Advertisement

Environmental Sustainability

, Volume 2, Issue 4, pp 343–353 | Cite as

Sustainability assessment of crop production in accord with energy, environment and economic performances in Nepal

  • Anil Pokhrel
  • Peeyush SoniEmail author
Original Article
  • 427 Downloads

Abstract

Increasing crop production usually warrants increased use of energy in the system that triggers the question of locating an ‘appropriate’ operating point by weighing the economics of output and its environmental impacts. Sustainability of diverse cereals, legumes and vegetable production systems were assessed in lowlands of Nepal in light of the efficiency of energy use, CO2e (carbon dioxide equivalent) emissions and financial return–cost ratios. The energy saving potential of 238 farms selected in the study area is calculated adopting Data Envelopment Analysis (DEA). Based on the per-hectare energy consumption, vegetables (12.37 GJ), cereals (10.60 GJ) and legumes (3.80 GJ) crop production systems are ranked in descending order. Rice offered the highest per-hectare energy output (68.99 GJ) among the selected crops. Potential saving of energy inputs in the study area is shown to be between 18 and 35% without conceding yield of the particular crop production system. The highest environmental impact is due to garlic production (2997.13 kg CO2e ha−1); while rice, maize, wheat, lentil, mungbean and onion crops emit 60, 63, 28, 5, 3 and 86% CO2e emissions (of that of garlic crop), respectively. As expected, the vegetable production remains the most profitable system with the return–cost ratio of 2.96–3.52, followed by legume (1.83–2.16) and cereal (1.21–1.96) production systems.

Keywords

Energy input–output Greenhouse gas emissions Return–cost ratio Technical efficiency Crop production 

Introduction

Energy consumption and greenhouse gas (GHG) emissions are the key issues in todays’ agriculture, with increasing use of environmentally detrimental farm inputs. Agriculture is the major source of global GHG emissions that emits about 5.3 GtCO2e (gigatonne of carbon dioxide equivalent) in the year 2014, where more than 18% of global GHG emissions are coming from South Asia (FAOSTAT 2019). Similarly, the energy consumption by agriculture in South Asia is increasing to respond to feed the growing population with limited agricultural land (FAOSTAT 2019). Increased use of energy in agricultural production system not only yield production and earnings to the farms, but it also undesirably affects the environment by significantly contributing to the global warming (Arora 2018; Arora et al. 2018).

In South Asian countries like Nepal, the CO2e emissions from agriculture has increased by 16% in the year 2016 than that of year 2007, where the consumption of energy in agriculture was more than double in the year 2012 from year 2007 (FAOSTAT 2019). Due to a high population growth and inadequate cultivated land, todays’ agriculture is becoming an energy exhaustive sector with many of its activities attributing to the energy intensive farm machinery and inputs. With an expected population of more than 40 million by the year 2030, the country requires 47% more cereals production compared with the year 2010 in order to fulfill its national requirement (Prasad et al. 2011; Pokhrel and Soni 2017). But, the productivity of major cereals like rice (Oryza sativa), wheat (Triticum aestivum) and maize (Zea mays) are relatively lower (3.4, 2.6 and 2.6 t ha−1, respectively) as compared to average yield of South Asian countries (3.9, 3.0 and 3.4 t ha−1, respectively), which is the major challenge to meet the demand for feeding the growing population (FAOSTAT 2019). The government looks on to increasing the energy intensive farm inputs, i.e. fertilizers, farms machinery and mechanized irrigation systems to meet its domestic food demand and to improve the livelihood of the farm households, where the increased yield with intensified production systems could feed the growing population (MESD 2018).

However, the consumption of agricultural inputs in crop production systems is rising, still there is no corresponding proportional growth (in crop production) (MESD 2018). Fertilizer (N, P2O5 and K2O) consumption has increased by 3 times in the year 2016 compared with year 2010 (FAOSTAT 2019). Similarly, use of diesel fuel and electricity in agriculture is growing yearly by 11% and 8%, respectively (WECS 2010). Furthermore, the irrigated area has improved from 24% (year 2006) to 30% (year 2010) in relation of the total agricultural land (World Bank 2017). As energy consumption and GHG emissions in crop production systems are growing at increasing rate, it is therefore imperative to evaluate the interrelated aspects as energy, environmental and monetary performances of the major cereals, legumes and vegetable cultivation systems for their sustainability in food production systems. For this purpose, we selected rice, wheat, maize, lentil (Lens culinaris), mungbean (Vigna radiata), garlic (Allium sativum) and onion (Allium cepa) as the major crops, primarily grown under lowland ecological domain of the country.

Expansions of farm productivity and profitability with minimum use of farm inputs are the basic aims of sustainable crop production systems (CPSs). The energy analysis of different CPSs is used to gauge the environmental effects and its underlying sustainability issues. To estimate the technical efficiencies of farms, Data Envelopment Analysis (DEA) has been used for quantifying energy inputs saving potential in CPSs (Khoshnevisan et al. 2013; Pokhrel and Soni 2017).

Different CPSs result in variation of desirable outputs (yield and income) and undesirable output (environmental effect) from different agricultural production activities at farm level. Numerous energy and environmental studies have been conducted worldwide while studying CPSs such as in: rice (Soni et al. 2013; Pokhrel and Soni 2019), wheat (Khoshnevisan et al. 2013; Saad et al. 2016; Pokhrel and Soni 2019), maize (Soni et al. 2013; Saad et al. 2016; Elsoragaby et al. 2019; Juarez-Hernandez et al. 2019), lentil (Pokhrel and Soni 2018), mungbean (Soni et al. 2013), garlic (Samavatean et al. 2011; Pokhrel and Soni 2018; Elsoragaby et al. 2019) and onion production system (Zarini et al. 2014). But, none of such analyses is known to be reported on these aspects in the context of Nepal.

This research is aimed to examine energy use efficiency, environmental impacts and financial benefits of different crop production systems in the lowland ecological regions of Nepal through analyzing energy input–output characteristics, CO2e emissions and return–cost ratio.

Methodology

Study site

This study was conducted in Rapti Sonari Rural Municipality of Banke district located at 28.00° north latitude and 81.91° east longitude in Province No. 5, Nepal. Rice in rainy, garlic, onion, wheat and lentil in winter, and mungbean and maize in spring seasons are the major crops of the study area growing in lowland. The study area received 1642 mm rainfall during the study period from June to May in which 84% of total precipitation occurred during June to September. A monthly maximum (39.7 °C) and minimum (36.3 °C) temperatures were recorded in June (the hottest month), whereas January was the coolest month with an average maximum of 17.9 °C and 8.9 °C minimum temperatures. The soil pH in the study area was 6.3 and contained 1.53%, 0.07%, 43 kg ha−1 and 310 kg ha−1 organic matter, nitrogen, phosphorus and potassium, respectively.

Data collection

A household survey was performed to collect the primary data by adopting a structured questionnaire. Altogether, 238 sample farms were chosen from the stratified random sampling technique comprising 42 for rice, 37 for wheat, 39 for maize and 30 farms for each lentil, mungbean, onion and garlic crops.

Energy analysis

The farm inputs considered in this study were seeds, farmyard manure (FYM), chemical fertilizers, labor, draft animal, farm machinery and tools, and diesel fuel for the estimation of total energy consumption in CPSs. Farmers did not practice any chemicals for the management of diseases and pests in all the CPSs. Similarly, the economic output of crops was used for the calculation of energy output. The physical units of all the inputs and output in CPSs are expressed into energy units with their standard energy coefficients to accomplish the energy analysis (Table 1).
Table 1

Energy coefficients (MJ unit−1) of farm inputs and outputs

Farm Inputs and outputs

Energy equivalent

1. Labor: male (h), female (h)

1.96, 1.57

2. Animal-bullock: large, medium (Pair h−1)

14.05, 10.10

3. Diesel fuel (l)

56.31

4. Chemical fertilizer (kg)

 Nitrogen (N)

66.14

 Phosphorus (P2O5)

12.44

 Potassium (K2O)

11.15

5. Farmyard manure (kg)

0.30

6. Farm tools and machinery

 Spade (h)

0.314

 Sickle (h)

0.836

 Plough (h)

0.627

 Cart (h)

5.204

 Moldboard Plough (h)

2.508

 Cultivator (h)

3.135

 Trailer (h)

17.431

 Thresher (h)

7.524

 Diesel engine (h)

0.581

 Tractor: 45 hp and above (h)

16.416

7. Seed, output (kg)

 Rice

14.7

 Wheat

15.7, 14.7

 Maize

15.30, 14.7

 Lentil

14.7

 Mungbean

15.52

 Onion

1.90

 Garlic

1.6

All values are obtained from Pokhrel and Soni (2017, 2019)

The energy balance, energy ratio, specific energy, energy productivity and energy intensiveness were calculated by using Eqs. (1)–(5).
$$\begin{aligned} Energy\,balance\,\, \left( {{\text{GJ}}\,{\text{ha}}^{ - 1} } \right) &= Energy\,output\,\, \left( {{\text{GJ}}\,{\text{ha}}^{ - 1} } \right) \\& \quad - Energy\,input \,\,\left( {{\text{GJ}}\,\,{\text{ha}}^{ - 1} } \right) \end{aligned}$$
(1)
$$Energy\,ratio = \frac{{Energy\,output\, \left( {{\text{GJ}}\,{\text{ha}}^{ - 1} } \right)}}{{Energy\,input \,\left( {{\text{GJ}}\,{\text{ha}}^{ - 1} } \right)}}$$
(2)
$$Energy\,productivity\, \left( {{\text{kg}}\,{\text{GJ}}^{ - 1} } \right) = \frac{{Crop\,yield \,\left( {{\text{kg}}\,{\text{ha}}^{ - 1} } \right)}}{{Energy\,input \,\left( {{\text{GJ}}\,{\text{ha}}^{ - 1} } \right)}}$$
(3)
$$Specific\,energy\, \left( {{\text{MJ}}\,\,{\text{kg}}^{ - 1} } \right) = \frac{{Energy\,\,input\, \left( {{\text{MJ}}\,{\text{ha}}^{ - 1} } \right)}}{{Crop\,\,yield\, \left( {{\text{kg}}\,{\text{ha}}^{ - 1} } \right)}}$$
(4)
$$\begin{aligned} &Energy\,intensiveness \,\,\left( {{\text{MJ}}\,{\text{USD}}^{ - 1} } \right) \\&\quad = \frac{{Energy\,input\,\left( {{\text{MJ}}\,{\text{ha}}^{ - 1} } \right)}}{{Total \,cost\,of\,production\,\left( {{\text{USD}}\,{\text{ha}}^{ - 1} } \right)}} \end{aligned}$$
(5)

Efficiency estimation of the farms

A non-parametric DEA technique was adopted to measure the efficiency of different CPSs (Khoshnevisan et al. 2013; Pokhrel and Soni 2017). The agricultural production farms required a set of inputs like seeds, labor, fertilizers, plant protection chemicals, water, farm machinery and tools to produce the product as yield. The DEA estimated the farms’ efficiency level in comparison to all other farms within the same CPS by assuming that all farms lie on or below the efficient frontier. Farms, which lie on efficient frontier are accepted as efficient farms, while the farms which lie beneath the efficient frontier are measured as inefficient farms (Pokhrel and Soni 2017).

There are three approaches in DEA for the estimation of farm efficiencies: (1) input-oriented DEA that considered the same level of outputs by reducing the level of inputs, (2) output-oriented DEA that considered increasing the level of outputs with the same level of inputs, and (3) non-oriented DEA that considered the increasing level of outputs by reducing the level of inputs. An input-oriented DEA technique was used for this study as farmers can only control over the level of inputs used in the production process, but cannot solely control over the outputs due to uncontrollable externalities. The farm efficiencies were estimated by considering energies of inputs (GJ ha−1) as an input and crop yield (kg ha−1) as output parameters in MaxDEA pro (Khoshnevisan et al. 2013; Pokhrel and Soni 2017).

According to Chauhan et al. (2006), technical efficiency is the ratio of sum of the weighted outputs to sum of weighted inputs (Eq. 6).
$$\theta_{j } = \frac{{u_{1} y_{1j} + u_{2} y_{2j} + \cdots + u_{n} y_{nj} }}{{v_{1} x_{1j} + v_{2} x_{2j} + \cdots + u_{m} y_{mj} }} = \frac{{\mathop \sum \nolimits_{r = 1}^{n} u_{r} y_{rj} }}{{\mathop \sum \nolimits_{s = 1}^{m} v_{s} x_{sj} }}$$
(6)
where x and y are input and output and v and u are input and output weights respectively, s is number of inputs (s = 1, 2,…, m), r is number of outputs (r = 1, 2,…, n) and j represents jth farms (j = 1, 2, …, k). The estimated values of technical efficiencies lie between 0 and 1.
There are two models in DEA, first, Charnes–Cooper–Rhodes (CCR) model, which assumes the constant return to scale (CRS). The efficiency level that is assessed from it is the technical efficiency (TE), where the performance of farms is evaluated relative to other farms. The second model is Banker–Charnes–Cooper (BCC) that supposes the variable return to scale (VRS). The linear form of estimating the input oriented technical efficiencies of farms in CCR model is written as (Cooper et al. 2007):
$$\begin{aligned} {\text{Minimize}}\,\,\theta \hfill \\ {\text{Subjected to}}: \hfill \\ \quad \quad \quad \quad Y\lambda \ge y_{j} \hfill \\ \quad \quad \quad \quad \theta x_{j } - X\lambda \ge 0 \hfill \\ \quad \quad \quad \quad \lambda \ge 0 \hfill \\ \end{aligned}$$
(7)
where \(\theta\) is a technical efficiency of jth farms’ and \(\lambda\) is the intensity vectors of the weights of the efficient farms. \(y_{j }\) is the s × 1 vector of the value of desirable outputs from jth farms’ and \(x_{j }\) is the m × 1 vector of the value of inputs used in jth farm’s. The data for all n farms are represented as: \(Y\) is the s × n matrix of desirable outputs and X is the m × n matrix of inputs.
A pure technical efficiency (PTE) that is projected from the BCC model, splits the overall farm efficiency into TE and scale efficiency (SE) (Banker et al. 1984). The farm is considered as a typical VRS activity due to potential returns of scale and adding an additional constraint of \(\sum \lambda_{j } = 1\) into Eq. (7) leads to as VRS frontier. The BCC model is more flexible and envelops the data in a tighter way than the CCR model and the efficiency score from the BCC model is equal to or greater than the CCR. In relation with pure technical efficiency, Chauhan et al. (2006) explained TE as:
$$TE = PTE \times SE$$
(8)

This breakdown indicates the basis of inefficiency. The scale inefficiency of farms can be caused by its size, which shows that certain part of the inefficiency is due to inadequate size of the farms. The overall efficiency of the farm can be enhanced by selecting the farm with suitable size at the similar level of inputs used.

Environmental impacts

Key GHGs like carbon dioxide (CO2), methane (CH4) and nitrous oxide (N2O) were considered for estimating the amount of GHG emissions under different CPSs. These gases were suitably expressed as equivalent to CO2. The CO2 emission coefficients of various environmental detrimental inputs supplied in the CPSs were used for the estimation of GHG emissions. The emissions from FYM were calculated through its nutrient contents (1.87% N, 0.52% P2O5, 0.9% K2O).

The CO2 emission coefficient for a kg of fertilizer-element was used at 1.3, 0.2 and 0.15 kg CO2e for N, P2O5 and K2O, respectively, including their production, transportation, storage and transfer processes (Soni et al. 2013; Pokhrel and Soni 2017). The CO2e emission from a kg of N used at 4.87 kg CO2e for direct N2O emissions from organic and synthetic fertilizers, 1.096 kg CO2e for direct N2O emissions from N runoff or leaching, and 0.487 kg CO2e from synthetic fertilizers and 0.974 kg CO2e from organic fertilizers for indirect N2O emissions from atmospheric decomposition of N volatilized as NO2 and NH3 were used (IPCC 2006; Pokhrel and Soni 2017, 2019). Similarly, the emission at 74.1 × 10−3, 21 × 10−5 and 19 × 10−5 kg CO2e from the release of CO2, CH4 and NO2 gases, respectively, were considered for each MJ of diesel fuel used (IPCC 2006; Pokhrel and Soni 2017, 2018, 2019), while it was 0.071 kg CO2e MJ−1 for the machinery used in the study (Dyer and Desjardins 2006).

The default seasonal integrated emission factor of 10 g CH4 m−2 was adopted for calculating the CH4 emissions from rice field under continuously flooded condition without organic amendments. The CH4 emission factor of India was used because of unavailability of emission factor of Nepal and similarity in rice cultivation practices. In relative to emission factors for continuously flooded condition, the scaling factors 0.5 for intermittently flooded single and 0.2 for multiple aeration condition, and 2 for the organic amendments were used to calculate CH4 emission under specific conditions (IPCC 2006).

Economic analysis

A total financial value of production inputs used in CPSs such as costs of seeds, fertilizers, labor, draft animal, diesel fuel and rented machinery were used for determining the total cost of production. For calculating total return, the financial value of main product was used. Similarly, the gross return was intended by subtracting total cost of production from the total return, whereas return–cost ratio was calculated by dividing total return by total cost of production.

Statistical analysis

Statistical analysis was performed to interpret the collected data of different CPSs with SPSS 16.0 statistical software. A Tukey’s honest significant difference test was done to measure the significant difference of mean values at P ≤ 0.05 level. Similarly, descriptive statistics was used to compute the standard errors of mean. A regression statistical technique was also used to investigate the relationship between GHG emissions and crop yield.

Results

Energy indices of CPSs

Energy sources used in different CPSs were differed significantly (P < 0.05). The value of seed energy was highest in wheat (2.46 GJ ha−1) due to its high energy content seeds of about 155 kg ha−1. Whereas, onion (0.02 GJ ha−1) consumed the lowest seed energy, as the lowest amount of seeds (12 kg ha−1) were used in onion production system. Labor energy requirements for onion (4.87 GJ ha−1) and garlic (3.24 GJ ha−1) were very high compared with other crops. Similarly, energy requirement of draft animal was higher in onion (1.45 GJ ha−1) and garlic (1.26 GJ ha−1) compared with other crops, because almost all farmers in the study area used draft animal for land preparation in garlic and onion CPSs in their small-sized farms.

The use of chemical fertilizers (3.81 GJ ha−1) and diesel (2.13 GJ ha−1) were relatively higher in wheat compared with other CPSs. Conversely, the farm tools energy input was higher for onion (0.62 GJ ha−1), garlic (0.50 GJ ha−1) and maize (0.48 GJ ha−1) than other crops because of higher use of cart, animal drawn plough, spade and sickle in fertilizer application, ploughing, weeding and harvesting activities (Table 2).
Table 2

Energy inputs (GJ ha−1) in different crops

Inputs

Cereals

Legumes

Vegetables

Rice

Wheat

Maize

Lentil

Mungbean

Garlic

Onion

Seeds

1.56 ± 0.036 b

2.46 ± 0.052 a

0.61 ± 0.018 d

0.77 ± 0.046 c

0.52 ± 0.012 d

0.86 ± 0.024 c

0.02 ± 0.001 e

Labor

1.39 ± 0.027 cd

1.04 ± 0.028 e

1.64 ± 0.029 c

0.73 ± 0.036 f

1.32 ± 0.034 d

3.24 ± 0.125 b

4.87 ± 0.117 a

Draft animal

1.12 ± 0.111 ab

0.51 ± 0.093 c

0.71 ± 0.110 bc

0.48 ± 0.093 c

0.82 ± 0.135 bc

1.26 ± 0.030 a

1.45 ± 0.055 a

Chemical fertilizers

2.99 ± 0.124 b

3.81 ± 0.135 a

3.52 ± 0.141 a

0.68 ± 0.087 c

0.39 ± 0.094 c

Farmyard manure

1.09 ± 0.130 c

0.57 ± 0.078 c

2.87 ± 0.323 b

6.08 ± 0.258 a

5.15 ± 0.356 a

Diesel fuel

1.19 ± 0.136 b

2.13 ± 0.114 a

1.23 ± 0.125 b

1.06 ± 0.133 bc

0.53 ± 0.117 cd

0.15 ± 0.017 d

0.47 ± 0.061 d

Farm machinery

0.08 ± 0.009 b

0.13 ± 0.007 a

0.07 ± 0.012 b

0.07 ± 0.009 bc

0.03 ± 0.008 cd

0.01 ± 0.001 d

0.02 ± 0.002 d

Farm tools

0.28 ± 0.008 c

0.29 ± 0.008 c

0.48 ± 0.012 b

0.16 ± 0.005 d

0.05 ± 0.005 e

0.50 ± 0.024 b

0.62 ± 0.021 a

Data are means ± standard error. Means followed by the same letter in each row do not differ by Tukey’s honest significant difference tests at P < 0.05

A significant difference was found among the crops on key energy indices (Table 3). The highest energy input was required for onion (12.64 GJ ha−1) and garlic (12.09 GJ ha−1), which used higher amount of farmyard manure and labor than other crops. Furthermore, energy requirements for garlic, maize and wheat were found statistically at par (P > 0.05). In contrast, mungbean (3.64 GJ ha−1) and lentil (3.96 GJ ha−1) CPSs consumed the least amount of energy input. The energy output was highest from rice (68.99 GJ ha−1), whilst it was found lowest from lentil (9.17 GJ ha−1) and mungbean (11.23 GJ ha−1), because of lowered yields of legume crops. Similarly, the highest energy balance occurred in rice (59.29 GJ ha−1), whilst it was lowest from garlic (2.78 GJ ha−1) and onion (3.72 GJ ha−1). The results on energy input–output indicated that the best energy ratio was achieved from rice (7.14), followed by maize (5.37). In contrast, garlic (1.22) and onion (1.27) CPSs had the lowest energy ratios.
Table 3

Energy indices of different crops

Items

Cereals

Legumes

Vegetables

Rice

Wheat

Maize

Lentil

Mungbean

Garlic

Onion

Energy input (GJ ha−1)

9.70 ± 0.165 c

10.95 ± 0.148 bc

11.14 ± 0.459 b

3.96 ± 0.103 d

3.64 ± 0.118 d

12.09 ± 0.298 ab

12.64 ± 0.493 a

Energy output (GJ ha−1)

68.99 ± 1.629 a

38.09 ± 0.471 c

56.58 ± 0.960 b

9.17 ± 0.357 f

11.23 ± 0.314 ef

14.88 ± 0.861 de

16.36 ± 1.238 d

Energy balance (GJ ha−1)

59.29 ± 1.555 a

27.14 ± 0.420 c

45.45 ± 0.742 b

5.21 ± 0.327 de

7.59 ± 0.237 d

2.78 ± 0.676 e

3.72 ± 0.949 de

Energy ratio

7.14 ± 0.154 a

3.49 ± 0.046 c

5.37 ± 0.204 b

2.33 ± 0.084 d

3.11 ± 0.060 c

1.22 ± 0.053 e

1.27 ± 0.067 e

Energy productivity (kg GJ−1)

485.70 ± 10.495 c

237.50 ± 3.120 e

365.05 ± 13.870 d

158.73 ± 5.725 f

214.51 ± 4.165 ef

760.52 ± 32.844 a

668.76 ± 35.351 b

Specific energy (MJ kg−1)

2.07 ± 0.045 d

4.24 ± 0.053 b

2.87 ± 0.095 c

6.58 ± 0.279 a

4.71 ± 0.095 b

1.40 ± 0.072 e

1.63 ± 0.095 de

Energy intensiveness (MJ USD−1)

20.64 ± 0.249 b

24.37 ± 0.223 a

22.25 ± 0.731 b

14.26 ± 0.226 c

11.25 ± 0.205 d

9.73 ± 0.205 d

10.25 ± 0.209 d

Data are means ± standard error. Means followed by the same letter in each row do not differ by Tukey’s honest significant difference tests at P < 0.05

Significantly, the higher energy productivity was recorded in vegetables than cereals followed by legumes. Moreover, the highest energy productivity (760.52 kg GJ−1) and the lowest specific energy (1.40 MJ kg−1) were found in garlic, among the CPSs. Conversely, the lowest energy productivity (158.73 kg GJ−1) and the highest specific energy (6.58 MJ kg−1) were recorded in lentil, which can be explained by lower productivity of legumes as compared to cereals and vegetables.

Farms’ efficiency under different crops: DEA approach

The analyzed data of technical efficiencies from DEA models revealed a significant difference (P < 0.01) in TE, PTE and SE scores of different CPSs (Table 4). The frequency distribution of TE showed that most of the farms were technically inefficient in all the crop production systems, where the majority farms in each category had TE scores less than 80% except mungbean farms. The frequency distribution result in CCR model indicated that more farms were technically efficient in mungbean (7%) followed by wheat (5%), whereas few farms were technically efficient in rice (2%) followed by maize (3%). Similarly, the frequency distribution result in BCC model indicated that more farms were technically efficient in mungbean, garlic and onion (17%), followed by maize (13%).
Table 4

Efficiency distribution of farms under different crop production systems

Particulars

Cereals

Legumes

Vegetables

Rice

Wheat

Maize

Lentil

Mungbean

Garlic

Onion

Technical efficiency (CCR)

 Efficient farms

1

2

1

1

2

1

1

 Inefficient farms

  TE > 90%

2

1

4

2

7

4

2

  TE 80–90%

10

8

5

11

13

6

3

  TE 70–80%

15

24

3

5

6

7

7

  TE 60–70%

10

2

9

5

2

5

4

  TE < 60%

4

17

6

7

13

 Number of farms

42

37

39

30

30

30

30

 Minimum efficiency score

0.57

0.67

0.47

0.43

0.64

0.39

0.31

 Average efficiency scores

0.751 ± 0.016 bc

0.802 ± 0.011 ab

0.662 ± 0.025 cd

0.749 ± 0.027 bc

0.852 ± 0.017 a

0.717 ± 0.031 bcd

0.649 ± 0.034 d

 Minimum efficiency score

0.57

0.67

0.47

0.43

0.64

0.39

0.31

Pure technical efficiency (BCC)

 Efficient farms

3

3

5

2

5

5

5

 Inefficient farms

  TE > 90%

8

3

3

4

4

9

6

  TE 80–90%

17

19

4

11

14

10

6

  TE 70–80%

12

11

5

7

5

6

6

  TE 60–70%

2

1

8

5

2

6

  TE < 60%

14

1

1

 Number of farms

42

37

39

30

30

30

30

 Minimum efficiency score

0.69

0.68

0.50

0.59

0.64

0.72

0.53

 Average efficiency score

0.854 ± 0.013 a

0.842 ± 0.012 a

0.722 ± 0.027 b

0.819 ± 0.021 a

0.867 ± 0.017 a

0.884 ± 0.016 a

0.835 ± 0.024 a

Scale efficiency

0.881 ± 0.015 bc

0.953 ± 0.007 ab

0.926 ± 0.018 ab

0.912 ± 0.019 ab

0.983 ± 0.007 a

0.806 ± 0.027 cd

0.776 ± 0.033 d

Data are means ± standard error. Means followed by the same letter in each row do not differ by Tukey’s honest significant difference tests at P < 0.05

Moreover, the results indicated that about 80% of total farms had TE score less than 0.80 in onion, whereas about 70% of total farms had efficiency less than 80% in rice, wheat and maize CPSs. Similarly, about 60, 50 and 27% of total farms had efficiency score less than 0.80 in garlic, lentil and mungbean CPSs, respectively. Thus, technically inefficient farms of different CPSs need to change the tendency of inputs use in their CPSs for improving TE and sustainability of production system.

The average TE score of farms was estimated significantly higher in mungbean (0.852), followed by wheat, whereas the lowest TE was estimated in onion (0.649). Further breakdown of TE score into PTE and SE indicated higher PTE was in garlic (0.884) followed by mungbean, whilst the lowest PTE was achieved in maize (0.722). The scale efficient farms were observed in mungbean, wheat, lentil and maize than other crops.

Environmental impacts of CPSs

The total GHG emissions from different CPSs varied significantly at P < 0.05 (Table 5). The highest GHG emissions corresponded to garlic and onion at 2997.13 and 2583.46 kg CO2e ha−1, respectively, due to the large amount of fertilizer used in vegetable CPSs. Farmers used higher amount of farmyard manure in vegetable crops like garlic (11–25 t ha−1) and onion (7–23 t ha−1) than in other crops, where it contributed more than 90% of total emissions in those crops. In contrast, mungbean and lentil CPSs released the least GHG at 74.73 and 141.31 kg CO2e ha−1, respectively, because farmers in the study area grow legumes with very low inputs of fertilizers and irrigation.
Table 5

GHG emissions (kg CO2e ha−1) from different crops

Emissions

Cereals

Legumes

Vegetables

Rice

Wheat

Maize

Lentil

Mungbean

Garlic

Onion

F value

Machinery

5.40

9.53

5.26

4.87

2.42

0.36

1.16

Diesel fuel

88.75

158.70

91.77

78.92

39.21

11.03

35.28

Fertilizers

835.07

675.26

1782.79

57.52

33.10

2985.74

2547.01

CH4 emissions from rice field

853.66

Total GHG emissions

1786.35 ± 74.716 b

843.49 ± 37.124 c

1879.82 ± 185.704 b

141.31 ± 12.195 d

74.73 ± 15.064 d

2997.13 ± 127.049 a

2583.46 ± 175.959 a

Data are means ± standard error. Means followed by the same letter in each row do not differ by Tukey’s honest significant difference tests at P < 0.05

Methane (CH4) that emitted from flooded rice field due to anaerobic decomposition of organic materials is one of the most important sources of GHG effect. In rice, CH4 emissions from field share about 48% in the total GHG emissions. The emissions from machinery and diesel fuel inputs were higher in wheat, among the CPSs.

The coefficient of determination (R2) between crop yield and total GHG emissions for rice, maize, wheat, lentil, mungbean, garlic and onion were found at 0.33, 0.41, 0.23, 0.07, 0.24, 0.36 and 0.35, respectively (Figs. 1, 2, 3).
Fig. 1

Crop yield versus total GHG emissions in cereal crops

Fig. 2

Crop yield versus total GHG emissions in legume crops

Fig. 3

Crop yield versus total GHG emissions in vegetable crops

Economic performance of CPSs

The results of crop yields and economic analyses of CPSs are summarized in Table 6. The vegetable crops like garlic and onion produced significantly higher yields than cereals, followed by legume CPSs. However, vegetable CPSs required the highest cost for production per hectare (USD 1221.90–1247.67 ha−1) owing to higher use of farm labor and FYM. In contrast, the least cost of production was recorded in legume CPSs, which used lower amount of farm inputs as compared to other CPSs. The average number of labor (labor days) per hectare used in the production of vegetable, cereal and legume CPSs were found as 290, 102 and 70, respectively, in the study area. Moreover, the costs of production of rice, wheat and maize were found statistically similar (P > 0.05).
Table 6

Crop yields and economic analyses of different crops

Indicators

Cereals

Legumes

Vegetables

Rice

Wheat

Maize

Lentil

Mungbean

Garlic

Onion

Crop yield (kg ha−1)

4693.00 ± 110.818 c

2591.49 ± 32.056 d

3849.23 ± 65.298 c

623.61 ± 24.269 e

773.45 ± 21.598 e

9299.00 ± 538.399 a

8611.83 ± 651.524 b

Total cost of production (USD ha−1)

470.33 ± 6.750 b

449.62 ± 4.808 b

497.69 ± 11.627 b

278.13 ± 6.804 c

322.31 ± 7.014 c

1247.67 ± 27.015 a

1221.90 ± 27.711 a

Total return (USD ha−1)

916.25 ± 21.636 c

542.98 ± 6.717 c

916.48 ± 15.547 c

593.91 ± 23.113 c

589.30 ± 16.455 c

4428.10 ± 256.381 a

3690.79 ± 279.250 b

Gross return (USD ha−1)

445.92 ± 20.528 c

93.36 ± 5.939 c

418.79 ± 13.827 c

315.79 ± 22.726 c

266.99 ± 11.951 c

3180.42 ± 240.233 a

2468.89 ± 260.016 b

Return–cost ratio

1.96 ± 0.046 c

1.21 ± 0.014 d

1.87 ± 0.040 c

2.16 ± 0.084 c

1.83 ± 0.033 c

3.52 ± 0.165 a

2.96 ± 0.180 b

Data are means ± standard error. Means followed by the same letter in each row do not differ by Tukey’s honest significant difference tests at P < 0.05

The total return and gross return were significantly higher from onion and garlic, among the CPSs, due to their higher yield and market value compared with other CPSs. Therefore, the highest return–cost ratios were recorded in garlic (3.52) followed by onion (2.96). Further, results showed that total returns and gross returns among the cereals (except wheat) and legume CPSs were observed statistically at par.

Discussion

The frequency of labor involvement and use of FYM during crop production process were higher in vegetable crops as compared to cereal and legume CPSs. Therefore, vegetable crops like onion and garlic required higher energy input than cereals, followed by legume CPSs. Elsoragaby et al. (2019) also stated higher energy use in garlic compared with rice, maize and wheat CPSs, which mostly depended on labor and FYM. Results indicated that farmers did not use chemical fertilizers in garlic and onion CPSs, whilst they did not use FYM in lentil and mungbean CPSs; which are a common practice in Nepal. The level of farm mechanization is in general very low in Nepal, though it is increasing in trend. It is relatively higher in wheat compared with other crops. Energy consumptions from the use of farm machinery and diesel were therefore higher in wheat than other crops.

The energy input–output and energy ratios of rice and wheat in the current study are similar to the results of Pokhrel and Soni (2019). In contrast, the higher energy requirements for different CPSs as compared to the current study are reported by many researchers such as: in rice (Yodkhum et al. 2018; Elsoragaby et al. 2019), wheat (Mondani et al. 2017; Yuan et al. 2018; Elsoragaby et al. 2019; Nasseri 2019), maize (Elsoragaby et al. 2019; Juarez-Hernandez et al. 2019) and garlic (Elsoragaby et al. 2019). The higher energy consumption in their CPSs was because of higher use of chemical fertilizers and diesel fuel in crop production process.

The energy ratios of different CPSs varied from 1.22 (garlic) to 7.14 (rice). The lowest energy ratio of garlic was due to the low energy content in the outputs. The variation in energy ratio of different CPSs is also reported in literature, due to the differences in crop management practices adopted. For example, the energy output-input ratios are reported at 4.97–7.14 for rice (Pokhrel and Soni 2019), 4.96–10.82 for maize (Saad et al. 2016; Elsoragaby et al. 2019), 2.20–5.53 for wheat (Nasseri 2019) and 0.7 for garlic (Elsoragaby et al. 2019).

Many researchers expressed the relation of energy input to the yield and cost of production in different CPSs. The results showed that 2.07 MJ energy was required to harvest a kilogram of rice. In contrast to this study, Yodkhum et al. (2018) reported the total energy use was at 1.80 MJ kg−1 for rice production. However, specific energy and energy productivity of rice and wheat in the current study are compatible with Pokhrel and Soni (2019). The differences of these indicators for different CPSs were because of change in type and amount of input and output generated in various crop production processes.

The energy productivity of lentil, garlic and onion CPSs in the study are found higher than reported by Koocheki et al. (2011), Samavatean et al. (2011) and Zarini et al. (2014) in respective CPSs. Similarly, energy intensiveness is also achieved higher in this study than that claimed by Samavatean et al. (2011) in garlic and Zarini et al. (2014) in onion. The lower energy productivity and energy intensiveness in their study was due to higher demand of agrochemicals and farm machinery in their crop production systems.

The DEA analysis was performed to estimate the energy saving potential of farms under various CPSs. Results of frequency distribution and average TE of farms under different CPSs are consistent with the finding of Nassiri and Singh (2009). A significant variation in TE among the farms within same crop indicated that the proper energy inputs were not applied in adequate quantity in their CPSs. Pokhrel and Soni (2017) also observed technical inefficiency of CPSs because of inappropriate production technology and inputs used. The average TE scores in this study signified that energy inputs can be saved from 18 to 35% without negotiating the economic outputs, under different CPSs, by raising TE of farms to efficient level. The use of farm inputs in adequate quantity and time in CPSs results in higher TE score. The average TE, PTE, and SE scores of rice and wheat in this study are compatible to Nassiri and Singh (2009) and Khoshnevisan et al. (2013), respectively.

The GHG emissions in the study revealed that vegetable CPSs released more GHG than cereals, followed by legumes. This can be explained by the fact that vegetable crops required more fertilizers than cereals, followed by legume production systems. The lower values of CO2e emissions from the legume crops in the study are in line with existing literature. Maraseni and Cockfield (2012) and Soni et al. (2013) reported the lower emissions from chickpea (Cicer arietinum) and soybean (Glycine max), respectively, due to lower use of fertilizers.

The result of CO2e emissions from rice in this study is compatible with the findings of other researchers such as for rice and wheat (Pokhrel and Soni 2019) and maize (Juarez-Hernandez et al. 2019). The variation in CO2e emissions in CPSs was mainly due to the fertilizer application rate and level of farm mechanization. The CO2e emissions from wheat cultivation in the current study seemed higher than reported by Taki et al. (2018), owing to higher use of fertilizers and diesel fuel. Conversely, Mondani et al. (2017) reported CO2e emissions (from wheat) higher than this study due to the increased level of farm mechanization in wheat production.

The contribution of fertilizer was found higher in the total GHG emissions, where its proportion was recorded as > 90% in garlic, onion and maize, > 80% in wheat and > 40% of the total emissions in rice, lentil and mungbean CPSs. Pokhrel and Soni (2019) noted fertilizers as the major source of increasing GHG releases in rice and wheat production systems. Taki et al. (2018) also claimed that the higher use of fertilizers significantly contributes to the total GHG emissions.

Similarly, the results of regression analyses indicated that the variation of GHG emissions from maize had a major impact on the yield as compared to other crops. A significant relationship is not seen in increasing crop yield with increasing GHG emissions. The yield disparity in those crops might be due to some other management practices. A good fertilization technique could help for producing a good crop with reducing environmental impacts.

The current results revealed that the crop production cost was significantly higher in garlic and onion CPSs than maize, rice and wheat, followed by mungbean and lentil. The higher cost of production in CPSs was associated with higher involvement of labor, use of agrochemicals and farm machineries. Results of this study conform the findings of Ramsden et al. (2017), as they noted higher cost of production of garlic compared with rice, maize and mungbean. Nevertheless, the highest gross returns and return–cost ratios were associated in garlic and onion CPSs because of better economic yields and market prices. The economic benefit of growing legume crops was higher than wheat cultivation, though it produced the lowest yield, which was mainly because of comparative advantages of lowered production costs. Similarly, return–cost ratio as the major economic indicators of CPSs was found to vary from 1.21 in wheat to 3.52 in garlic. These variations of return–cost ratios in CPSs are due to differences in management practices adopted.

Conclusions

Diverse CPSs required different types and amount of crop production inputs to produce both the desirable and undesirable outputs. The vegetable crops like onion and garlic are the highest energy consumers, but the energy output is higher in rice and maize CPSs. Most of the farms under each crop are technically inefficient, with efficiency scores less than 0.80. Profitability of vegetable CPSs is higher than cereals, followed by legumes but the CO2e emissions are found maximum from vegetable crops.

There is a great potential of increasing energy use efficiency and crop output with reducing GHG emissions. An adaptation of improved crop production practices could help in reducing energy use and environmental effects without compromising the financial outputs for environmentally sustainable agricultural production system.

Notes

Acknowledgements

The authors gratefully acknowledge necessary support received from Agricultural Learning Exchanges in Asian Regional Networking (AgLEARN) project of the USAID/RDMA and Asian Institute of Technology (AIT), Thailand. Cooperation of the farmers during field survey and data collection is thankfully appreciated.

References

  1. Arora NK (2018) Environmental Sustainability-necessary for survival. Environ Sust 1:1–2Google Scholar
  2. Arora NK, Fatima T, Mishra I, Verma M, Mishra J, Mishra V (2018) Environmental sustainability: challenges and viable solutions. Environ Sust 1:309–340Google Scholar
  3. Banker RD, Charnes A, Cooper WW (1984) Some models for estimating technical and scale in efficiencies in data envelopment analysis. Manag Sci 30:1078–1092CrossRefGoogle Scholar
  4. Chauhan NS, Mohapatra PKJ, Pandey KP (2006) Improving energy productivity in paddy production through benchmarking: an application of data envelopment analysis. Energy Conv Manag 47:1063–1085CrossRefGoogle Scholar
  5. Cooper WW, Seiford LM, Tone K (2007) Data envelopment analysis: a comprehensive text with models, applications, references and DEA-Solver software, 2nd edn. Springer, New YorkGoogle Scholar
  6. Dyer JA, Desjardins RL (2006) Carbon dioxide emissions associated with the manufacturing of tractors and farm machinery in Canada. Biosyst Eng 93:107–118CrossRefGoogle Scholar
  7. Elsoragaby S, Yahya A, Mahadi MR, Nawi NM, Mairghany M (2019) Energy utilization in major crop cultivation. Energy 173:1285–1303CrossRefGoogle Scholar
  8. FAOSTAT (2019) Data. http://www.fao.org/faostat/en/#data. Accessed 24 Apr 2019
  9. IPCC (2006) IPCC guidelines for national greenhouse gas inventories. https://www.ipcc-nggip.iges.or.jp/public/2006gl/vol4.html. Accessed 11 Sept 2018
  10. Juarez-Hernandez S, Uson S, Pardo CS (2019) Assessing maize production systems in Mexico from an energy, exergy, and greenhouse-gas emissions perspective. Energy 170:199–211CrossRefGoogle Scholar
  11. Khoshnevisan B, Rafiee S, Omid M, Mousazadeh H (2013) Applying data envelopment analysis approach to improve energy efficiency and reduce GHG (greenhouse gas) emission of wheat production. Energy 58:588–593CrossRefGoogle Scholar
  12. Koocheki A, Ghorbani R, Mondani F, Alizade Y, Moradi R (2011) Pulses production systems in term of energy use efficiency and economical analysis in Iran. Int J Energy Econ Policy 1:95–106Google Scholar
  13. Maraseni TN, Cockfield G (2012) Including the costs of water and greenhouse gas emissions in a reassessment of the profitability of irrigation. Agric Water Manag 103:25–32CrossRefGoogle Scholar
  14. MESD (2018) Statistical information on Nepalese agriculture 2016/2017. Government of Nepal, Ministry of Agriculture, Land Management and Cooperatives, Monitoring, Evaluation and Statistics Division, Agriculture Statistics Section, Singhdurbar, Kathmandu, NepalGoogle Scholar
  15. Mondani F, Aleagha S, Khoramivafa M, Ghobadi R (2017) Evaluation of greenhouse gases emission based on energy consumption in wheat Agroecosystems. Energy Rep 3:37–45CrossRefGoogle Scholar
  16. Nasseri A (2019) Energy use and economic analysis for wheat production by conservation tillage along with sprinkler irrigation. Sci Total Environ 648:450–459CrossRefGoogle Scholar
  17. Nassiri SM, Singh S (2009) Study on energy use efficiency for paddy crop using data envelopment analysis (DEA) technique. Appl Energy 86:1320–1325CrossRefGoogle Scholar
  18. Pokhrel A, Soni P (2017) Performance analysis of different rice-based cropping systems in tropical region of Nepal. J Environ Manag 197:70–79CrossRefGoogle Scholar
  19. Pokhrel A, Soni P (2018) An investigation into a resource and environmentally sustainable system: zero-tillage lentil and garlic production in Nepal. Agroecol Sustain Food Syst 42:982–1002CrossRefGoogle Scholar
  20. Pokhrel A, Soni P (2019) Energy balance and environmental impacts of rice and wheat production: a case study in Nepal. Int J Agric Biol Eng 12(201):207Google Scholar
  21. Prasad SK, Pullabhotla H, Ganesh-Kumar A (2011) Supply and demand for cereals in Nepal, 2010–2030. http://ebrary.ifpri.org/cdm/ref/collection/p15738coll2/id/12494. Accessed 12 Jan 2018
  22. Ramsden SJ, Wilson P, Phrommarat B (2017) Integrating economic and environmental impact analysis: the case of ricebased farming in northern Thailand. Agric Syst 157:1–10CrossRefGoogle Scholar
  23. Saad AA, Das TK, Rana DS, Sharma AR, Bhattacharyya R, Lal K (2016) Energy auditing of a maize-wheat-greengram cropping system under conventional and conservation agriculture in irrigated north-western Indo-Gangetic Plains. Energy 116:293–305CrossRefGoogle Scholar
  24. Samavatean N, Rafiee S, Mobli H, Mohammadi A (2011) An analysis of energy use and relation between energy inputs and yield, costs and income of garlic production in Iran. Renew Energy 36:1808–1813CrossRefGoogle Scholar
  25. Soni P, Taewichit C, Salokhe VM (2013) Energy consumption and CO2 emissions in rainfed agricultural production systems of Northeast Thailand. Agric Syst 116:25–36CrossRefGoogle Scholar
  26. Taki M, Soheili-Fard F, Rohani A, Chen G, Yildizhan H (2018) Life cycle assessment to compare the environmental impacts of different wheat production systems. J Clean Prod 197:195–207CrossRefGoogle Scholar
  27. WECS (2010) Energy sector synopsis report 2010. Water and Energy Commission Secretariat, KathmanduGoogle Scholar
  28. World Bank (2017) Indicators. http://data.worldbank.org/indicator. Accessed 24 Sept 2017
  29. Yodkhum S, Sampattagul S, Gheewala SH (2018) Energy and environmental impact analysis of rice cultivation and straw management in northern Thailand. Environ Sci Pollut Res.  https://doi.org/10.1007/s11356-018-1961-y CrossRefGoogle Scholar
  30. Yuan S, Peng S, Wang D, Man J (2018) Evaluation of the energy budget and energy use efficiency in wheat production under various crop management practices in China. Energy 160:184–191CrossRefGoogle Scholar
  31. Zarini RL, Akram A, Kolour RT, Firouzjaee RA (2014) Energy use, economic analysis and mechanization index of onion production in Iran. J Middle East Appl Sci Technol 1:344–351Google Scholar

Copyright information

© Society for Environmental Sustainability 2019

Authors and Affiliations

  1. 1.Grain Legumes Research ProgramNepal Agricultural Research Council (NARC)KathmanduNepal
  2. 2.Department of Agricultural and Food EngineeringIndian Institute of Technology KharagpurKharagpurIndia

Personalised recommendations