This decision is based on the goal of minimizing loss of fish welfare and end product quality; aborting an initiated operation is certain to incur an extra load on the fish while the benefit of a quicker response is uncertain. In addition, the vessels may need to recommission before arriving at the emergency location. Recommissioning will depend on organizational resilience and ability to repurpose assets for operations they were not designed for . This may cover change of crew, picking up equipment, supplies, disinfecting the vessel or the likes. Supplementing the emergency response capability with DERVs on stand-by means that there are vessels that are available to respond to emergencies immediately. However, their emergency response contributions still depend on their positions relative to the emergency location and the impact of bad weather conditions. Examples of emergency types for sea-based fish farming and relevant emergency responses are presented in Table 1. The time frame parameter indicates a rough generalization of how long a situation can be sustained before significant fish welfare consequences are experienced, and amount gives an indication of the possible scope of consequences. Fig. 1 shows the development of three example emergencies as the amount of lost fish as a function of time. The shape and steepness of such development functions in relation to the progress of the emergency response determines the amount of lost fish during an emergency. The model evaluates the emergency response of the sea-based fish farming system at regular intervals, Δt RI, over a given period [t0, t0 +T], as presented in Fig. 2.
Emergency response capabilities change as the state of the fish farming system changes with time; therefore, livestock fodder system the first step of the method makes a prediction of how the fish farming system develops during normal operation based on the input for the initial state, task schedules and weather covering the period. Emergency response is thereafter simulated, and three emergency response measures are recorded at the different testing times, also referred to as response initiation times, e.g., t RI 1 in Fig. 2. The first measure is the first response time, defined as the time it takes from response initiation until the first vessel has commissioned and arrived at the emergency fish farm. The second is the response progress, which covers what response activities that are performed and when, for example the times and amounts for when fish is transported away from the emergency fish farm. Finally, the third is the response completion duration, defined as the time from response initiation until the emergency is over, for example when the last fish is pumped up from the emergency fish farm. Both the simulation of the normal operations in the fish farming system and the emergency response simulation in Fig. 2 are discrete event simulations where the system state changes at discrete points in time . A system state can be illustrated as a snapshot of the system, for example, including the position and status of each vessel and the weather conditions at that point in time, so that the development of a system over time can be described by a series of such snapshots. However, because the simulations are event driven, the system state changes do not occur at regular intervals. The system state is constant for the whole period between two system state changes, e.g., between the event at t2 and t3 in Fig. 2. Changes in the system state happens every time a vessel commences or ends a given operation or changes geographical position with more than one nautical mile. Any change in the initial sea-based fish farming system, including changes to the task schedule or the weather time series, will result in a different list of predicted system states. Uncertainty in the evaluation of the emergency preparedness of the system is reduced by applying several sets of historical data for the task schedules and hind cast weather time series.
The emergency response simulation is run once for each simulated emergency event, logging all details of the response. An emergency event is partly defined by the time at which it occurs, thus two identical emergencies occurring at different times are two different emergency events. Hence, every emergency event must be matched with the correct predicted system state for each emergency response simulation. Understanding emergency preparedness is crucial both to ensure good fish welfare and a sound operational practice in sea-based fish farming. The insight gained from model-based simulations enables the stakeholders to quantitatively assess their ability to effectively handle the various situations that might arise, and how to prepare for such situations. Based on the results of the case study, the method can be used to evaluate both the responses to individual emergencies and the general emergency preparedness level of a fish farming system. It can be used to indicate how well a basic operational system is set up for emergency response, and the improvement in emergency response capabilities from having additional emergency response resources. In Table 4, we see that the effect of having a DERV is more significant for the smaller system, which is expected as the relative capacity of an extra vessel is higher than in the larger system, and the emergency does not scale with the system size. Whether the first response times, response progress or response completion durations advocate for additional resources or other measures must however be seen in relation to specific emergency events and their required response times and statuses. A cost-benefit analysis of possible emergency response measures, for instance adding a DERV, would be one way to make such evaluations. However, formulating a cost benefit analysis is not straight forward due to both the cost and benefit side being highly dependent on, e.g., the system boundaries and to what degree a vessel is going to be dedicated. Testing for two different system sizes is of interest because regulations can divide fish farms into geographical areas, e.g., in the case of Norway where there are defined production areas. Biosecurity restrictions related to crossing the production area borders can be both costly and time consuming. This means that response vessels, to a large extent, can be assumed not cross production area borders within the time span of an emergency response situation.
Given quick response initiation the emergency response of most of the tested cases could be characterized as acceptable, based on the time frames of Table 1. For both weather scenarios and system sizes, the longest response completion durations for emergencies up to 3200 tons were in the order of two days. However, for the 12 800 tons emergencies, response completion durations were found to be as high as a week. The case results could be regarded as optimistic bounds as the response strategy made all vessels respond to the emergency event. Also, the results are based on predictions of the vessel activities, i.e., the mission schedules. New missions may suddenly arise, and the weather forecasts are not certain. The further into the future the evaluations go, the more uncertain are the predictions. However, the assumption that commenced operations may not be aborted prematurely might make the vessels less responsive than they are in reality. In a real-life scenario, two conditions are likely to delay the emergency response, making the response times longer than shown in the results. First, the hazard must be identified, and then the appropriate decision makers in the companies must decide to implement response actions. Early detection of HABs is not easy as the identification of the algae type and concentration usually is done by taking water samples and sending them to laboratories for analysis.Systems for early detection based on satellite imaging of algal concentrations, artificial intelligence identification of algae types, and monitoring of the potential for algal blooms are being developed. Potential for algal blooms is evaluated based on secondary indicators such as water temperature, oxygen levels and the level of blue-green algae. After a threat or unwanted event has been identified emergency response resources are not deployed until the appropriate decision makers give the order. In situations like severe HABs, the potential large scale of the required emergency response means that the response is costly and is likely to negatively affect other parts of the business, e.g., occupying company resources that are needed in normal operation. This means that a thorough assessment of the situation must be made before initiating a full emergency response, and action may not be deemed beneficial until the emergency has escalated. Considering the two delaying factors in real-life situations, response time could probably be improved if DERVs were positioned according to real-time assessments of harm potential and the probability of an emergency. Such a problem would resemble the maximal covering problem addressed in Probability of emergency could, e.g., be based on the degree to which environmental conditions favor a HAB, as proposed in .
Analyses of emergency response performance can be useful in understanding and quantifying risk . Enabling operators to show insurers that they reduce the consequences of adverse events can also provide benefits for both parties. Stakeholders should be aware that the method is not meant to give exact information far into the future,fodder system trays rather it is meant to indicate the emergency preparedness level of a sea-based fish farming system. Therefore, a sufficient number of evaluations should be performed, with different input data, so that they trust the results and the value of the information in the results. However, this depends on what the interests of the stakeholders are and what they want to study. If testing for general preparedness, then the uncertainty of task schedules and weather forecasts is less of a problem since hind cast data can be used. If they want to perform what-if analyses on specific emergencies, the evaluation period should not be stretched too far. The current model of agricultural intensification, based on agrochemical inputs, large monocultures and landscape homogenisation, has successfully increased yields, but is associated with severe losses of biodiversity and ecosystem services, even in neighbouring nature reserves. Current trends can only be reversed by a concerted effort to fundamentally redesign farming systems and agricultural landscapes; that is, a paradigm shift in agriculture. Certified organic farming, that is, banning synthetic agrochemicals to achieve sustainability in agricultural systems in general and biodiversity conservation in particular, is often claimed to be the fundamental alternative to conventional farming. However, the contribution of certified organic agriculture to stop the losses in biodiversity appears to be exaggerated in the public perception. In fact, switching from conventional to organic practices increases local species richness by just a third, but leads to considerable yield losses, so that more land is needed to produce the same amount of food. Surprisingly, a wealth of biodiversity friendly measures that can enhance biodiversity and can be implemented in conventional agriculture, have so far been poorly adopted in current agricultural systems. Here, we challenge the widespread appraisal that organic farming is the fundamental alternative to conventional farming for promoting or restoring biodiversity in agricultural landscapes. After considering measures essential for biodiversity-friendly farming, we propose more effective solutions towards biodiversity friendly landscapes and ways to integrate local and landscape scales in existing organic and conventional farming systems as well as in agricultural policies.Certified organic farming can enhance biodiversity when compared to conventional farming. On average, organic farming across the world’s crops increases local species richness by ~34% and abundance by ~50%, with plants and bees benefitting most and other arthropods and birds to a smaller degree. Benefits also vary with crop type and landscape context. Organic farming strives for environmental benefits, sustaining soil fertility and biodiversity, and prohibits synthetic fertilisers, synthetic pesticides, and genetically modified organisms. In particular, the replacement of herbicides by mechanical weeding is important for biodiversity conservation, because higher weed cover benefits many organisms. Practices such as crop diversification, small fields, green manure, low fertiliser input, and restoration of natural landscape elements are often recommended by organic food organisations and can be more prevalent on organic than conventional farms,but they are not formal part of certification regulations.