*3.2. Traditional and Diversified Crop Production and Economic Returns*

Over the 20-year management period (2018–2037), the range of estimated profit for breadfruit and banana was positive (Table 5). Profit from raw taro (without sales of additional products) was negative for both scenarios. However, including additional taro products such as poi increased profits, as did allowing for volunteer family effort (15% reduction in costs), although the lower bound of the estimated profit range remained negative (see Figures S1–S4 ). When considering the three crops collectively and assuming no volunteer effort, combined profit ranged from –\$6.33 million to \$14.22 million (Figure 3). Under the family program, the estimated range of profits increased to –\$4.09 to \$16.46 million (Figure 3).

**Table 5.** Estimated total production and profits over 20 years for banana, breadfruit, and taro.

**Figure 3.** Total profits in different crop scenarios (**left**) and profit by crop type (**right**). Note that *taro A* and *B* are used to differentiate the two yield-cost-price scenarios developed in collaboration with Kako'o ' ¯ Oiwi (see SI Methods for details) and ¯ *family* denotes implementation of the family volunteer farming program.

#### *3.3. Energy Savings*

Estimated annual energy savings, owing to the use of organic N-fertilizer in place of synthetic N-fertilizer in taro production, ranged from 0.16 to 0.25 kWh/m2. Over the 20-year management period, this amounted to a cumulative energy savings in the range of 0.20–0.33 million kWh. In terms of energy offsets, the low taro yield scenario would offset 46–48 tons of heavy fuel oil, while the high yield scenario would offset 115–120 tons over the 20-year project time horizon. This is equivalent to 5.3 million kWh (low) to 13.6 million kWh (high).

Aside from interest in supporting the State's renewable energy goals, Kako'o ' ¯ Oiwi is interested ¯ in becoming more self-sufficient in their on-farm energy use to increase the profitability of their programs. The energy required for poi milling in taro production scenario A (1.12 kg/m<sup>2</sup> yield) totaled 37,361 kWh over 20 years. Powering this process with solar energy would require a 3410-watt PV system at an installation cost of \$10,327. Cumulative utility bill savings of \$9908 combined with \$5094 in revenue from selling excess energy back to the grid resulted in a payback period of 15 years. For taro production scenario B (1.79 kg/m2 yield), energy input totaled 59,778 kWh over 20 years. The higher input would require a larger (5450-watt) PV system at an installation cost of \$16,523. This resulted in cumulative utility bill savings and revenue from grid supply sales equal \$15,853 and \$8151 respectively. The payback period remained unchanged at 15 years.

#### *3.4. Sediment and Nutrient Retention*

Overall, we found that the full agriculture restoration scenario decreased sediment export by 38% compared with the current scenario, but that nutrient export could increase by as much as 240% due to fertilizer inputs. However, nutrient export in the restored agriculture scenario was still less than half of export in the urban development scenario (Figure 4). Specifically, sediment exported to Kane'ohe ¯ bay from the wetland was estimated at 1070, 668 and 2365 tons year−<sup>1</sup> for the current, agriculture restoration and urban scenarios, respectively (Table 2). Nutrient export was 2500, 8515, and 17,995 kg year−<sup>1</sup> for the current, agriculture restoration, and urban scenarios.

Sediment export at He'eia stream mouth during baseflow for the current scenario was based on average TSS concentration exported from the wetland (18.5 mg L−<sup>1</sup> from 2013 to 2017; [59]) and combined mean daily discharge of Ha'iku and 'Ioleka'a streams (1.8 2.87 cfs; [60]), translating to approximately 75 tons year−<sup>1</sup> of sediment. We assumed this was 7% of the sediment budget given that much of export is not associated with baseflow [65], which translated into a total current export of 1070 tons year−1. Thus current net input and export (2335 tons input and 1070 tons export), translated to an accumulation of 1265 tons of sediment in the wetland per year (see SI Methods).

For the restored agricultural scenario, Slaets et al. [61]'s accumulation rate for lo'i retention in the agricultural restoration scenario translated to a future accumulation of 1670 tons of sediment per year (for the ~600,000 m<sup>2</sup> of retention space available in this scenario). Sediment retention was null in the urban scenario and conversion of the upland areas within Kako'o ' ¯ Oiwi from non-native vegetation to ¯ urban also increased sediment export by 31 tons year<sup>−</sup>1.

#### Nitrogen

Predicted N export from upper He'eia watershed into the wetland using the InVEST NDR model under the current scenario was 4990 kg year−1. Using the Department of Health [59] average N concentrations and baseflow estimates, we calculated that 2500 kg year−<sup>1</sup> of this import is currently retained by the wetland, resulting in an export of ~2500 kg year<sup>−</sup>1.

For the restored agriculture scenario, we estimated that an additional 6015 kg year−<sup>1</sup> of N (from fertilizer) would be added to the system that may not be taken up by crops (based on 23% uptake by plants), resulting in an export of 8515 kg year−1. Urban development contributed a wastewater derived nitrogen load of 12,960 kg year−<sup>1</sup> in addition to the current predicted export, resulting in a total export of 17,955 kg year-1 (Figure 4).

**Figure 4.** Summary results of sediment accumulation (tons/year), sediment export (tons/year) and nitrogen export (kg/year) under the current, restored agriculture, and urban scenarios.

#### **4. Discussion**

Interest in biocultural approaches to restoration are growing across the world [2,9,20,43], and Hawai'i is emerging as a hotspot for biocultural restoration of traditional agriculture [7,15]. This can be attributed to an indigenous Hawaiian cultural renaissance around food and ahupua'a and broader moku (district or region) management [3,66] coupled with the State of Hawai'i's recent commitment to doubling local food production [31]. Statewide efforts to increase local food while also moving towards renewable energy, watershed and marine ecosystem protection, and local (re) connection to place situate these systems-level biocultural restoration efforts as transformative projects for sustainability and resilience at multiple scales.

A defining principle of biocultural approaches to conservation and restoration is the need to start from a cultural, place-based approach while also acknowledging diverse actors and multiple goals across spatial scales [9,43]. In this study we set out to assess future scenarios of restoration in terms of outcomes valued by the local community and a community-based non-profit as well as the State of Hawai'i as formalized in sustainability goals around food, water, and energy [31]. In the context of biocultural restoration of lo'i kalo in He'eia, O'ahu, we identified a diverse set of locally relevant environmental, cultural, and economic goals, many of which align with statewide sustainability goals (see Table 1). We found a number of benefits, including those stemming from the biocultural restoration process, as well as key challenges facing this and similar biocultural restoration projects as they begin to scale.

From the perspective of participating families in the 'ohana program, the most important benefits associated with cultivating kalo were associated with the *process* of restoring lo'i kalo rather than just the cultural or economic benefits of the end product. In particular, families valued the opportunity to (re)connect to important biocultural landscapes and build social connections among like-minded people. That cultural benefits emerge from the process of restoring reciprocal relationships to place rather than just the end products of biocultural restoration has been noted elsewhere [7,20,67]. The attention to restoring ritual and the cultural protocols alongside ecological systems is an important theme emerging from this study and other articles in this issue [7,20]. While work on "relational values" [68] and to some extent "nature's contribution to people" [27] has acknowledged this, the vast majority of literature in ecosystem services has focused on the value of the end state rather than the process of getting there.

In regards to the process of restoration, we found a suite of benefits across all categories in the indigenous Hawaiian cultural ecosystem services framework of Pascua et al. [17], as well as some extensions. This included benefits classified as 'ike (knowledge): families particularly valued the chance for place-based experiential learning of how to grow kalo and provide opportunities for their children to be in and learn in these environments. There was also substantial reference to themes related to mana (spiritual landscapes) as families described the work as a cultural practice linked to ancestral practices and the Kumulipo (legend of Hawaiian origin), and to pilina kanaka (social connections) as ¯ the program was discussed as strengthening connections between and among families. Participants also pointed to physical and mental health benefits (ola mau), with many references to the lo'i as a place of individual and collective mental renewal as well as physical health benefits. This connection created a deeper understanding of kuleana (responsibility) to the land and of cultural identity as kanaka (indigenous Hawaiians). The site became a place of healing, love, and self-reflection, providing ¯ a higher purpose for the people working to restore the area. Participants described a sense of pride and accomplishment, a sense of belonging (safe space), and valued the opportunity to progress as kanaka (indigenous Hawaiians). ¯

While attention to process is key, an important cultural as well as economic benefit of the restoration is also clearly the production of traditional crops, an outcome highly valued by Kako'o ¯ 'Oiwi and aligned with the State's goal of doubling local food production. Even in the lowest return ¯ scenario, there are still 339 tons of bananas (mai'a), 151 tons of breadfruit (ulu), and 1628 tons of taro (kalo) produced. The food produced would represent an important step towards the State's goal of doubling local food production and would increase the total area dedicated to taro production across the State by 50% [31].

The focus on organic production methods and renewable energy as part of the broader biocultural approach, also produced important energy savings benefits that align with the State's renewable energy goals. In comparison to energy use that would have occurred with the use of industrial fertilizer and oil for electricity, organic fertilizer and solar power also saved 0.87–1.41 million MJ. Adding in potential savings through avoided food imports, this saving is substantial and clearly demonstrates the potential of biocultural restoration of traditional agricultural systems to produce food in a way that also provides synergies for energy sustainability goals.

Our finding that the most financially beneficial model included diversified crop production utilizing a "hybrid" production model that includes the family program also points to important potential synergies between local community and cultural benefits and the program's financial sustainability. Our economic analysis suggests that if the family program were to go to scale, it could roughly increase returns by \$2 million USD, which would help to sustain a project that provides many benefits locally and more broadly. However, to go to scale the number of families must increase substantially. While Kako'o ' ¯ Oiwi is committed to providing these opportunities and "growing ¯ farmers," families interviewed also expressed concern that some of the greatest components of the project (close social connections and a sense of sanctuary) could change as more families get involved. Managing scaling in the context of the cultural significance of the quality of the process of restoration is critical to the long-term success of the project.

Finally, our results suggest that biocultural restoration of lo'i kalo in He'eia can provide benefits in terms of the broader social-ecological system including downstream fishpond (managed by Paepae of He'eia) and the Kane'ohe Bay coral reef ecosystem. In elevated concentrations, both nitrogen and ¯ sediment can have adverse impacts on coral reefs and fisheries and likely make these systems less resilient to climate change [65,69,70]. When compared to a hypothetical urban scenario (which was once the fate of He'eia wetland and could be in the future if sustainable models are not developed), we found important benefits of lo'i kalo restoration in terms of reduced nitrogen and sediment loads to these nearshore environments. This aligns with previous research in the Pacific showing that taro fields retain sediment [71], as well as studies of sediment retention in similar rice paddy systems [72]. Lo'i kalo and similar systems have also been shown to have a high capacity to store water compared with invasive wetland grasses [73] and urban environments [74], suggesting that these systems retain and slow more water because of their construction as basins and, therefore, more sediment than alternative land uses addressed in our study. The approach presented in this article accounted for baseflow and small storm conditions; further research is needed to understand how the conversion back to lo'i would be affected by larger storm events.

Within the restored agriculture scenario, however, we found an important tradeoff between enhanced sediment retention (and reduced sediment export to the bay) and increased nitrogen export (due to fertilizer inputs). While the nutrient loads are much lower than they would be with urban expansion or with other forms of conventional agriculture, they are higher than current land cover, and could have impacts on important downstream systems. However, the model used did not directly consider nutrient uptake of drainage channels or by complex microbial and wetland plant communities, and thus can be considered conservative [75]. Careful design of lo'i and the wetland system could mitigate some of the nutrient tradeoffs while also increasing sediment retention of the system. Field design insights from natural and constructed wetlands used to treat waste discharge sites could also be incorporated into lo'i kalo design to further reduce nutrient export [76].

While addressing this potential nitrogen tradeoff is paramount, leaving the area as invasive grasses provides no economic and little to no direct cultural or community benefit and would likely leave the area more susceptible to urban development pressures. The current plant community also provides little to no ecological habitat value, whereas the restored system (including lo'i kalo as well as native wetland restoration) is expected to increase habitat for native fish (e.g., 'o'opu akupa; *Eleotris sandwicensis*), insects (e.g., *Pantala flavescens*), plants (e.g., neke; *Cyclosorus interruptus*, 'ahu'awa; *Cyperus javanicus)* and birds, including the endangered Hawaiian stilt (ae'o; *Himantopus mexicanus knudseni*). Overall, our research suggests that the restored system would substantially contribute to prioritized local cultural, economic, and ecological goals while also helping to meet the State of Hawai'i's sustainability goals around food, water, and energy. Thus, the project thus represents a locally viable and beneficial opportunity to meet broad societal environmental objectives, which provides broad lessons for the worldwide challenge of local implementation of the SDGs, in an equitable and effective way.

#### **5. Conclusions**

Biocultural approaches to conservation and restoration explicitly recognize the interconnection between biological and cultural diversity and between social and ecological systems that have often been obscured in Western-based conservation efforts. While not a new concept, theories of biocultural restoration that emphasize cultural and place-based perspectives, knowledge, and values are emerging in a context of contemporary conservation and restoration efforts [2,29,43]. Here we demonstrated how existing frameworks of evaluation of synergies and tradeoffs in land management from the food-energy-water and ecosystem services frameworks can be adapted to illuminate potential synergies and tradeoffs among multiple cultural, environmental, and economic goals associated with biocultural restoration projects. An important contribution of such an integrated assessment is that it highlights the potential of biocultural restoration to both achieve locally-relevant cultural, economic, and ecological goals while also contributing meaningfully to broader sustainability goals defined by formal policies.

Kako'o ' ¯ Oiwi is at the center stage of biocultural restoration of social-ecological systems in Hawai'i. ¯ Our collaborative case study from He'eia, Hawai'i suggests that biocultural restoration of traditional agriculture has the potential to simultaneously meet multiple community and statewide sustainability goals, including increasing local food production, reducing energy consumption, increasing cultural connection to place, and decreasing sediment delivery to downstream coastal systems. Yet, there are important tradeoffs to consider in the form of nutrient export, which will be much less than alternate land uses (like urban and conventional agriculture), but still likely an increase from the current fallow, degraded system. By understanding and adapting in light of potential tradeoffs, it is clear that the process of (re)connecting to place inherent in a biocultural approach provides a suite of community and cultural benefits that are essential to the long-term social and financial sustainability of this multi-benefit system.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2071-1050/10/12/4502/s1.

**Author Contributions:** Conceptualization, L.L.B., K.F., C.W., K.B., K.K.-S., N.R., G.C. and T.T.; Data curation, K.F.; Formal analysis, K.F. and C.F.; Funding acquisition, K.F., K.B., K.K.-S. and T.T.; Investigation, L.L.B., K.F., C.F., C.W., K.B. and K.L.L.O.; Methodology, K.F.; Project administration, L.L.B., K.F. and K.K.-S.; Visualization, K.F.; Writing—original draft, L.L.B., K.F., C.F., C.W. and T.T.; Writing—review & editing, L.L.B., K.F., C.F., C.W., K.B., K.K.-S., N.R., G.C., K.L.L.O. and T.T.

**Funding:** The APC was funded by Hawai'i Community Foundation.

**Acknowledgments:** This research would not have been possible without the work and dedication of Kako'o ¯ 'Oiwi staff and volunteers who carry out this restoration and who graciously collaborated with us in this research. ¯ We particularly are grateful to the families in the 'ohana program who participated in interviews with Casey Ching and welcomed us into the lo'i with their families. The knowledge and insight gained was inspiring in this work and beyond. We thank Pua'ala Pascua for input and guidance in the beginning of this project in our use of her cultural ecosystem services framework and Kawika Winter, Kevin Chang, and 3 anonymous reviewers for comments on the manuscript. Thanks also to Victoria Ward who assisted with figures and to Sarah Medoff who assisted in the economic analysis and manuscript compilation. Funding for this project was through National Science Foundation Coastal SEES Program (#SES-1325874), National Science Foundation EPSCoR Program (#1557349), DLNR Commission on Water Resource Management Water Security Grant (#66092), NOAA Coastal Resilience Grant (NA17NMF4630007), and the Ha'oli Mau Loa Foundation graduate research fellowship with the Department of Natural Resources and Environmental Management, University of Hawai'i at Manoa. This is contributed paper WRRC-CP-2019-12 of the Water Resources Research Center, University of ¯ Hawai'i at Manoa, Honolulu, Hawai'i. ¯

**Conflicts of Interest:** The authors declare no conflict of interest.
