Unveiling low-to-high-frequency data sampling caveats for aquaculture environmental monitoring and management

Sampaio, Fernanda Garcia; Araújo, Carlos A. S.; Dallago, Bruno Stéfano Lima; Stech, José; Lorenzzetti, João A.; Alcântara, Enner; Losekann, M. E.; Marin, Diego Bedin; Leão, Joaquim Antônio Dionísio; Bueno, Guilherme Wolff

doi:10.1016/j.aqrep.2021.100764

Cited by 5 publications

(8 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Data gaps or lack of observational data are not unique to aquaculture and similar challenges have been highlighted in other studies (Jones et al, 2016;Dorigo et al, 2017;Mottram et al, 2021). In addition, measurements do not necessarily represent true conditions as there can be representativity issues as well as inconsistencies and errors during data collection and analysis (Sampaio et al, 2021;Skogen et al, 2021), and it is important to also consider the context of how measurements are used in any model evaluation process. Many aquaculture datasets have relied on manual data collection and recording, with data availability and quality affected by many factors, including challenging weather conditions, monitoring equipment, lack of common protocols, and differences in how people collect and store the data.…”

Section: Challenges In Comparing Modelled Projections and Observationsmentioning

confidence: 99%

“…Furthermore, this study focused on a monthly comparison due to availability of data and visualization purposes, but it is important to acknowledge that many aspects of the health, welfare and behavior of aquatic species are influenced by the temperature of their surrounding environment at a much shorter time scale. Hence, for aquaculture purposes, and for variables like temperature, time averages can mask conditions that challenge the fish (Sampaio et al, 2021).…”

Section: Challenges In Comparing Modelled Projections and Observationsmentioning

confidence: 99%

“…Choice of approach will depend on several factors, including data availability, data quality and how the comparison will be interpreted. This study used comparisons of 5year monthly averages (and standard deviations), but monthly averages could miss conditions relevant to aquaculture production (Sampaio et al, 2021). Decisions on temporal resolution will be constrained by what data is available, but it is important not to forget the overall goal.…”

Section: Comparison Between Model Outputs and In-situ Measurementsmentioning

confidence: 99%

See 2 more Smart Citations

Context matters when using climate model projections for aquaculture

Falconer,

Ytteborg,

Goris

et al. 2023

Front. Mar. Sci.

View full text Add to dashboard Cite

At present, specific guidance on how to choose, assess and interpret climate model projections for the aquaculture sector is scarce. Since many aspects of aquaculture production are influenced by the local farm-level environment, there is a need to consider how climate model projections can be used to predict potential future farming conditions locally. This study compared in-situ measurements of temperature and salinity from Norwegian salmon farms and fixed monitoring stations to simulations from a regional ocean climate model for multiple locations and depths in southern Norway. For locations considered in this study, a similar seasonal cycle in terms of phasing was visible for modelled and measured temperatures. For some depths and times of the year the modelled and measured temperatures were similar, but for others there were differences. The model tended to underestimate temperature. On occasion there were differences between average modelled and measured temperatures of several degrees and aquaculture users would need to consider the implications of using the modelled temperatures. As for salinity, the model does not include localized freshwater inputs, so the model overestimated salinity for locations close to shore and was not able to represent more brackish water conditions in shallower depths. It was not possible to draw a general conclusion as to whether the model was suitable for aquaculture purposes, as the similarities and differences between the modelled and measured values varied by variable, area, depth, and time. These findings made it clear that aquaculture users would have to implement a process to determine whether they could use climate model outputs for their specific purpose. A model vetting framework is presented that can be used to support decisions on the use of climate model projections for aquaculture purposes. The vetting framework describes four stages that can be used to establish the necessary context regarding the aquaculture requirements and model capabilities, and then check how the model is simulating the conditions of interest at farm sites. Although the focus was aquaculture, the findings are relevant for other sectors and the framework can guide use of climate models for more local-scale assessment and management in coastal locations.

show abstract

Section: Challenges In Comparing Modelled Projections and Observationsmentioning

confidence: 99%

Section: Challenges In Comparing Modelled Projections and Observationsmentioning

confidence: 99%

Section: Comparison Between Model Outputs and In-situ Measurementsmentioning

confidence: 99%

See 1 more Smart Citation

Context matters when using climate model projections for aquaculture

Falconer,

Ytteborg,

Goris

et al. 2023

Front. Mar. Sci.

View full text Add to dashboard Cite

show abstract

“…Recent studies show that high-frequency phone surveys per single user are extremely useful for limiting the bias in the collection of agricultural data. An example is with regard to labor inputs or the harvesting of continuous crops such as cassava, for which the use of long recalls is highly inaccurate [51][52][53]; or for water quality measurement in the case of aquaculture management [54]; or for plot size and productivity [55][56][57][58]; or for enforcing labor contracts [59]. The recent integration of active artificial intelligence (AI) is further facilitating the interpretation of land use data and is making it more commonplace to access data that are no more than a few days old.…”

mentioning

confidence: 99%

“…1 Kenya (32); Lebanon(43); Madagascar(7); Malawi(7,49); Mozambique(53); Nepal (55); Peru (3); Senegal(7); Sierra Leone (42); South Africa(18); Tanzania(7); Togo(7). 2 Afghanistan (40); Bangladesh(19); Ethiopia(40); Ghana(24,31,54); Mozambique(40); Peru (3); Sierra Leone (42); Uganda(19); Zimbabwe(40). 3 Liberia (17); Peru (3) 4.…”

mentioning

confidence: 99%

Benefits and Challenges of Making Data More Agile: A Review of Recent Key Approaches in Agriculture

Serfilippi¹,

Giovannucci²,

Ameyaw

et al. 2022

Sustainability

View full text Add to dashboard Cite

Having reliable and timely or ongoing field data from development projects or supply chains is a perennial challenge for decision makers. This is especially true for those operating in rural areas where traditional data gathering and analysis approaches are costly and difficult to operate while typically requiring so much time that their findings are useful mostly as learning after the fact. A series of innovations that we refer to as Agile Data are opening new frontiers of timeliness, cost, and accuracy. They are leveraging a range of technological advances to do so. This paper explores the differences between traditional and agile approaches and offers insights into costs and benefits by drawing on recent field research in agriculture conducted by diverse institutions such as the World Bank (WB), World Food Program (WFP), United States Agency for International Development (USAID), and the Committee on Sustainability Assessment (COSA). The evidence collected in this paper about agile approaches—including those relying on internet and mobile-based data collection—contributes to define a contemporary dimension of data and analytics that can contribute to more optimal decision-making. Providing a theoretical, applied, and empirical foundation for the collection and use of Agile Data can offer a means to improve the management of development initiatives and deliver new value, as participants or beneficiaries are better informed and can better respond to a fast-changing world.

show abstract