Transdisciplinary approaches to local sustainability: aligning local governance and navigating spillovers with global action towards the Sustainable Development Goals

In an evolving world, effectively managing human–natural systems under uncertainty becomes paramount, particularly when targeting the United Nations 2030 Agenda for Sustainable Development Goals (SDGs). The complexity in multi-actor decision-making and multi-sectoral settings, coupled with intricate relationships and potential conflicting management approaches, makes understanding the local implications of progressing towards the global SDGs challenging. We used a transdisciplinary approach for knowledge co-production with local stakeholders to assess the impact of local action to boost sustainability in the Goulburn–Murray region, Victoria, Australia, and its alignment with global action towards the SDGs. Together, we co-developed 11 local actions geared towards achieving four locally important environmental and socioeconomic SDGs, with a particular emphasis on addressing potential ‘spillovers’—unintended effects that influence SDGs across scales. Through system dynamics modelling, we evaluated the interplay between these local actions and global scenarios, emphasising their synergies, trade-offs, and the resulting impact on SDG indicators. Key findings indicate a predominant synergy between global and local actions across most SDG indicators. However, certain areas like dairy production, riverine algal blooms, and agricultural profit displayed trade-offs. Local actions significantly impacted indicators, such as crop production, dairy output, agricultural land use, and agricultural profitability. Findings highlighted the need for complementary actions in areas, such as water availability management, skilled workforce, and salinity control. This study underscored the importance of harmonising local initiatives with global sustainability objectives and can inspire local governance to champion resilience policies that harmoniously integrate local actions with global sustainability goals, adapting to evolving uncertainty scenarios.


Introduction
Successful realisation of the United Nations 2030 Agenda for SDGs hinges on local initiatives (UN 2015).However, complexity exists in understanding the local implications of global SDG achievement (Sachs et al. 2019).Every nation and local region shapes the advancement or impediment of the 2030 Agenda's success through competing development agendas (Messerli et al. 2019).Local scale implementation of the SDGs, championed by local communities, cities, and organisations, provides a tailored approach that aligns with the distinct characteristics of each region (Moallemi et al. 2019).However, progressing on one SDG can generate 'spillovers'-unintended effects that in uence the same or other SDGs across scales (Engström et al. 2021; SDG Watch Europe 2019).There is an urgent need for research that effectively governs for spillovers and empowers local regions to consider global implications when implementing local SDG policies (Engström et al. 2021).
Sustainability science uses scenario analysis to address future uncertainties and craft policies for sustainable development (Swart et al. 2004).Several recent studies have undertaken scenario analysis for sustainability at various scales ( Gao and Bryan (2017) utilised an integrated land system model to evaluate numerous scenarios aiming to assess multiple SDG targets.However, as Moallemi et al. (2020) underscored, diverse local priorities can sometimes clash with overarching policies set at national or global levels.To effectively achieve the aspirations set forth in the SDGs, there must be a harmonious integration of local actions with global sustainability objectives, while fostering coordination among all stakeholders to ensure a sustainable future for all.
Policymakers may encounter challenges in crafting evidence-based sustainability policy, often due to a lack of detailed information about potential uncertainties, requiring careful consideration and analysis (Chappin et al. 2017;Forni et al. 2016).The intricacies and uncertainties of the SDGs can bene t from the application of systems models, commonly known as integrated assessment models (IAMs) ( However, relying solely on integrated assessment models proves inadequate to guide us through the complexity of the system (Koasidis et al. 2023) within intricate societal, economic, and environmental frameworks characterised by a range of stakeholders with diverse needs, values, and priorities (Messerli et al. 2019).A transdisciplinary approach (Game et al. 2018) is required to enable the development of local sustainable pathways while elevating the salience and legitimacy of the information derived from integrated assessment models (Cash et al. 2003).The 2030 Agenda implementation requires engaging all stakeholders (Independent Group of Scientists appointed by the Secretary-General 2019).Engaging both scientists and non-scienti c knowledge in transdisciplinary collaboration (Michas et al. 2020), co-design (Moser 2016), and co-production (Norström et al. 2020) represents a transformative strategy for merging science and policy in order to ful ll humanity's collective aspiration for sustainable development.The aim of these concepts is gravitating towards knowledge co-production as a promising strategy to advance in the intricate challenges arising in the eld of sustainability while also exerting a meaningful in uence on the decision-making processes of policymakers (Moallemi et al. 2021a).Norström et al. (2020) highlighted that achieving high-quality knowledge co-production for sustainability requires being "local, pluralistic, goal-oriented, and interactive".A discernible void persists in the realm of local sustainability, pertaining to the realisation of such a transdisciplinary approach of knowledge co-production.
In this study, we pioneered a unique transdisciplinary approach to unravel the spillovers of local sustainability actions within broader global scenario frameworks.Utilising a local IAM based on system dynamics, the Local Environmental and Socioeconomic Model (LESEM) (Bandari et al. 2023), we focused on the Goulburn-Murray region in Victoria, Australia.Our approach was distinctive in its rigorous integration of knowledge co-production principles at each step, ensuring our research was deeply embedded in local sustainability contexts.We codeveloped 11 local actions towards achieving four locally relevant environmental and socioeconomic SDGs.Furthermore, we explored the interplay of four combinations of global scenarios, including SSP and RCP scenarios, global diet, and food waste, with these local actions.Ultimately, our goal is to inspire local governance to embrace both local and global sustainability goals, championing policies that align with the evolving scenarios and consistently support both local and global sustainability objectives.

Study area
The Goulburn-Murray region is situated adjacent to the River Murray in northern Victoria, Australia.This region represents the country's largest irrigation district, encompassing six local government areas-Moira, Greater Shepparton, Loddon, Campaspe, Gannawarra, and Swan Hill (Fig. 1).This region is recognised as Australia's primary food production hub, characterised by extensive cropping, livestock production, and horticulture (GMIDWL 2018).Nonetheless, the agricultural sector and economy in the Goulburn-Murray region have endured signi cant impacts from recent economic, policy, and environmental changes, such as climate change, declining commodity prices, water reform policies, uctuating water prices, droughts, variations in water availability, and unpredictable international agriculture commodity markets (RPG 2020).The region serves as a valuable case study for investigating the complex interplay of diverse future sustainability actions and scenarios at both local and global scale.

Methodological approach
Our transdisciplinary approach to the identi cation and assessment of local actions under global scenarios followed four key steps, as shown in Fig. 2. In Step 1, we co-designed and employed a local IAM (i.e., LESEM) to analyse sustainable development across local priority SDGs.In Step 2, we co-produced and quanti ed local actions through a participatory approach with local stakeholders in the Goulburn-Murray region.We de ned the assumptions for global scenarios (Step 3).We elaborated our scenario assumptions including both global scenarios and local actions and utilised these assumptions in the model.In Step 4, we utilised LESEM to simulate and capture the interplay of implementing the local actions under the global scenario frameworks across multiple SDGs.

Exploring the multisectoral dynamics underlying the SDGs through modelling
We employed the LESEM, developed via a participatory approach, with a speci c focus on four highest priority SDGs in the Goulburn-Murray region: Agriculture (SDG 2), Water Availability (SDG 6), Economic Growth (SDG 8), and Life on Land (SDG 15).These SDGs address socioeconomic development outcomes and environmental impacts.Using LESEM, we modelled priority SDG indicators across four SDGs.
The selection of SDGs and their indicators was guided by conducting a comprehensive contextual analysis, which included interviews and workshop with local stakeholders, examination of scienti c papers and reports, and review of policy documents (Bandari et al. 2022).The objective was to identify the socioeconomic and environmental issues that held signi cant importance to the local stakeholders in relation to the SDGs and assess their interactions.The LESEM utilises the system dynamics approach (Sterman 2002) and operates at a local level with an annual timeframe, covering the mid-term to long-term period from 2010 to 2050.The LESEM consists of seven main sub-models: Demographics, Agriculture, Water Availability, Land Use, Economy, Fertiliser Use, and Water Quality.Further details about the modelling can be found in Bandari et al. (2023).These sub-models capture the essential characteristics and issues of the study area, identi ed through the knowledge co-production process and contextual analysis.We applied LESEM to simulate and analyse the effects of various local actions under global scenarios to address local concerns regarding water availability, blue-green algal blooms, salinity, land-use change, skilled workforce in agricultural sector, agricultural commodities production, and local economy.

Identifying and parameterising local actions
In response to collective sustainability challenges in the Goulburn-Murray region, we have collaboratively engaged with local stakeholders in a comprehensive knowledge co-production approach.This approach encompassed all steps of the research process, including identifying local challenges and opportunities, co-designing a system dynamics model, devising local actions aimed at advancing four priority SDGs, and analysing results.We conducted three face-to-face workshops with stakeholders (Fig. 3) where we actively engaged with local experts to identify and understand speci c challenges and opportunities.By involving stakeholders directly, we aimed to co-create local actions that are most relevant and appropriate for the Goulburn-Murray region.The initial workshop, conducted in March 2022, focused on gaining a comprehensive understanding of the system and its dynamics.During this session, we engaged stakeholders to gather valuable insights and knowledge about the local context, challenges, and opportunities related to sustainability.The initial workshop involved 18 local stakeholders representing organisations such as Murray Dairy, Goulburn Valley Water, the Department of Energy, Environment and Climate Action (DEECA), Agriculture Victoria, Goulburn-Murray Water, Regional Development Victoria, and Goulburn Broken Catchment Management Authority (GBCMA).
In July 2022, Deakin University ran a workshop with a subcommittee of the Goulburn-Murray Resilience Taskforce to develop in uential local actions and identify global driving forces that could impact the local sustainability in the future.The subcommittee included ten local stakeholders from organisations such as Goulburn Broken Catchment Management Authority (GBCMA), North Central Catchment Management Authority, Regional Development Victoria, Goulburn-Murray Water, Goulburn-Murray region action working group, Goulburn Valley Community Energy, and RM Consulting Group (RMCG).A group of local sustainability actions (Table 1.List of local actions and their assumptions) were identi ed through stakeholder engagement to address four priority SDGs: SDG2 (agricultural activities), SDG6 (clean water and sanitation), SDG8 (decent work and economic growth), and SDG15 (life on land).These actions were designed to address speci c challenges and opportunities within the local context and contribute to achieving the priority SDGs.To further re ne these actions, we solicited input from workshop participants regarding the types of actions to be tested, the system parameters they would potentially in uence, and the extent of their impact.We speci ed 11 local actions and assessed their effects on eight SDG indicators.Note that there were implicit assumptions in what is being tested -e.g., increase water-use e ciency by 10% (Table 1.List of local actions and their assumptions).The participants engaged in a discussion about the potential scenarios in which the actions might not achieve the desired results.They also determined which actions would be most effective in advancing the aspirations of the region in the short and long term.
In September 2023, Deakin University hosted a hybrid workshop in collaboration with RM Consultancy Group -both in-person and onlineas the concluding session of this project.The workshop was attended by 21 individuals representing diverse organisations and community members.The event was focused on presenting the ndings from the modelling phase and discussing about possible futures for the Goulburn-Murray region in 2050.These collaborative approaches aimed to stimulate the region to take resilience actions in preparing for and adapting to diverse potential futures.

Local action
Local action assumption Action 1: Improve agricultural productivity Achieve an annual 2% increase in agricultural productivity.
Action 2: Increase water-use e ciency on-farm Achieve total increase of 10% in water-use e ciency on-farms by the year 2050 (0.25% per year).
Action 3: Increase farm water reuse Achieve a total increase of 15% in farm water reuse on-farm by the year 2050 (0.375% per year).
Action 4: Reduce conveyance water loss Reduce conveyance water loss by 5% by the year 2050 (0.375% per year).
Action 5: Reduce input costs Reduce agricultural input costs by 10% via the use of modern technology and promotion of sustainable practices (0.25% per year).
Action 6: Premium produce growth Achieve a total increase of 10% in agricultural commodity prices by the year 2050 (0.25% per year) through producing higher quality or premium produce which commands a higher price.
Action 7: Increase migration rate Increase migration rate by 20% by the year 2050 (0.5% per year).
Action 8: Increase birth rate Increase the birth rate by 10% by the year 2050 (0.25% per year).
Action 9: Reduce fertiliser application Reduce fertiliser application by 20% by the year 2050 (0.5% per year).
Action 10: Reduce nitrogen and phosphorous generated from animal manure Achieve a total decrease of 20% in animal manure and N & P potential loads from dairy shed e uent by the year 2050 (0.5% per year).
Action 11: Reduce nitrogen and phosphorous runoff fraction Achieve a total decrease of 20% in nitrogen and phosphorous loss by the year 2050 (0.5% per year).

Global scenario assumptions
Our approach involves using both global scenarios and local actions to examine different projections of output variables from the perspective of local sustainability and assess their sustainability performance.We characterised global scenarios using two combinations We characterised two global scenarios: Global BAU and Global ACTION and these incorporate assumptions that play a crucial role in shaping future socio-economic and environmental outcomes for the study area (Table 2).We speci ed the combination of SSP2 and RCP4.5 for the Global BAU scenario and SSP1/RCP2.6 for Global ACTION.SSP1 and SSP2 were used to estimate food demand based on the gross domestic product and population.Temperature and rainfall data were downloaded for RCP2.6 and RCP4.5 using 8 general circulation model estimates from the WorldClim CMIP6 database and these were average to provide ensemble layers from 2010 to 2050.These layers were then used to calculate potential evapotranspiration using the Priestley-Taylor method ( Food demand is also a key driver in uencing the amount and type of agricultural production in the study area.Food demand consisted of both a diet component (Willett et al. 2019) and a waste component (FAO 2011) and time series demand multipliers were created under the SSP1 and SSP2 scenarios.The total production of agricultural commodities required from the study was calculated using an unpublished panel data timeseries econometric and gravity model which regresses historical GDP, population and trade relationships with Australian agricultural production.Future Australian production of crops and livestock (including crops used as livestock feed) was projected using this model based on projections of GDP and population under SSP1 and SSP2.Future demand for agricultural production from the study area was then downscaled from this based on the total historical Goulburn-Murray regional production as a proportion of total Australian agricultural production for each commodity.We considered two distinct types of diets: a BAU diet and a Flexitarian diet (FLX), and these change the relative amounts of agricultural commodities demanded.The BAU diet represents the continuation of trends in recent historical conventional dietary preferences.The Flexitarian diet involves predominantly plant-based eating patterns with the exibility to incorporate animal products in moderation.Finally, we assumed two scenarios of food loss and waste as BAU waste and Halve waste.Reducing the amount of waste under Halve waste means that less total production was required to meet demand (as less is wasted).• BAU diet

• BAU waste
Food demand under the scenario SSP2-BAU Diet-BAU waste is increasing for all agricultural commodities, including crops, dairy, beef, and sheep by the year 2050 (Table S1).
Water yield RCP4.5 According to the projections, the average water yield under BAU scenario (RCP4.5) is projected to decrease by 0.2% annually (Table S5).
Livestock & crop yield RCP4.5 Under the BAU scenario, agricultural commodity yields are projected to increase for all agricultural commodities (UNCCD 2017), including both irrigated and dryland commodities by 2050 (Table S3).
Global ACTION Food demand • SSP1 • FLX diet

• Halve waste
In the context of SSP1-FLX Diet-Halve waste, the demand for FLX products is increasing across various agricultural commodities, except for winter oilseed, summer oilseed, hay (predominantly utilised for non-human consumption), dairy, and beef (Table S2).Notably, the rate of increase in food demand for FLX products in this scenario surpasses that of the SSP2-BAU Diet-BAU waste scenario.In the context of Australia's scenarios, SSP1 exhibits higher population (36.59  Based on the projections, the average water yield under scenario RCP2.6 is expected to experience an annual decrease of 0.2% (Table S5).

Livestock & crop yield
RCP2.6 By 2050, under the RCP2.6 scenario, agricultural commodity yields are predicted to increase for all irrigated commodities and most dryland agricultural commodities, with the exception of dryland beef, dairy, stone fruit, and hay (Table S4).On the other hand, for most agricultural commodities, livestock and crop yields under scenario RCP2.6 are lower compared to scenario RCP4.5, except for irrigated grapes, dryland stone fruit, dryland non-cereal crops, dryland winter oilseeds, and dryland winter cereals.These speci c commodities show higher yields in scenario RCP2.6 compared to RCP4.5.In the RCP2.6 (low emission scenario), temperatures are projected to increase by approximately 1.5 to 2 degrees, while under RCP4.5 (medium emission scenario), temperatures are expected to rise around 2.5 to 3 degrees.

Local and global scenario combinations
We combined the Local BAU (Table 3) and Local ACTION scenarios (Table 1) with the Global BAU and Global ACTION scenarios ( By incorporating these four scenarios into our analysis, we aimed to capture a broad range of potential projections and better understand the implications of different future local and global scenarios.S8.

Agricultural commodity yield
The agricultural commodity yields, measured in head/ha for livestock and tonnes/ha for crops, was calculated for average of year 2005, 2010, and 2015, as detailed in Table S9.

Water requirement
The water requirement of agricultural commodities (Million litres/ha or Million litres/head per year) for average of year 2005, 2010, and 2015, as detailed in Table S10.

Surface farm water reuse rate
Based on historical data from 2015 to 2019, an average surface farm water reuse rate of 0.12 of the total surface water usage across all users was utilised (VSG 2019).

Conveyance water loss
An average conveyance water loss rate of 0.1 of the total surface water storage calculated based on historical data from 2015 to 2019 (VSG 2019).

Migration rate
Based on primary data sourced from the Australian Bureau of Statistics census data, the mean migration rate from 2010 to 2020 stood at 0.00352, representing a portion of the total population in each age cohort (ABS 2022).
Birth rate Based on primary data sourced from the Australian Bureau of Statistics census data, the mean birth rate is 0.043 from 2010 to 2020 (ABS 2022).

Commodity prices
The commodities in the region were classi ed into four primary groups, namely crops, sheep, beef, and dairy.We further categorised the agricultural commodities into dryland and irrigated commodities.The irrigated crops comprised 15 crops, including winter cereals, rice, winter legumes, summer legumes, winter oilseeds, summer oilseeds, hay, other noncereal crops, pears, apples, stone fruit, tropical stone fruit, grapes, vegetables, and citrus.The dryland crops comprised seven crops, including winter cereals, winter legumes, winter oilseeds, hay, other noncereal crops, stone fruit, and grapes.The commodity prices for base year 2010 is available in Table S6.

Nitrogen and phosphorus fertiliser application
Nitrogen and phosphorus fertiliser application (kg/ha) for base year 2010 as detailed in Table S11 (Navarro & Marcos Martinez 2021).

Nitrogen and phosphorus generated with manure per cow
Nitrogen generated per cow is 70 kg per year and phosphorus generated per cow per year is 9 kg per year (GBWQWG 1995).

Total phosphorus and nitrogen generated with manure per sheep
Nitrogen generated per sheep is 10 kg per year and phosphorus generated per sheep per year is 1.5 kg per year (GBWQWG 1995).

Phosphorus and nitrogen runoff fraction in irrigated and dryland areas
Data was gathered pertaining to the runoff fraction from both irrigated and dry land areas, with a particular focus on phosphorus and nitrogen runoff nding its way into rivers and lakes.Based on extensive long-term monitoring data acquired from drained catchments, it has been established that the runoff factor stands at 20% in irrigated areas and 7.5% in dryland areas (GBWQWG 1995).

Scenario analysis
After implementing our scenario assumptions (Table 1.List of local actions and their assumptions, Error!Reference source not found.,Table 3, Fig. 4) as speci c model parameters, we ran the LESEM for the four scenarios.Our simulations allowed us to evaluate and assess the scenarios based on a comprehensive set of socioeconomic and environmental indicators over time.In our selection process, we identi ed a total of eight SDG indicators to assess progress towards various sustainability goals (Table 4).In relation to SDG2, the emphasis was on two indicators that measured dairy and crop production.For SDG6, two indicators were considered, re ecting different dimensions of water availability and river water salinity.The SDG8 was examined through two indicators related to economic growth and skilled workforce.Lastly, SDG15 was represented by two indicators capturing aspects of environmental conservation through changes in agricultural land and blue-green algal bloom.By selecting these indicators across multiple SDGs, we aimed to gain a comprehensive understanding of progress and challenges in key areas of sustainable development.

Scenario realisations
Using the LESEM model for scenario quanti cation resulted in internally consistent outcomes across seven sub-models and enabled the evaluation of progress towards eight key SDG indicators across four priority SDGs (Fig. 5).The crop production results, as an indicator of SDG2, reveal that in the year 2050, the amount of crops produced in the  S5.
In the Global BAU | Local BAU scenario, the value of river water salinity as another indicator of SDG6 increased from 47 µS/cm in 2010 to 66 µS/cm by the year 2050.The primary reasons for this increase are twofold: rst, there is a reduction in water yield under both RCP4.5 and RCP2.6, and second, there is an increasing demand for food across all four scenarios, leading to a rise in agricultural practices.

Local action sensitivity analysis
The result of local action sensitivity analysis demonstrated that the most impactful action was agricultural productivity, and this action affected ve SDG indicators: agricultural land, agricultural pro t, total crop production, total dairy production, and blue-green algal bloom (Fig. 6).Among all SDG indicators, agricultural pro t (SDG8) exhibited the highest increase, with a growth of 46% and 35% by 2050 and 2030, respectively.The projections revealed that increasing agricultural productivity was associated with a decrease in agricultural land (SDG15) by 33% by 2050 and 17% by 2030.Additionally, the action of agricultural productivity had a direct impact on the indicators of total crop production and total dairy production.By 2050, the projections showed an increase of approximately 34% in total crop production and 22% in total dairy production (SDG2) as a result of this action.Similarly, by 2030, there was an estimated increase of approximately 25% in total crop production and 16% in total dairy production.The second most impactful action, Action 6 (commodities price), showed a notable impact on agricultural pro t (SDG8) with an increase of approximately 13% by 2050 and 7% by 2030.
Action 2, which focused on enhancing water use-e ciency on-farm, had impacts on multiple SDGs, including total crops and dairy production (SDG2), agricultural land (SDG15), agricultural pro t (SDG8), and blue-green algal bloom (SDG15).Our analysis revealed that this action led to a reduction of approximately 7% in agricultural land (SDG15) by 2050 and around 2% by 2030.Actions 7 and 8 had only a slight impact on most of the SDG indicators assessed, indicated their limited in uence on the selected indicators.However, these actions exhibited a stronger impact on the skilled workforce indicator (SDG8).Actions 9, 10, and 11 had impacts on the blue-green algal bloom indicator (SDG15).On the other hand, actions 3 and 4 in uenced several SDG indicators, including total crops and dairy production (SDG2), net water availability (SDG6), river water salinity (SDG6), agricultural water use (SDG6), agricultural land (SDG15), agricultural pro t (SDG8), and blue-green algal bloom (SDG15).Furthermore, the sensitivity analysis was conducted on SDG indicators under the Global ACTION | Local ACTION scenario compared to those under the Global BAU | Local BAU scenario (Figure S1).

Discussion
By assessing both local actions and global scenarios with a system dynamics model (i.e., LESEM as an IAM tool), was able to explore a wide range of possible future trajectories and identify key drivers of change.We adopted a transdisciplinary methodology, partnering with local stakeholders, to co-produce knowledge aimed at realising sustainable results.We introduced potential future scenarios to enlighten local policymakers about the implications of various combinations of local and global scenarios pertaining to four key SDGs.Each scale of action serves unique purposes, and in speci c SDGs, they share common goals in addressing the diverse array of challenges faced in the pursuit of sustainability.We sought to address a research void by employing a transdisciplinary approach (Moallemi et

Limitations and future work
We utilised the system dynamics model, LESEM, as a simpli ed representation of the real-world system, speci cally within the Goulburn-Murray context.Although LESEM proved to be a valuable tool for policymaking, further development is needed to encompass a broader range of indicators across all SDGs (Allen 2019).For example, adding energy or circular economy which was suggested by local stakeholders and establishing additional multisectoral connections.Parameterisation remains a central challenge in all modelling processes, a task complicated further when addressing systems with societal components (Verburg et al. 2016).In modelling process, we faced a limitation concerning the model parameterisation and also availability of comprehensive data.We scrutinised the signi cant uncertainties surrounding various speci ed model parameters (Bandari et al. 2023).However, we recognised that our study does not encompass all types of uncertainties, particularly those extreme forms characterised by unknown circumstances or a state of complete ignorance, which are inherently unrepresentable in models (Stirling 2010).As a recommendation for the future, we propose testing other global scenarios, such as SSP3, SSP4, and SSP5, to gain a comprehensive understanding of their potential impacts on the sustainability goals.Expanding the range of SDG indicators under each of the four priority SDGs offers a more comprehensive view of progress, potential trade-offs, and allows us to assess interconnections and synergies in sustainable development.Furthermore, we suggest the development of more context-speci c and localised actions.

Conclusion
We used the LESEM system dynamics model to explore potential future trajectories concerning sustainable development in the Goulburn-Murray region of Victoria, Australia.Through the integration of local initiatives and global outlooks, we pinpointed pivotal drivers pertaining to four SDGs, including Agriculture (SDG 2), Water Availability (SDG 6), Economic Growth (SDG 8), and Life on Land (SDG 15).Collaborating closely with local stakeholders via a knowledge co-production approach, we jointly crafted strategies to achieve these SDGs.Our ndings underscored the effective in uence of local actions such as agricultural productivity and water use e ciency on priority SDG indicators, simultaneously emphasising the necessity of other complementary synergistic actions, particularly in areas like water resource management, water salinity control, and blue-green algal bloom mitigation.The study's results highlighted the synergies and trade-offs between local and global actions in the pursuit of sustainability.Our results highlighted a strong synergy between global and local initiatives for the majority of SDG indicators.However, speci c sectors such as dairy production, riverine algal blooms, and agricultural pro tability show evident trade-offs.Our methodology, deeply rooted in an inclusive knowledge co-production with local stakeholders and bolstered the credibility of our data from the LESEM system dynamics model.By the results of modelling and future scenarios, we aimed to enhance community awareness and advocate for resilience policies that align with both local and global sustainability goals.Crucially, the data suggested the imperative of crafting policies that cater to both local and global requirements for a sustainable future.For instance, by amplifying agricultural productivity, implementing e cient water management strategies, and mitigating agricultural expenses, we can usher in advancements in food production, economic prosperity, and ecological balance, bene ts that have the potential to resonate on a global scale.

Declarations Figures
Land Hat eld-Dodds et al. 2015; Lamontagne et al. 2019).Moallemi et al. (2022) modelled global socioeconomic and climatic change scenarios and their uncertainty for sustainable development.At the national level, van Beek et al. 2020) to comprehend the interplay of synergies and trade-offs both among and within the SDGs (Neumann et al. 2018) and capturing the joint impacts of global uncertainties and local actions on local priorities.Previous studies have shown systems modelling (Babatunde et al. 2017; Greeven et al. 2016; Sterman et al. 2012; van Beek et al. 2020; Wiedmann 2009) to be effective in modelling intricate feedback interactions, exploring sector interconnections, grasping non-linearities and radical shifts (Moallemi et al. 2021b), with wide-ranging applications within the broader sustainability topics (Eker et al. 2019) or those closely tied to the SDGs (Allen 2019; Randers et al. 2019).

of
Shared Socioeconomic Pathways (SSPs) (O'Neill et al. 2017; Riahi et al. 2017) and Representative Concentration Pathways (RCPs) (van Vuuren et al. 2011).The SSPs and RCPs provided a comprehensive framework for understanding and analysing the potential future scenarios in terms of key socioeconomic and climate driving forces and for characterising global action on sustainability.The SSPs outline different trajectories for future socioeconomic development, offering ve pathways encompassing differing levels of climate change adaptation and mitigation, extending until 2100 (O'Neill et al. 2017).The RCPs depict various climate forcing levels that correspond to different potential futures.They offer insights into the potential climate outcomes based on different emissions scenarios and serve as a basis for understanding the range of possible climate futures (van Vuuren et al. 2011).

2. 8 .
Sensitivity analysis We conducted a sensitivity analysis using the LESEM to examine the percentage change in SDG indicators under Global BAU | Local ACTION scenario compared to SDG indicators under Global BAU | Local BAU.The reference point for comparison was the year 2010.This involved calculating the percentage change in each SDG indicator for the years 2030 and 2050 under the Global BAU.One-by-one we turned on each local action and compared SDG performance against the Local BAU scenario in percentage terms.The different sectors (i.e., SDGs) within the model have interconnected dynamics, which means that any action implemented in one sector will have ripple effects throughout the entire model.
Global BAU | Local BAU scenario (1331 kilotonnes [kt]).This increased to 1442 kt under the Global ACTION | Local BAU scenario, 1938 kt under the Global BAU | Local ACTION scenario, and 2091 kt under the Global ACTION | Local ACTION scenario.In the Global BAU | Local BAU scenario, dairy production, another indicator for SDG2, increased from 1480 ML in 2010 to 1830 ML by the year 2050.Under the Global ACTION | Local BAU scenario production was 1236 ML by 2050, which is lower than the BAU and showed a decreasing trend over time resulting from the substantive reduction in demand under a exitarian diet shift.Dairy production under the Global BAU | Local ACTION scenario of 2354 ML was highest of all scenarios illustrating the effects of strong demand for animal-based foods in the global diet and increasing productivity and resource-use e ciencies at the local scale.Under the Global ACTION | Local ACTION scenario dairy production by the year 2031, the implementation of local actions such as improving agricultural productivity had led to an increase in dairy production, reaching 1716 ML.However, beyond 2031, the impact of global actions and the shift towards a exitarian diet overshadowed the effect of local actions and led to a decline in dairy production 1560 ML by the year 2050.Under the Global BAU | Local BAU scenario, the net water availability as an indicator of SDG6 was projected to decrease from 4350 Gigalitres (GL) in 2010 to 3201 GL by 2050.The net water availability exhibited a marginal change under the Global BAU | Local ACTION scenario compared to the Global BAU | Local BAU scenario, with a recorded level of 3274 GL by 2050.Similarly, minor differences were observed between the scenarios of Global ACTION | Local ACTION, approximating 3258 GL, and Global ACTION | Local BAU, around 3186 GL.These variations were attributed to the water yields noted under RCP4.5 and RCP2.6, as outlined in Table use of the study area.The Goulburn-Murray region is speci ed with a black boundary.The inset map shows the region location in the context of the state of Victoria, Australia.

Figure 2 Overview
Figure 2

Figure 5 Projections
Figure 5 (Fischer et al. 2021ime-series temperature, rainfall, and potential evapotranspiration data layers were combined with a digital elevation model, solar radiation data, and soil data from the Soil and Landscape Grid of Australia to calculate average water yield time series spatial layers at 1km 2 grid cell resolution under the RCP4.5 and RCP2.6 scenario from 2010 to 2050 using the InVEST model(Sharp et al. 2018).The Global Agro-Ecological Zones (GAEZ) 4 model was applied to quantify the impact of climate change on agricultural productivity and generate the agricultural commodity yield multipliers under the RCP4.5 (BAU) and RCP2.6 scenario compared to the base year of 2010 for various crops, pasture, beef, sheep, and dairy from 2010 to 2050(Fischer et al. 2021).

Table 2 )
, specifying four combinations: Global BAU | Local BAU; Global BAU | Local ACTION; Global ACTION | Local BAU; and Global ACTION | Local ACTION (Fig.4).In the Global BAU scenario, the model adhered to the Global BAU settings (SSP2_RCP4.5_BAUdiet_BAU waste).The Global ACTION utilised the model parameters from the Global ACTION (SSP1_RCP2.6-FLXdiet -Halve waste).The scenario SSP1_RCP2.6_FLXdiet_Halve waste referred to the scenario in which we examined food demand under SSP1, the FLX diet, Halve waste, and livestock, crops, and water yield under RCP2.6.On the other hand, the Local ACTION scenario implements Local ACTION by applying each of the 11 local actions parameterised with our stakeholder group.Lastly, in the Local BAU scenario, the model parameters are set based on the BAU trends.

Table 3
The livestock productivity, including beef, sheep meat, wool (unit: tonnes/head), and dairy (unit: litres/head) for average of year 2005, 2010, and 2015, as detailed in Table

Table 4 :
List of SDGs and related indicators Notably, the scenarios of Global ACTION | Local ACTION and Global ACTION | Local BAU exhibited lower salinity levels compared to the scenarios of coupled with advancements in livestock and crop yield, played a pivotal role in driving the increase in agricultural commodities production.Furthermore, the high food demand was perpetuated by the continuation of BAU food waste without any global action under the global scenario of SSP2_RCP4.5_BAUDIET_BAU WASTE.In the scenario Global ACTION | Local ACTION, the projected agricultural pro t was approximately 3031$M by the year 2050, making it the second highest among the considered scenarios.The ndings further revealed that agricultural pro t under both scenario Global ACTION | Local BAU and scenario Global BAU | Local BAU was lower compared to the other scenarios, with values of 1564$M and 1875$M, respectively, by the year 2050.Under the Global BAU | Local BAU scenario, it was projected that the skilled workforce, a key indicator for SDG8, would escalate from 2784 people in 2010 to 3246 people by 2050.Additional projections indicate that the number of skilled people would reach 3354 under the scenarios of Global BAU | Local ACTION and Global ACTION | Local ACTION, whereas it is projected to be 3246 people under the Global ACTION | Local BAU scenario by the year 2050.The skilled workforce indicator remains unaffected by global scenarios, including Global BAU and Global ACTION.Under the Global BAU | Local BAU scenario, the agricultural land was projected to shrink from 794,000 ha in 2010 to 724,000 ha by 2050.The agricultural land projections showed a more substantial decrease in agricultural land under the scenarios of Global BAU | Local ACTION and Global ACTION | Local ACTION, down to 450,000 ha, compared to the scenario of Global ACTION | Local BAU, which was projected to be around 691,000 ha by the year 2050.The nal indicator for SDG15 is the blue-green algal bloom, which can signi cantly affect water quality.In the Global BAU | Local BAU scenario, the blue-green algal bloom was projected to rise from 4841 units/ML in 2010 to 4984 units/ML in 2050.The projections of algal blooms indicated a relatively minor increase under the scenarios of Global BAU | Local ACTION and Global ACTION | Local ACTION, reaching 4920 units/ML.In comparison, the scenario of Global ACTION | Local BAU was projected to be around 4990 units/ML by the year 2050.
Global BAU | Local BAU and Global BAU | Local ACTION.This slight difference is explained by the amount of water yield under RCP4.5 and RCP2.6 in each scenario.Despite these variations, there was an evident convergence towards a similar salinity level across all four scenarios approximately 65 µS/cm by the year 2050, indicating a common trend in water salinity increase.In the Global BAU | Local BAU scenario, the agricultural pro t, an indicator associated with SDG8, escalated from 556 $M in 2010 to 1875 $M by the year 2050.Under the scenario Global BAU | Local ACTION, the agricultural pro t was projected to reach 3351 $M by the year 2050, making it the highest among all four scenarios.Moreover, the substantial demand for food under the global BAU scenario, especially in animal products, (Engström et al. 2021e development of local sustainability solutions, while enhancing the relevance and credibility of the data sourced from integrated assessment models(Koasidis et al. 2023).The ndings provide deep insights into shaping local sustainability policies amidst global and local uncertainties, while the re ned methodologies from this study equip communities to harmonise developmental aspirations with resilience strategies in a dynamic future.4.1.Synthesising synergies and trade-offs between local and global actionsThe Goulburn-Murray region, with its strong agricultural and food production focus (DELWP 2019a), has seen signi cant changes due to various factors (RPG 2020).Scenario analyses indicated a shift in crop production under the Global ACTION scenario(Doelman et al. 2018).This high demand is in uenced by a global economic acceleration and dietary shifts towards a low meat exitarian diet (UNCCD 2017).Implementing both global and local action resulted in increased crop production.The synergy of global actions, like the increased demand under exitarian diets, and local initiatives, such as enhanced agricultural productivity (Aither 2019; DAWR 2017), ampli ed crop production despite the reduction in crop loss and waste.This alignment between local and global sustainability actions also bene ts the local region through an expanded agricultural sector.The results highlight the need for proactive global and local measures to boost agricultural productivity, meet global food demand, and promote sustainability.Dairy is the cornerstone of the region's agriculture (RMCG 2016).Scenario analysis revealed differences in dairy production between the Global ACTION | Local BAU and Global BAU | Local ACTION scenarios.The increase in dairy production under Global BAU | Local ACTION is linked to continued high animal-sourced food demand as per SSP2(Doelman et al. 2018) and BAU diet, and trend food loss and waste(Alston et al. 2018) while the local action component sees improved farming practices and water-use e ciency contribute to rising dairy productivity to meet burgeoning demand.The Global ACTION | Local BAU scenario sees declining dairy production due to decreased demand and lower productivity and resource-use e ciencies and does not align with the region's dairy-driven economy.This spillover highlights the need for local communities to be aware of the potential in uence of cross-scale interactions.To effectively respond to these spillovers, it is essential to incorporate local actions that align with both global and local sustainability goals.Adapting to variable water availability is already integral to the Goulburn-Murray region's agricultural practices (DELWP 2019a; RPG 2020).Given water's pivotal role in agriculture, economy, and the environment, projecting its availability under various future scenarios is crucial for sustainable planning.While local actions offer bene ts, they may not su ce in signi cantly improving net water availability or salinity, stressing the need for broader coordinated efforts and sustainable salinity management (DELWP 2019b).To achieve the SDG6 targets and protect freshwater ecosystems, collaboration between stakeholders, including local communities, Rural Water Authority, Catchment Management Authorities, and both State and Federal Governments is essential.Our ndings highlighted the importance of multi-level coordination and cooperation for sustainable development (DELWP 2019a).The economy of the study area is strongly tied to agricultural activities (RMCG 2016).Hence, crafting policies that bolster agricultural pro tability is pivotal for local sustainability (RPG 2020).The marked agricultural pro t driven by Local ACTION scenario highlights the alignment between region-speci c local actions, such as improved agricultural productivity and reduced costs.While the Global ACTION | Local ACTION scenario shows signi cant pro tability, the trade-offs associated with reduced demand for animal-based products impacted pro tability on the study area due to heavy dependence on dairy (RMCG 2019).Best performance across multiple require harmonising local strategies with global actions (Chassagne 2020).The ndings stressed the need for strategies that both strengthen the local economy and align with global sustainability goals, ensuring mutual bene ts.The assessment of agricultural land for SDG15 revealed the balance needed between agricultural land development and environmental preservation.Sustainable land-use practices are crucial given how land development impacts biodiversity and landscape(Baral et al. 2014).The projections indicated a more pronounced decrease in agricultural land under the Global ACTION | Local ACTION scenario, as compared to the other scenarios, illuminating the potency and alignment of both global and local actions.The local actions, which might include measures to enhance agricultural productivity or introduce sustainable farming practices, not only bolstered production to meet agricultural targets but also limited the sprawl of agricultural land which is important for environment as well(Walker et al. 2009).Reduction in landintensive livestock demand and reduced loss and waste under the Global ACTION scenario also reduced the total amount of land required to meet demand.When global and local sustainability actions align, they not only boost agricultural productivity but also promote sustainable land use, ensuring a balance between development and environmental conservation.4.2.Aligning local and global actions for achieving the SDGsIn an interconnected world, actions to one SDG at one scale or in one region can have spillover effects, in uencing other SDGs at different scales or in different regions due to the interlinkages between the 2030 Agenda's goals(Sachs et al. 2020).To truly ensure policy coherence for sustainable development, it is essential to consider all externalities and spillover effects (SDG Watch Europe 2019).Spillovers Upon evaluating the impacts of local and global actions, we discovered trade-offs associated with the adoption of global actions, such as dietary changes and food waste reduction which caused a decrease in dairy demand.Assessment of agricultural pro t and dairy production revealed that these two SDG indicators are in uenced by both global and local scenarios.Local actions positively impacted the dairy production and local economy, while global actions in this context had a contrasting effect.Speci cally, the shift towards a exitarian diet, decreasing food waste, and reducing the dairy demand through global actions did not effectively improve the region's economy, mainly due to the signi cant role of dairy production in the local economy.The decrease in dairy production may not be in line with the aspirations of the local people, who value and rely on dairy farming for their livelihoods and cultural practices (RMCG 2016).Considering potential future global scenarios that might result in reduced demand for animal product foods (UNCCD 2017) or other global sustainability actions con icting with local bene ts, local communities should proactively consider strategies to adapt their agricultural practices, diversify their products, and explore alternative local sustainability actions to be aligned with global SDGs(Ningrum et al. 2023).Being proactive and exible in responding to evolving circumstances is crucial for achieving long-term sustainable development goals and ensuring resilience in the face of changing global dynamics.By pursuing sustainability approaches that amplify positive effects and diminish negative spillovers, the gap between the 2030 Agenda's global vision and its local implementation is bridged(Engström et al. 2021).
(Moallemi et al. 2020chosen policies: diverse approaches to achieve an SDG target within an area can result in varying spillover effects (SDG Watch Europe 2019).In response to the need for research that helps local regions identify the spillover effects of their 2030 Agenda implementation(Engström et al. 2021), we showcased how different local sustainability actions under global uncertainties shape future trajectories, speci cally regarding eight local SDG indicators, and their broader implications for global SDGs.Indeed, there are both trade-offs and synergies involved when considering the implementation of both local and global actions(Moallemi et al. 2020).