; 3) That data for all life cycle stages were available for assessment. The latter criterion was particularly important because in Mexico there are no databases containing information on fuel consumption for agricultural and transportation machinery. The assessment was limited to four sugar mills that fulfill these criteria. Table 1 shows the location, total industrialized area and mean sugar production of the four case studies used to calculate the carbon footprint of sugar production in Mexico. Note that these mills might not be a statistically representative sample of all mills in Mexico, since significant differences exist in levels of technology in cane production (from low to high mechanization and fertilization), in industrialized area (ranging from about 2400e39,000 ha/y) and in net cane crushed (229,000e2,200,000 ton cane/y). Sugarcane is generally cultivated in five-year cycles followed by replanting. Cane yields vary between 64 and 106 t/ha a year. Irrigation is by gravity or pumping (or both) while soil tillage is fully mechanized and harvest is mostly manual in almost all our cases. This last practice makes it necessary to burn sugarcane trash in the fields before and after harvesting to ease the cutter's work and to clear residues. Truck loading is mechanized. The industrial process starts with sugarcane crushing to extract juice, which is then clarified and concentrated. On the concentrated juice, up to three successive crystallizations are carried out, followed by separation of sucrose crystals and molasses by centrifugation (sugar and molasses are produced in each centrifugation). The latter can be used for ethanol production. 2.2. Life cycle assessment The life cycle assessment (LCA) method can be used to measure total environmental performance of a product from cradle to grave (Khatiwada and Silveira, 2011). This method accounts for energy and material inputs required in the development of a product as well as by-products and emissions that occur during the production process. In our four case studies, the system boundaries are consistent with other studies and include agricultural practices, sugarcane harvesting, cane milling and sugar conversion (Fig. 1) (Khatiwada and Silveira, 2011; Nguyen et al., 2010; Yuttitham et al., 2011). The agricultural stage includes the following sources of emissions: From fuel used for farm machinery, from the production of fertilizers and pesticides, nitrogen monoxide (N2O) emissions from nitrogen fertilizer volatilization (calculated as 1% of the N applied; IPCC, 2006), emissions from energy for irrigation, from harvest machinery and from sugarcane burning for harvesting. The industrial stage (sugar milling and sugar conversion) includes emissions from electricity generation and fuel use. Emissions from transport correspond to the use of trucks for carrying sugarcane (Fig. 1). CO2 emissions from biomass burning were considered to be neutral since photosynthesis during plant growth involves carbon fixation from the atmosphere (Khatiwada and Silveira, 2011). For biomass combustion, only CH4 and N2O emissions were considered. In all the cases studied, the agricultural fields had been cleared prior to 1960, so land use change emissions were not included in

2634

C.A. García et al. / Journal of Cleaner Production 112 (2016) 2632e2641

Table 1 Location, total industrialized area and mean sugar production of four case studies used to calculate the carbon footprint of sugar production in Mexico. Sugar mill

Ubication

Central Motzorongo La Gloria Tamazula Emiliano Zapata

18 19 19 18

280 250 410 390

3100 4200 0500 1300

N, N, N, N,

96 430 4300 O 96 240 0100 O 103 140 4300 O 99 110 2300 O

Tezonapa, Veracruz Úrsulo Galv an, Veracruz Tamazula, Jalisco Zacatepec, Morelos

Emissions Emissions from cane trash from inputs burning

Mean industrialized area (ha)

Mean production (ton sugar/y)

17,990 14,860 12,104 10,260

111,406 159,780 154,495 159,292

Chemicals Electricity from Fuel and lubricants the grid oil

Transportation Stage INPUTS: Fertilizers Pesticides Diesel Energy for Irrigation

Sugar Agricultural Stage

Industrial Stage Molasses

Fuel

Emisions from fuel

Bagasse burning emissions

Electricity Fuel oil emissions emissions

Fig. 1. System boundaries, inputs, outputs and sources of greenhouse gas emissions used to calculate carbon footprint values.

the analysis (CNIAA, 2013). This approach is consistent with other studies (Plassmann et al., 2010; Yuttitham et al., 2011). The functional unit is defined as 1 kg of sugar at the sugar mill gate. We considered just the main greenhouse gases, CO2, CH4 and N2O (with global warming potential of 1, 23 and 296 respectively), considering a time horizon of 100 years (European Commission, 2011). Other GHGs were excluded from the calculation since they contribute with only a small portion of global emissions (about 2%) (IPCC, 2014b). 2.3. Data Our study primarily encompasses agricultural practices in sugarcane farmland as presently adopted by cane farmers in the cases listed in Table 1, as well as the prevailing industrial operations in the respective sugar mills. The following ten year data were obtained from the National ~ eros” (UNC for its acronym in Spanish): Harvested Union of “Can area (ha), sugarcane productivity (t/ha), sugar yield (t/ha), cane mechanically harvested (%), cane mechanically loaded (%), burned sugarcane for harvest (%), sugar mill yield (%), raw sugar production (t), molasses production (t), external electricity use (kWh/ t cane) and petroleum consumed in sugar mills (L/t cane). Data on agrochemical application and fuel consumption for agricultural and transportation machinery are based on interviews and questionnaires with sugarcane growers and sugar mill technicians conducted during field visits, as well as personal communications with experts. Emissions from inputs, cane burning and the electric power grid were taken from the literature and from the BioGrace model (European Commission, 2011) (see Tables 2 and 3 for details). 2.4. Uncertainty and sensitivity analyses The results of a LCA study can be affected by different sources of uncertainty such as the quality of the available data, methodological choices, initial assumptions made on the allocation rules and system boundaries definition (Cellura et al., 2011). In this study, the uncertainty assessment was performed using Monte Carlo simu€rklund, 2002). This lation, which is a popular approach in LCA (Bjo method is a computational-based technique that relies on the

repetition of several individual model iterations, where each iteration is randomly constructed using a set of values selected from the probability distribution of each parameter. The uncertainty assessment comprised five stages. The first stage is the procedure of gathering all available information from the sugarcane production process, while the second stage focuses on the selection of the most relevant variables or factors for the sugarcane production life cycle assessment (see Section 2.3). The third stage is the characterization of the corresponding probability distributions for the selected variables. In this study, this characterization is based on information found in (Seabra et al., 2011) (see Table 4). Finally, the fourth and fifth stages are the Monte Carlo simulation and the corresponding analysis of its results. The number of simulations chosen for the Monte Carlo procedure was set to one million. The Monte Carlo analysis described above is complemented with an assessment of the contribution of each parameter towards the overall uncertainty of the sugarcane production process. For this purpose, we performed a single-parameter sensitivity analysis which is the usual procedure within the LCA community, used commonly to understand the uncertainty propagation in a given process (Steen, 1997). We varied each of the parameters from Table 4 separately, using their associated distribution and keeping the remaining parameters constant. We then obtained the corresponding total GHG emissions and estimated their minimum and maximum values. For further detail in uncertainty and sensitivity analysis method we refer the reader to the Supplementary Material. 2.5. Cogeneration scenarios Sugar mills may become net producers of electricity through the combustion of sugarcane bagasse (Brizmohun et al., 2015;  n et al., 2014). These surpluses, or net Ramjeawon, 2008; Rinco electricity production (NEP), can range from 60 kWh/t cane to 140 kWh/t cane in boilers with high pressures and temperatures (Khatiwada et al., 2012; Seabra et al., 2011). Cogeneration could be important for the economic sustainability of the sugarcane industry by providing additional income and reducing its exposure to volatile global sugar prices (Renouf et al., 2013). Two scenarios were developed in order to explore the effect that installing efficient

C.A. García et al. / Journal of Cleaner Production 112 (2016) 2632e2641

2635

Table 2 Parameters of the agricultural, transport and industrial stages used to calculate carbon footprint values in our four case studies. Sugar mill and supply zones Agricultural stage (sugarcane production) Harvested area (ha) Sugarcane productivity (ton/ha) Bagasse in cane (%) Pumping irrigation (% of harvested area) Energy for irrigation kWh/ha Trash burning (% of harvested area) Dry based matter in trash (%) Diesel consumption (L/ha) N fertilization rate (kg/ha) % of N to N2O P2O5 fertilization rate (kg/ha) K2O fertilization rate (kg/ha) CaCO fertilization rate (kg/ha) Pesticides application rate (kg active ingredient/ha) Transportation stage Diesel consumption (L/ton) Industrial stage Sugarcane crushed (ton/year) Sugar recovery (kg sugar/ha) Molasses production (ton) Fuel oil consumption (L/ton cane) Electricity from the grid (kWh) Lubricants (kg/ha)c Lime (kg/ha)d a b c d

Central Motzorongo

La Gloria

Tamazula

Emiliano Zapata

Source

17,990 64.4 26.8 5 219 95 14a 131b 105 1 51 153 1000 2.0

14,860 89.1 27.5 18 1185 91

12,104 106.4 31.3 25 1591 51

10,260 106.5 30.2 10 307 89

149 116 1 61 94 1000 2.5

154 210 1 70 140 1000 2.0

143 180 1 45 30 0 3.5

(UNC, 2014) (UNC, 2014) (UNC, 2014) Questionnaires Questionnaires (UNC, 2014) (Seabra et al., 2011) (García et al., 2011), Questionnaires Questionnaires (IPCC, 2006) (CONADESUCA, n.d.; García et al., 2011) (CONADESUCA, n.d.; García et al., 2011) Questionnaires, personal communication Questionnaires

2.2

1.9

2.9

2.2

Questionnaires

1,001,650 7181 42,230 4.5 221,083 0.66 57

1,324,335 10,489 45,316 0 1,529,395 0.91 78

1,218,769 13,068 40,551 2.9 3,030,910 1.1 94

1,200,334 13,062 40,127 2.0 2,628,077 1.1 94

(UNC, 2014) (UNC, 2014) (UNC, 2014) (UNC, 2014) (UNC, 2014) Calculated Calculated

Includes only CH4 and N2O emissions. We consider an ethanol density of 0.984 kg/L. Calculation based on 10.3 g/ton cane (Seabra et al., 2011). Calculation based on 880 g/ton cane (Seabra et al., 2011).

Table 3 Emission factors from the agricultural, transport and industrial stages of sugarcane production. These were used to calculate carbon footprint values. Sugar mill and supply zones

Central Motzorongo

Agricultural stage (sugarcane production) Emissions from electrical grid (kg CO2e/kWh)

La Gloria

Tamazula

Emiliano Zapata

Source (García et al., 2011)

0.489 Cane trash burning emission factor (kg CO2e/kg dry matter)

(Macedo et al., 2008) 0.083

Diesel emission factor (gCO2e/MJ)

(European Comission, 2011) 87.64a

N fertilizer emission factor (kg CO2e/kg)

(European Comission, 2011) 5.9

P2O5 fertilizer emission factor (gCO2e/kg)

(European Comission, 2011) 1010.7

K2O fertilizer emission factor (gCO2e/kg)

(European Comission, 2011) 576

CaCO fertilizer emission factor (gCO2e/kg)

(European Comission, 2011) 129.5

Pesticides emission factor (gCO2e/kg)

(European Comission, 2011) 10,971

Industrial stage Fuel oil emission factor (gCO2e/MJ)

(European Comission, 2011) 84.98b

Bagasse emission factor (kg CO2e/kg)

(Khatiwada and Silveira, 2011) 0.025

Lubricants emission factor (gCO2e/kg)

(European Comission, 2011) 0.947

Lime emission factor (kg CO2e/kg)

(European Comission, 2011) 1030.2

a b

We use a LHV of 35.9 MJ/L. Considering a fuel oil LHV of 39.3 MJ/L.

boilers and selling surplus electricity could have on the carbon footprint of the cases studied. The first scenario considers sales of NEP to the national grid of 60 kWh/t cane, and the second one considers a NEP of 117 kWh/t cane (based on Guerra et al., 2014). The resulting CFs of these scenarios for our case studies were

calculated applying allocation of emissions per energy content and economic value to the sugar, molasses and NEP. Table 5 shows total production of sugar, molasses and electricity in our two cogeneration scenarios. Lower heating values and economic costs are also shown.

2636

C.A. García et al. / Journal of Cleaner Production 112 (2016) 2632e2641

Table 4 Probability distributions of input parameters used in the Monte Carlo simulation for calculating the carbon footprint of our four case studies. Parameter

Distribution Values Central Motzorongo Mean/min

Sugarcane yield (t/ha) Bagasse in cane Energy for irrigation Trash burning (% of area) Diesel consumption (L/ha) N application (kg/ha) N2O emission factor (%) P2O5 application (kg/ha) K2O application (kg/ha) CaCO application (kg/ha) Pesticides application (kg/ha) Diesel for sugarcane transportation (L/t) Sugar production (%/t cane) Fuel oil consumption (L/t) Electricity from the grid (kWh)

La Gloria

Standar Mean/min deviation/max

Normal Normal None Triangular

64.4 26.8 219.5 0.90

Triangular

131

260

Triangular Triangular Triangular Triangular None Normal

84 0.4 45 30 1000 2.5

210 4 70 156

Normal

1.00

2.175

Normal

11.2

None Normal

7.4 1.1

0.707 0.727

1.900

126,610

6.9 0.6 0.96

131

Standar Mean/min deviation/max 5.4 0.8

210 4 70 156 0.707

84 0.4 45 30 1000 2.500

0.361

2.950

0

2.9 1,355,456

260

131

210 4 70 156

12.3

Standar deviation/max

106.5 30.2 307.3 0.85

0.54

131

0.2

1,529,395

Emiliano Zapata

106.4 31.3 1591.2 0.48

260

11.8

4.4864 221,083

Standar Mean/min deviation/max

89.1 27.5 1185.2 0.86

84 0.4 45 30 1000 2.500

0.5

Tamazula

0.707

84 0.4 45 30 0 2.500

0.354

2.200

0.2

4.7 2.2 0.93 260 210 4 70 156 0.707 1.414

12.2

0.5

2.017 3,030,911

664,494

2,628,078

1,173,047

Table 5 Total production of sugar, molasses and electricity in our two cogeneration scenarios (net electricity production of 60 kwh/t cane and 117 kWh/t cane). Lower heating values and economic costs are also shown, as they were used to allocate emissions values. Central Motzorongo

La Gloria

Tamazula

Emiliano Zapata

129,823 42,230

156,154 45,316

158,140 40,551

132,853 40,126

70 162

79 185

77 180

66 153

Average production Sugar (ton) Molasses (ton) Electricity 60 kWh/ton cane Annual surplus (GWh/y) 117 kWh/ton cane Annual surplus (GWh/y)

3. Results In this section we present the results of the carbon footprint (CF) evaluation for the case studies described and analyze the main sources of greenhouse gas (GHG) emission. We also present the results of the uncertainty and sensibility analyses, where we identified the parameters that have the greatest influence on CF values. Finally, we discuss the results of the cogeneration scenarios and how these may influence the CF.

LHV (GJ/t)

Economic value (USD/t)

16.7 7.2

657 172

e e

101 101

which represents nearly 30% of total emissions in three cases. The industrial stage for La Gloria sugar mill had the lowest contribution of carbon emissions (14%). The CF value of the agricultural stage for each sugar mill is shown in Fig. 3. This includes: Cane trash burning for manual harvest, diesel consumption for machinery, irrigation, fertilizer

Sugarcane producƟon

Sugarcane transportaƟon

Sugar mill

0.7

3.1. Carbon footprint 0.6

0.5

kgCO2e/kg sugar

Fig. 2 shows how the three stages of the sugarcane production process, agricultural, transportation and industrial, contribute towards total carbon emissions (kg CO2e). Differences are observed when we compare the CF from the four different mills: La Gloria mill reported the lowest CF (0.45 kg CO2e/kg sugar) and Motzorongo mill presented the highest (0.63 kg CO2e/kg sugar). Amongst these production stages, it is the agricultural stage that presents the highest contribution to the CF in all cases (in a range of 0.29e0.36 kg CO2e/kg sugar), a bulk that translates into 59e74% of total CO2e emissions for the sugarcane production process (Fig. 2). Sugarcane transportation from field to sugar mill reflected the lowest contribution of carbon (0.05e0.08 kg CO2e/kg sugar) reaching 13% of overall CO2e emissions. Finally, the industrial stage (sugar mill) had emissions between 0.06 and 0.18 kg CO2e/kg sugar,

0.4

0.3

0.2

0.1

0

Motzorongo

La Gloria

Tamazula

Emiliano Zapata

Fig. 2. Contribution of each stage towards the carbon footprint (kg CO2e/kg sugar) of sugarcane production in our four case studies.

C.A. García et al. / Journal of Cleaner Production 112 (2016) 2632e2641

Cane trash burning

Diesel consumpƟon

IrrigaƟon

N ApplicaƟon

N2O emissions

P2O5 ApplicaƟon

K2O ApplicaƟon

CaO ApplicaƟon

PesƟcides

Sugarcane transportaƟon

2637

0.45

0.4

kgCO2e/kg sugar

0.35

0.3

0.25

0.2

0.15

0.1

0.05

0

Motzorongo

La Gloria

Tamazula

Emiliano Zapata

Fig. 3. Carbon footprint (kg CO2e/kg sugar) from the agricultural stage of sugar production in our four case studies.

application and N2O emissions from nitrogenous fertilizers. When transportation is not considered, cane trash burning represents the highest portion of CO2 emissions, nearly a third of total carbon. Although this tendency is explicit for Motzorongo, La Gloria and Emiliano Zapata mills, Tamazula mill showed that only 14% of its GHG emissions came from cane trash burning due to its lower manually harvested area. When considering fertilizer application (P2O5, K2O, N, CaO) and N2O emissions, we find that these factors contribute with 40e50% of GHG emissions from the agricultural stage. Pesticides application contributes only marginally to the CF, with 1% of total CO2e emissions. In short, the mayor contributors towards the carbon footprint of the agricultural stage are cane trash burning, fertilizer application, N2O emissions, fossil fuel use and energy for irrigation; these findings are shown in Fig. 3. Fig. 4 shows how the industrial stage contributes towards the CF of the sugarcane mills. This stage includes fuel oil consumption,

Fig. 4. Carbon footprint (in kg CO2e/kg sugar) from the industrial stage of sugar production in our four case studies.

electricity from the grid, bagasse burning, chemicals and lubricants used during the sugarcane production process. Fuel oil consumption is a central parameter for Motzorongo, Tamazula and Emiliano Zapata mills: It represents 41e66% of total GHGs emissions from the industrial stage. La Gloria mill reports no fuel oil consumption. Use of electricity from the grid and application of chemicals and lubricants contributed with 0.01e0.06 kg CO2e/kg sugar for overall mills. In our case studies, most of the electricity is obtained from a cogeneration process where bagasse is used to generate energy, and only about 5% of total electricity consumption comes from de grid. The four mills show the same contribution in emissions from bagasse burning (0.06 kg CO2e/kg sugar), 10e13% of the total CF. 3.2. Uncertainty and sensitivity analyses The main results from the Monte Carlo simulation are shown in Fig. 5 as normalized histograms representing the aggregated emissions for the agricultural and industrial stages, as well as for the total amount of emissions from the whole process. The first observation from these figures is that, for the four sugar mills, all the analyzed emissions have a lognormal distribution. An additional observation is the fact that, for the industrial stage and total GHG emissions, Motzorongo mill reports distributions with the largest variation, and consequently its emissions have the largest uncertainty. Regarding the agricultural stage, it can be seen from Fig. 5 that although the ranges of the resulting emissions are different for the four mills, the width of the distributions is very similar. This means that in the four sugar mills evaluated the level of uncertainty associated with the agricultural stage is comparable. Fig. 6 shows the sensitivity analysis performed with regards to the variation in total GHG emissions due to changes in each of the corresponding selected variables. While the same scale is used on the x-axis, the order of the parameters in the y-axis is different for the four sugar mills assessed. Motzorongo is the mill with the largest sensitivity in CF results to changes in sugar production, sugarcane yield, N2O emission factor and amount of nitrogen (N)

2638

C.A. García et al. / Journal of Cleaner Production 112 (2016) 2632e2641

0.15

0.35

Industrial Stage

0.3

Agricultural Stage

0.25 Probability

Probability

0.1

0.05

0.2 0.15 0.1 0.05

0 0.2

0.4

0.6 0.8 kgCO2e/kgsugar

0 0.05

1

0.1

0.15 0.2 kgCO2e/kgsugar

0.25

0.1 0.08

Total

0.1

0.06

Probability

Probability

Transport

0.04

0.05

0.02 0

0

0.02

0.04 0.06 0.08 kgCO2e/kgsugar Motzorongo

0.1

0.12

La Gloria

0

0.4

Tamazula

0.6

0.8 1 kgCO2e/kgsugar

1.2

Emiliano Zapata

Fig. 5. Monte Carlo simulation results (kg CO2e/kg sugar) for carbon emissions from sugar production in our four case studies.

Fig. 6. Results of the Monte Carlo sensitivity analysis (in kg CO2e/kg sugar), reflecting sensitivity to variation in parameter values for our four case studies.

application. For all four mills, the parameters to which the final emissions are more sensitive are all within the agricultural stage, which means that the carbon footprint of sugar production is mostly influenced by the parameters of the agricultural stage (mainly sugarcane yield, N2O emissions and N application).

3.3. Cogeneration scenarios Results of GHG emissions allocation to sugarcane by-products are presented in Table 6. It is important to note that the allocation of emissions to molasses and electricity reduces the carbon

C.A. García et al. / Journal of Cleaner Production 112 (2016) 2632e2641

2639

Table 6 Greenhouse gas emissions allocated to sugar, molasses and electricity production in our two cogeneration scenarios.

60 kWh/ton cane export Sugar kg CO2e/kg/(% allocation) Molasses kg CO2e/kg/(% allocation) Electricity kg CO2e/kWh/(% allocation) 117 kWh/ton cane export Sugar kg CO2e/kg/(% allocation) Molasses kg CO2e/kg/(% allocation) Electricity kg CO2e/kWh/(% allocation)

Central Motzorongo

La Gloria

Energy content

Economic value

Energy content

Economic value

Energy content

Economic value

Energy content

Economic value

0.50/(80%) 0.17/(11%) 0.108/(9%)

0.54/(86%) 0.14/(7%) 0.082/(7%)

0.36/(81%) 0.17/(10%) 0.082/9%)

0.39/(87%) 0.11/(7%) 0.063(6%)

0.46/(8%) 0.25/(9%) 0.099/(9%)

0.49/(88%) 0.16/(6%) 0.083/(6%)

0.38/(81%) 0.18/(10%) 0.089/(9%)

0.42/(87%) 0.12/(7%) 0.068/(6%)

0.46/(73%) 0.20/(10%) 0.196/(17%)

0.50/(80%) 0.13/(7%) 0.151/(13%)

0.33/(75%) 0.16/(9%) 0.147/(16%)

0.36/(81%) 0.11/(6%) 0.115(13%)

0.42/(76%) 0.23/(8%) 0.177/(16%)

0.46/(82%) 0.15/(6%) 0.152/(12%)

0.35/(75%) 0.16/(10%) 0.159/(15%)

0.38/(82%) 0.11/(6%) 0.124/(12%)

footprint of sugar, resulting in values between 0.36 and 0.54 kg CO2e/kg sugar with a net electricity production (NEP) of 60 kWh/t cane (with the same amount of bagasse). When we consider a NEP value of 117 kWh/t cane, the CF ranges between 0.33 and 0.46 kg CO2e/kg sugar. The method of allocation of emissions by economic value produces slightly higher values of CF for sugar (nearly 8% higher) compared to allocation by energy content. According to our method, electricity generation from our NEP scenarios would have emissions in a range of 0.082e0.196 kg CO2e/ kWh for our four cases (see Table 6). 4. Discussion The results present carbon footprint values for sugarcane production within a range of 0.45e0.63 kg CO2e/kg sugar. Of the four cases studied, Motzorongo mill generated the most carbon emissions and La Gloria mill the fewest. In these four cases, the agricultural stage contributes with the greatest portion of the CF values (59e74% of total emissions), followed by the industrial stage (16e32%) and finally the transportation stage (10e13%). The CF values of sugar for our case studies are relatively higher than what has previously been reported for sugar produced in other countries. CF values reported for sugar production in Brazil are between 0.23 and 0.24 kg CO2e/kg sugar (De Figueiredo et al., 2010; Seabra et al., 2011); for Mauritius, the CF reaches 0.255 kg CO2e/kg sugar (Plassmann et al., 2010) and in Thailand is the CF is within the range of our Mexican case studies, 0.55 kg CO2e/kg sugar (Yuttitham et al., 2011). For these international CF values, as well as our case studies, the greatest emissions output occurs during the agricultural stage. In our four cases, 53e68% of agricultural stage emissions were due to the combination of nitrogenous fertilizer use, N2O emissions and cane burning for manual harvest. These processes represent in turn 38e49% of the total carbon footprint for the sugarcane production process. Yuttitham et al. (2011), De Figueiredo et al. (2010) and Seabra et al. (2011) also reported that the main contributor to the CF in their analyses was the agricultural stage, including biomass burning, fertilizer application, field emissions (including N2O) and fossil fuel use. It is interesting to note that for cases reported in Brazil, nitrogenous fertilizers are applied in lower quantities compared to what is used in Mexico in all cases (0.777 kg N/t cane in Brazil versus 1.3e2.0 kg N/t cane in Mexico; Seabra et al., 2011), and cane yield is higher in those Brazilian cases than for Motzorongo mill and slightly lower than La Gloria mill. This would indicate that there is an opportunity to reduce GHG emissions from nitrogenous fertilizers. Similar opportunities must be identified and exploited, such as maintaining the C/N ratio at an adequate level, considering direct application of press mud cake (or previously composted), incorporating vermicomposting, manures and green manuring as trash

Tamazula

Emiliano Zapata

(tops and leaves) into the soils. There are other possibilities such as establishing legume intercropping as a source of organic nutrients, increasing diversity of crop rotations, use of biofertilizers and zeolites, among others (Prado et al., 2013; Yang et al., 2013). Emissions from cane burning for manual harvest contribute with 8e20% of total CF values for our case studies. This activity generates a considerable amount of emissions in Brazil (0.048 kg CO2/kg sugar), but these are still significantly lower than three of our Mexican case studies where CF values are as high as 0.10 kg CO2/kg sugar. One of the reasons behind this difference is that some of Brazil's emissions from burning are assigned to the ethanol production process, and not to sugarcane production. An additional reason is that in Brazil, cane is burned for manual harvest in 65% of the total harvest area, whereas in three of the Mexican case studies, 89e95% of the area is used for burning. Tamazula mill is the exception where cane is burned only in 51% of the total harvest area, and therefore it presents the lowest CF (0.05 kg CO2/kg sugar) for this agricultural activity. Sugar cane burning could be substituted by a method known as green harvest, where all or part of the straw is left on the soil. According to some studies, green harvest can potentially reduce GHG emissions (Capaz et al., 2013; De Figueiredo and La Scala Jr., 2011; De Oliveira Bordonal et al., 2012). This harvest method has other advantages, such as preventing liberation of gasses and polluting particles, with a consequential improvement of nearby populations' health conditions (Galdos et al., 2013); decreasing erosion; increasing land fertility; reducing soil evaporation, which in turn improves water use efficiency, among others (Leal et al., 2013). Other sources of emissions in the agricultural stage are fuel use in agricultural activities and energy use for irrigation. In order to diminish emissions from diesel use in agricultural work and sugarcane transportation, it would be important to explore no-till agricultural as an alternative (De Oliveira Bordonal et al., 2012). Implementing drip irrigation, an efficient water use alternative, would be a highly important strategy to mitigate emissions from energy used for irrigation. Estimated emissions for sugarcane transportation from crop field to sugar mill are less relevant and very similar in the four mills studied (10e13% of their total GHG emissions). In order to reduce these GHG emissions, some useful suggestions include improving the roads, fostering the use of trains for sugarcane transportation, as well as refurbishing trucks, loaders and harvesters (Tieppo et al., 2014; Mele et al., 2011). The industrial stage of sugarcane production contributes in a lesser degree to total emissions. Sugar milling was found to contribute with nearly 30% of CO2 emissions from the industrial stage. The only exception was La Gloria mill, where this process has been improved by increasing thermal and energy efficiency, with a resulting elimination of fuel oil use. On the other hand, Motzorongo

2640

C.A. García et al. / Journal of Cleaner Production 112 (2016) 2632e2641

mill showed the highest emissions from sugarcane processing (31% of its total GHG emissions), mainly due to the high use of fuel oil as supplementary fuel for heat and electricity generation. In the case of Thailand's sugar industry (Yuttitham et al., 2011), the carbon footprint of its industrial stage (0.06 kg CO2/kg sugar) is significantly lower than the corresponding CF for three of our Mexican case studies (0.07e0.20 kg CO2/kg sugar). This situation can be explained by the lack of supplementary fossil fuel use for heat and electricity during the industrial process in Thailand. This can be confirmed by the fact that, as mentioned above, La Gloria mill does not consume any additional fossil fuels during the industrial stage, hence its CF value for this stage (0.07 kg CO2/kg sugar) is very similar to what has been found in Thailand. This effectively shows that taking action to eliminate fossil fuel use in sugar mills will contribute to minimize their carbon footprint. The agricultural stage is not only associated with the greatest amount of GHG emissions, but it is also the stage with the largest uncertainty in the estimation of emissions. This uncertainty is mainly explained by the application of nitrogen fertilizers to the crop. As was noted in Section 3.2, CF values are especially sensitive to variations regarding sugarcane yield, N2O emissions and fertilization with nitrogen. These results confirm that actions taken to decrease nitrogenous fertilizer amounts without affecting sugarcane yield are highly important. Carbon footprint values for sugar production could be minimized by implementing efficient cogeneration in sugar mills through installation of high-temperature and pressure boilers which use bagasse as fuel, potentially reaching CF values of 0.33e0.54 kg CO2e/kg sugar. This could also represent an opportunity for low-carbon electricity generation with emissions that range between 0.082 and 0.196 kg CO2e/kWh, which is significantly cleaner than what is generated with fossil fuels in Mexico (1.090 kg CO2e/kWh for coal, 0.959 kg CO2e/kWh for fuel oil and 468 kg CO2e/kWh for natural gas; estimations done with data from Santoyo-Castelazo et al., 2011). Efficient cogeneration has the potential to minimize the carbon footprint of Mexican sugar mills depending on their net electricity production (NEP). This is a promising option that must be studied in greater detail.

combustion in the boilers. Finally, the transportation stage only represents between 10 and 13% of total emissions. Few systematic scientific efforts have been made to analyze the potential role of the Mexican sugar agroindustry in quantifying GHG emissions in sugar production. This is relevant since Mexico's ambitious goal of a 30% reduction of GHG emissions by year 2020 was established. The main opportunities for GHG emission mitigation in the sugar agroindustry can be found in the agricultural stage, specifically with regards to minimizing nitrogenous fertilizer use and limiting cane burning for manual harvest. Within the industrial stage it is important to eliminate fossil fuel use and promote efficient cogeneration. Studying such techniques for improving sugar cane yield, making fertilizer use more efficient, minimizing cane burning and developing efficient cogeneration in sugar mills with bagasse as fuel is scientifically relevant. Applying concrete public policy measures to these areas of opportunity would allow production of low carbon sugar in Mexico. The results of this study may also be used as reference by other countries with similar sugar production conditions. Acknowledgments Carlos García wishes to thank the Direccion General de Asuntos mico of Universidad Nacional Auto noma de del Personal Acade xico for the financial support via a Postdoctoral Fellowship Me Program (No. 505002358). The authors also want to thank Andrea Alatorre and Alfredo rrez for the edition, English review and conceptual Fuentes Gutie contributions to the manuscript. Thank you as well to the three anonymous reviewers and the associate editor for their valuable comments. Appendix A. Supplementary material Supplementary material related to this article can be found at http://dx.doi.org/10.1016/j.jclepro.2015.09.113. References

5. Conclusions In this article we carried out an assessment of the carbon footprint (CF) of sugar production for four cases in Mexico using life cycle assessment (LCA) method. We also made uncertainty and sensitivity analyses for the main parameters in sugar production, which is relevant since uncertainty and variability have been discussed as important aspects to consider within the field of life cycle assessment (LCA) methodology. We also explored two cogeneration scenarios for net energy production (NEP) and its effect on the CF. The results show that CF values vary within a range of 0.45e0.63 kg CO2e/kg sugar, which are relatively higher than other reported cases from other countries, which range from 0.234 in Brazil to 0.55 in Thailand. The agricultural stage contributes with the largest portion of the CF of sugar production in all four cases. The main sources of GHG emissions in the agricultural stage were from the manufacturing and application of fertilizers, fossil fuel use, energy use for irrigation and biomass burning. This stage also presents the greatest uncertainty in results according to the Monte Carlo simulation, mainly due to nitrogenous fertilizer manufacturing and N2O emissions. The sensibility analysis shows that sugarcane yield on the field is also important. The industrial stage is second in importance regarding emissions: 14e30% of total GHG emissions occur in this stage. This is due to the use of complementary fuel oil for heat and electricity generation, followed by emissions of CH4 and N2O from bagasse

€rklund, A.E., 2002. Survey of approaches to improve reliability in LCA. Int. J. Life Bjo Cycle Assess. 7, 64e72. Brizmohun, R., Ramjeawon, T., Azapagic, A., 2015. Life cycle assessment of electricity generation in Mauritius. J. Clean. Prod. 16, 1727e1734. Campbell, J.E., Lobell, D.B., Field, C.B., 2009. Greater transportation energy and GHG offsets from bioelectricity than ethanol. Science 324, 1055e1057. Capaz, R.S., Carvalho, V.S.B., Nogueira, L.A.H., 2013. Impact of mechanization and previous burning reduction on GHG emissions of sugarcane harvesting operations in Brazil. Appl. Energy 102, 220e228. Special Issue on Advances in sustainable biofuel production and use e XIX International ́ Symposium on Alcohol Fuels e ISAF. Cellura, M., Longo, S., Mistretta, M., 2011. Sensitivity analysis to quantify uncertainty in life cycle assessment: the case study of an Italian tile. Renew. Sustain. Energy Rev. 15, 4697e4705. Cherubini, F., Bird, N.D., Cowie, A., Jungmeier, G., Schlamadinger, B., WoessGallasch, S., 2009. Energy- and greenhouse gas-based LCA of biofuel and bioenergy systems: key issues, ranges and recommendations. Resour. Conserv. Recycl. 53, 434e447. Cherubini, F., Strømman, A.H., 2011. Life cycle assessment of bioenergy systems: state of the art and future challenges. Bioresour. Technol. 102, 437e451. CNIAA, 2013. Sugar Industry Mexican Handbook. Edited by CNIAA. 495 p. n de Costos de Produccio n de Can ~ a de Azucar. CONADESUCA. Sistema de Informacio De Figueiredo, E.B., La Scala Jr., N., 2011. Greenhouse gas balance due to the conversion of sugarcane areas from burned to green harvest in Brazil. Agric. Ecosyst. Environ. 141, 77e85. ~o, R., La Scala Jr., N., 2010. Research greenDe Figueiredo, E.B., Panosso, A.R., Roma house gas emission associated with sugar production in southern Brazil. Carbon Balance Manag. 5, 3e7. De Oliveira Bordonal, R., de Figueiredo, E.B., La Scala, N., 2012. Greenhouse gas balance due to the conversion of sugarcane areas from burned to green harvest, considering other conservationist management practices. GCB Bioenergy 4, 846e858. n, 2012. Ley General de Cambio Clima tico. Diario Oficial de la Federacio

C.A. García et al. / Journal of Cleaner Production 112 (2016) 2632e2641 Dunkelberg, E., Finkbeiner, M., Hirschl, B., 2014. Sugarcane ethanol production in Malawi: measures to optimize the carbon footprint and to avoid indirect emissions. Biomass Bioenergy 71, 37e45. European Commission, 2011. Biograce GHG Calculation Tool 4b. Intelligence Energy Europe. http://www.biograce.net/. Fang, K., Heijungs, R., de Snoo, G.R., 2014. Theoretical exploration for the combination of the ecological, energy, carbon, and water footprints: overview of a footprint family. Ecol. Indic. 36, 508e518. Galdos, M., Cavalett, O., Seabra, J.E.A., Nogueira, L.A.H., Bonomi, A., 2013. Trends in global warming and human health impacts related to Brazilian sugarcane ethanol production considering black carbon emissions. Appl. Energy 104, 576e582. García, C.A., Fuentes, A., Hennecke, A., Riegelhaupt, E., Manzini, F., Masera, O., 2011. Life-cycle and greenhouse gas emissions and energy balances of sugarcane ethanol production in Mexico. Applied Energy 88 (6), 2088e2097. Guerra, J.P.M., Coleta Jr., J.R., Arruda, L.C.M., Silva, G.A., Kulay, L., 2014. Comparative analysis of electricity cogeneration scenarios in sugarcane production by LCA. Int. J. Life Cycle Assess. 19, 814e825. IPCC, 2014a. Climate Change 2014 Impacts, Adaptation and Vulnerability Part A: Global and Sectorial Aspects. IPCC, 2014b. Climate Change 2014 Synthesis Report. IPCC, 2013. Climate Change 2013 the Physical Science Basis. IPCC, 2006. Guidelines for National Greenhouse Gas Inventories. In: Agriculture, Forestry and Other Land Use, vol. 4. Islas, J., Manzini, F., Masera, O., 2007. A prospective study of bioenergy use in Mexico. Energy 32, 2306e2320. xico: Estudio Sobre la DisJohnson, T.M., Alatorre, C., Romo, Z., Liu, F., 2010. Me  n de Emisiones de Carbono (MEDEC). The World Bank, Washington, minucio DC. Khatiwada, D., Seabra, J., Silveira, S., Walter, A., 2012. Power generation from sugarcane biomass e a complementary option to hydroelectricity in Nepal and Brazil. Energy 48, 241e254. Khatiwada, D., Silveira, S., 2011. Greenhouse gas balances of molasses based ethanol in Nepal. J. Clean. Prod. 19, 1471e1485. Klenk, I., Landquist, B., de Imana, O.R., 2012. The product carbon footprint of EU beet sugar (part I). Sugar Ind. Zuckerind. 137 (3), 169e177. Leal, M.R.L.V., Galdos, M.V., Scarpare, F.V., Seabra, J.E.A., Walter, A., Oliveira, C.O.F., 2013. Sugarcane straw availability, quality, recovery and energy use: a literature review. In: Biomass Bioenergy, 20th European Biomass Conference, 53, pp. 11e19. Macedo, I.C., Seabra, J.E.A., Silva, J.E.A.R., 2008. Greenhouse gases emissions in the production and use of ethanol from sugarcane in Brazil: the 2005/2006 averages and a prediction for 2020. Biomass and Bioenergy 32 (7), 582e595. n-Gos nez, L., 2011. Multiobjective model Mele, F.D., Kostin, A.M., Guille albez, G., Jime for more sustainable fuel supply chains. A case study of the sugar cane industry in Argentina. Ind. Eng. Chem. Res. 50 (9), 4939e4958. Nguyen, T.L.T., Gheewala, S.H., Sagisaka, M., 2010. Greenhouse gas savings potential of sugar cane bio-energy systems. J. Clean. Prod. 18, 412e418. Octaviano, C., Paltsev, S., Gurgel, A.C., n.d. Climate change policy in Brazil and Mexico: results from the MIT EPPA model. Energy Econ. http://dx.doi.org/10. 1016/j.eneco.2015.04.007. Plassmann, K., Norton, A., Attarzadeh, N., Jensen, M.P., Brenton, P., EdwardsJones, G., 2010. Methodological complexities of product carbon footprinting: a

2641

sensitivity analysis of key variables in a developing country context. Environ. Sci. Policy 13, 393e404. Prado, R.D.M., Caione, G., Campos, C.N.S., 2013. Filter cake and vinasse as fertilizers contributing to conservation agriculture. Appl. Environ. Soil Sci. 2013, 1e8. Ramjeawon, T., 2008. Life cycle assessment of electricity generation from bagasse in Mauritius. J. Clean. Prod. 16, 1727e1734. Rein, P.W., 2010. The carbon footprint of sugar. Sugar IndistrySuckerind 13, 427e434. Renouf, M.A., Pagan, R.J., Wegener, M.K., 2013. Bio-production from Australian sugarcane: an environmental investigation of product diversification in an agro-industry. J. Clean. Prod. 39, 87e96. rrez Lllerela, J.P., García, C.A., 2012. Acciones apropiadas para la Riegelhapt, E., Gutie n de emisiones de gases de efecto invernadero y uso eficiente del agua mitigacio xico. Report for: Instituto Nacional de en la agroindustria azucarera de Me Ecología y Cambio Clim atico and Interamerican Development Bank. Mexico. n, L.E., Becerra, L.A., Moncada, J., Cardona, C.A., 2014. Techno-economic analRinco ysis of the use of fired cogeneration systems based on sugar cane bagasse in south eastern and mid-western regions of Mexico. Waste Biomass Valoriz. 5, 189e198. €m, J., Steffen, W., Noone, K., Persson, Å., Chapin, F.S., Lambin, E.F., Rockstro Lenton, T.M., Scheffer, M., Folke, C., Schellnhuber, H.J., Nykvist, B., de Wit, C.A., €rlin, S., Snyder, P.K., Costanza, R., Hughes, T., van der Leeuw, S., Rodhe, H., So Svedin, U., Falkenmark, M., Karlberg, L., Corell, R.W., Fabry, V.J., Hansen, J., Walker, B., Liverman, D., Richardson, K., Crutzen, P., Foley, J.A., 2009. A safe operating space for humanity. Nature 461, 472e475. Santoyo-Castelazo, E., Gujba, H., Azapagic, A., 2011. Life cycle assessment of electricity generation in Mexico. Energy 36, 1488e1499. Seabra, J.E.A., Macedo, I.C., Chum, H.L., Faroni, C.E., Sarto, C.A., 2011. Life cycle assessment of Brazilian sugarcane products: GHG emissions and energy use. Biofuels Bioprod. Biorefining 5, 519e532. mez-Merino, F.C., Valdez-Balero, A., Silva-Rojas, H.V., TrejoSentíes-Herrera, H.E., Go llez, L.I., 2014. The agro-industrial sugarcane system in Mexico: current staTe tus, challenges and opportunities. J. Agric. Sci. 6 (4), 26e54. Soam, S., Kumar, R., Gupta, R.P., Sharma, P.K., Tuli, D.K., Das, B., 2015. Life cycle assessment of fuel ethanol from sugarcane molasses in northern and western India and its impact on Indian biofuel programme. Energy 83, 307e315. Steen, B., 1997. On uncertainty and sensitivity of LCA-based priority setting. J. Clean. Prod. 5, 255e262. Tieppo, R.C., Andrea, M.C.S., Gimenez, L.M., Romanelli, T.L., 2014. Energy demand in sugarcane residue collection and transportation. Agric. Eng. Int. CIGR J. 52e58. UNC, 2014. Estadísticas azucareras de Mexico. Available from. : http://www.caneros. org.mx/estadisticas.html. Veysey, J., Octaviano, C., Calvin, K., Martinez, S.H., Kitous, A., McFarland, J., van der Zwaan, B., n.d. Pathways to Mexico's climate change Mitigation targets: a multimodel analysis. Energy Econ. http://dx.doi.org/10.1016/j.eneco.2015.04.011. Wiedmann, T., Minx, J., 2007. A Definition of “Carbon Footprint”. ISA UK Research Report 07e01. Cent. Integr. Sustain. Anal. Durh. UK. Yang, S.D., Liu, J.X., Wu, J., Tan, H.W., Li, Y.R., 2013. Effects of vinasse and press mud application on the biological properties of soils and productivity of sugarcane. Sugar Tech. 15 (2), 152e158. Yuttitham, M., Gheewala, S.H., Chidthaisong, A., 2011. Carbon footprint of sugar produced from sugarcane in eastern Thailand. J. Clean. Prod. 19, 2119e2127.