Due to its physical, natural and social characteristics, Mexico has a real potential to be a leader in aquaculture. Most of the aquaculture production for this species, with more than 200 aquaculture plants in Sonora. This study focuses on the application of LCA + DEA methodology to assess the ecoefficiency of 38 semi- intensive shrimp farms located in the state of Sonora. LCA results showed that feed management and electricity consumption are the main critical points in almost all the impact categories. Further improvement actions were evaluated, the replacement of wheat meal for Dried Distiller Grains with Solubles (DDGS) resulted in environmental impact reductions ranged from 2% to 57%, depending on the impact category. On the other hand, the installation of photovoltaic panels in the area was evaluated, looking for a shift towards a less carbon-intensive energy production. Overall, the implementation of these improvement measures will contribute to increased environmental protection and resource efficiency.
Shrimps are produced in three models of farming systems: extensive, semi-intensive and intensive. The differences lie in the level of technology applied, the control of physical-chemical and biometric variables, water consumption and the frequency of meal dosage. In recent years, the expansion of aquaculture systems has been accompanied by an intensification of the system and has generated social concerns on the associated sustainability issues. In this context, Life Cycle Assessment (LCA) is considered an appropriate methodology to evaluate the environmental impacts associated with shrimp farming in Mexico.
Within this framework, the goal of the current study was to apply the large amount of data available to carry out an environmental and eco efficiency assessment of 38 semi-intensive farms located in Sonora using a combined LCA and DEA approach.

The environmental and eco efficiency analyses were conducted in order to detect critical activities in the environmental profile of the process, identify operational inefficiencies, set input reduction objectives and compute the environmental impacts of inefficient practices in shrimp farming.
The results of the eco-efficiency analysis will allow a realistic proposal of alternatives to improve environmental performance by identifying those facilities that under similar conditions may act as reference for their peers. This document also proposes the definition of a roadmap for more sustainable aquaculture production with a view to future environmental certification.
Materials and methods
System overview
In Mexico, most of the national shrimp production is concentrated in the northwest region, specifically in the states of Sonora and Sinaloa, where semi-intensive farms are the most abundant. The vegetation associated with shrimp farming in the state of Sonora is mostly semi-desertic, xeromorphic and succulents shrubdominated, arranged in a matrix, so environmental impacts associated with land use change are not expected to be relevant. For this reason, in the present study, these impacts were excluded, attention was focused on impacts related to nutrient emissions to water.
The LCA + DEA framework
The methodology is structured in 5 steps:
i) data collection and construction of life cycle inventory for each DMU;
ii) determination of the life cycle environmental impacts of each DMU through the LCA methodology;
iii) implementation of the DEA model to obtain the efficiency scores and operational objectives for each DMU. These operational objectives represent reductions in input consumption while maintaining output production;
iv) impact assessment of LCI for new virtual DMUs based on the operational reductions established in step 3;
v) interpretation of the results obtained, comparison among DMUs and verification of inefficient practices.

LCA methodology
The ISO 14040 and 14044 standards have been used as the basic methodology to carry out environmental assessment. These standards define the LCA phases as: goal and scope definition, inventory analysis, impact assessment and interpretation.
Goal and scope definition
The main objective of this case study is to analyze the significant environmental burdens of shrimp aquaculture and link them to operational inefficiencies. A secondary objective is to identify operational improvement actions to reach, totally or partially, the proposed theoretical goals.
Data collection and life cycle inventory
A total of 38 Mexican shrimp farms were inventoried in this case study. All the facilities are grouped into nine local boards of aquatic health. The information provided compiles relevant data to understand the operation of the different farms and comprises the following variables: Farming area (ha), stock density (organisms/m2 ), total shrimp production (t), survival rate (%) and Feed Conversion RatioFCR.
Water exchange in ponds was based on agricultural records of the region. Direct emissions of suspended solids, nitrogen and phosphorus were obtained following the guidelines provided by farm managers.
Impact assessment
To convert the extensive list of life cycle inventory results into a useful list of environmental indicators, the following impact categories were selected: Global warming (GW), Stratospheric ozone depletion (SOD), Terrestrial acidification (TA), Freshwater eutrophication (FE), Marine eutro-phication (ME), Marine ecotoxicity (MET), Fossil resources scarcity (FRS) and Water consumption (WC).
DEA model selection
Based on different models described in the DEA methodology, three. of the most used ones were tested for the available dataset: Slacks-Based Measure (SBM), Charnes-Cooper Rhodes (CCR) and Epsilon-Based Measure (EBM). Finally, SBM model was selected as it follows a non-radial approach, which allows greater discrimination power to assess the efficiency of DMU than radial methods Input/output selection.

The DEA matrix used in this study was composed of 7 inputs and 1 output. These units were chosen for their operational importance and associated environmental impacts, according to the previous life cycle analysis.
Improvement actions
Once the critical stages in the environmental profile were deter- mined. The variation of the life cycle impact was estimated with respect to two fundamental elements: the formulation of the feed and the energy requirements of the larvae tanks. It should be noted that environmental burdens from water discharge are derived from the portion of feed that is not consumed by the animals and remains in the pond water.
All this leads to the proposal to replace some components of the feed with others of lesser environmental impact that result in similar levels of growth and survival. The shift from electricity production to photovoltaic panel generation was evaluated as another improvement action.
Results
Environmental burdens of current DMUs
SS1. Feed and SS2. Larvae are primarily responsible for environmental burdens in most impact categories, except for freshwater eutrophication and water consumption. The environmental burdens in the GW category come mainly from the electricity requirements of SS1 and SS2. These electrical consumptions are related to the milling of wheat and soybean grains to obtain meals and the need for aeration in the larvae tanks to maintain optimal growth conditions.
With respect to FE, SS3. Aquaculture is the main contributor due to direct phosphorus emissions (95%).
DEA calculation and efficiency scores
Of all the shrimp farms evaluated, just over 13% (5 of 38 farms) were found fully efficient (Φ = 1). However, although only 5 farms were considered fully efficient, the efficiency index can be considered high in general, as only four farms have efficiency values below 0.6 and an average efficiency of 0.79 is achieved.
Environmental burdens of virtual DMUs
As expected, the greatest reductions occurred on the farms with the lowest efficiencies, such as DMU 23 (69.9%) and DMU 31 (50.7%). While the smallest reductions were found on farms that were already close to full efficiency (DMU 6 and 30).
Improvement actions
In the view of the results, only the replacement of wheat meal by barley meal or by DDGS seems to be environmentally friendly. Analyzing barley meal in detail, the reductions in environmental impacts are limited, although a 14% decrease in the SOD category stands out.
The installation and use of photovoltaic panels would result in a 15% reduction in carbon footprint, in addition to a 10% reduction in TA and 23.2% in FRS. Bearing in mind that the high impact in this category is derived from a structure whose useful lifetime is quite long, it can be concluded that the implementation of these photovoltaic panels in the facilities will have a positive effect on the environmental impact.
Discussion
Chemicals are responsible for improving productivity in aquaculture systems by improving larval survival rates, feeding efficiency, and pathogen control, but they also have a negative impact on the environment due to their ecotoxicity.
The carbon footprint values reported in this study are slightly higher than those obtained in a previous study that evaluated organic shrimp production in Taiwan.

Therefore, it can be considered that the farms evaluated in this study have, in general, a good environmental performance, at least in terms of carbon footprint and terrestrial acidification, at similar levels to organic and certified production.
With regard to efficiency scores, the results obtained from the DEA study showed that only 5 of the 38 farms evaluated were considered efficient, which represents a low value compared to previous LCA/DEA studies applied to the agri-food sector. Although it was found that few DMUs were fully efficient, it is important to note that most DMUs achieved efficiency values above 0.5.

In fact, only 5 were found to be below 0.6. Therefore, the average efficiency value of the sample evaluated is a reasonably high value of 0.79
While it is true that inefficient farms do not have significantly lower production/feed values than efficient ones, the combination of the three ratios clearly gives the worst results.
This makes it clear that, in order to seek operational and environmental efficiency, action must be taken on all possible lines of action, prioritizing a balanced improvement of all variables.
Conclusions
The results showed that feed formulation and electricity consumption in larval tanks are the main “hotspots” of the process. As a result of the eco-efficiency analysis, several improvement actions were proposed that resulted in the convenience of installing photovoltaic panels and decreasing the food conversion ratio by substituting wheat meal in the feed. Substitution by DDGS proved to be the most promising option, ensuring reductions of between 2% and 57% depending on the impact categories.
This is a summarized version developed by the editorial team of Aquaculture Magazine based on the review article titled “ECO-EFFICIENCY ASSESSMENT OF SHRIMP AQUACULTURE PRODUCTION IN MEXICO” developed by: ANTONIO CORTÉS, RAMÓN CASILLAS-HERNÁNDEZ, CRISTINA CAMBESES-FRANCO, RAFAEL BÓRQUEZ LÓPEZ, FRANCISCO MAGALLÓN-BARAJAS, WALTER QUADROS-SEIFFERT, GUMERSINDO FEIJOO, MARIA TERESA MOREIRA. The original article was published on JULY 2021, through ELSEVIER B.V. under the use of a creative commons open access license.