Investing in Landesque Capital: Spatial Suitability for Biocultural Restoration on Public Lands in Hawai‘i

1. Introduction

Across the Hawaiian archipelago, the physical traces of a highly productive past remain visible in the landscape. Stone walls of dryland field systems stretch across leeward slopes, terrace complexes (lo‘i) step down windward valleys, and the walls of fishponds (loko i‘a) encircle coastal bays. These features are not merely archaeological sites; they are the remnants of what economists Sen (1959) and Brookfield (1984) term "landesque capital"—permanent modifications to the land that provide returns over long periods. In pre-contact Hawai‘i, these systems functioned as the primary infrastructure of society, underpinning a population estimated between 400,000 and 800,000 (Kirch 2007).

Today, however, the mechanisms for funding public infrastructure in Hawai‘i have become decoupled from these landscape-based systems. The State of Hawai‘i’s Capital Improvement Project (CIP) budget—funded primarily through General Obligation bonds—allocates billions of dollars annually for "gray" infrastructure such as highways, harbors, and public buildings. In contrast, the restoration of traditional agricultural systems, which provide comparable public benefits in terms of flood mitigation, aquifer recharge, and food security, is largely excluded from this capital funding stream. Instead, biocultural restoration projects are forced to rely on piecemeal, short-term funding sources like Grants-in-Aid (GIA), which prohibit the long-term planning and "heavy lifting" (e.g., wall restoration, dredging) required to revitalize these systems.

This paper addresses this policy gap by reframing biocultural restoration as a form of capital infrastructure investment. We argue that restoring a fishpond wall or an irrigation network on public land is functionally equivalent to repairing a state reservoir or sea wall and should be eligible for the same long-term financing mechanisms.

To operationalize this concept, we conduct a spatial suitability analysis of state-owned lands to identify areas where biocultural restoration is most ecologically and socially viable. By intersecting government land ownership records with data on indigenous agricultural potential (Kurashima et al. 2019) and State Land Use Districts, we identify a "restoration-ready" land base. Furthermore, we examine the historical context of the "Land Hui" movement as a precedent for community-based management of these resources, suggesting that a return to neo-traditional tenure arrangements—where the state holds title but communities hold long-term stewardship rights—offers a viable pathway for revitalizing Hawai‘i’s landesque capital.

2. Context: A History of Investment, Decline, and Resilience

2.1 The Rise of Landesque Capital

Prior to Western contact, Hawaiian society was characterized by intensive investment in agricultural infrastructure, which Allen (1991) describes as the "material engine" of the Hawaiian archaic state. This was not subsistence farming in a simple sense, but a system of "public works" maintained by communal labor under chiefly direction.

The development of this infrastructure followed distinct geological pathways (Kirch 1994). On the older, eroded islands of Kaua‘i and O‘ahu, investment focused on lo‘i kalo (irrigated taro terraces), which utilized sophisticated hydraulic engineering to divert stream water into valley-wide complexes. These systems were highly effective: Palmer et al. (2009) found that irrigation water in these systems supplied nutrients (calcium) at rates 100 times higher than natural soil weathering, demonstrating their function as high-performance nutrient delivery infrastructure.

On the younger islands of Maui and Hawai‘i, investment shifted to "farming the rock" through vast dryland field systems (such as the Kohala and Kona field systems), where field walls and mulching strategies modified microclimates to support sweet potato and ulu production (Kirch & Sahlins 1992; Vitousek et al. 2004). These investments were capital-intensive: they required immense upfront labor to construct but yielded increasing returns over generations, effectively "banking" labor in the landscape.

2.2 The Collapse (1820–1880)

The disintegration of these systems was neither accidental nor purely environmental. It was a structural collapse driven by converging factors between 1820 and 1880:

1. Political Shock (1819): The abolition of the kapu system (ʻAi Noa) removed the religious authority of the konohiki (land managers) to command the communal labor necessary for cleaning ditches and repairing walls, destabilizing the maintenance regimes required for large-scale infrastructure (Earle 1978).
2. Demographic Collapse: Introduced diseases reduced the native population by over 70%, decimating the labor force required to maintain labor-intensive infrastructure (Kirch 2007).
3. The Sandalwood Trade: The rapid integration into the global economy shifted labor from maintaining "landesque capital" (food production) to extracting natural capital (sandalwood) for immediate trade, leading to the first historical famines (Rice 2020).
4. The Māhele (1848): The transition to fee-simple property fragmented the ahupua‘a system. The traditional right of access to kula (uplands) and kai (sea) was severed from the kuleana (house/farm) plots awarded to commoners, rendering the integrated resource system dysfunctional for many families (Roversi 2012).

By 1880, the era of large-scale communal investment in agricultural infrastructure had effectively ended, replaced by plantation agriculture. Yet, the durability of the initial investment remained visible: rice farmers in the late 19th century utilized the "shell" of abandoned lo‘i as "successional capital," repurposing the terraces for global export crops (Bayliss-Smith et al. 1997). However, the rise of industrial sugar ultimately decommissioned much of this system through "dewatering"—constructing massive ditches that diverted water away from traditional watersheds, physically destroying the functional capacity of the lo‘i (MacLennan 2014).

2.3 The Response: The Land Hui Movement

In the wake of the Māhele, Native Hawaiians did not passively accept dispossession. They innovated new legal structures to maintain communal land tenure within the framework of Western property law. Between the 1860s and 1920s, groups of Hawaiians formed "Land Huis"—cooperative associations that pooled resources to purchase entire ahupua‘a (Roversi 2012; Watson 1932).

The Hui system functioned as an early form of biocultural restoration and resilience. A Hui would hold the land in common, often adopting a constitution and bylaws that mirrored traditional resource management: electing a luna (manager) to oversee water rights, enforce grazing limits, and maintain shared infrastructure, while allocating individual plots for family use. This hybrid governance structure allowed communities to retain access to the full vertical resource zone (mountain to sea) despite the legal fragmentation of the surrounding landscape.

Although many Huis were eventually dissolved by partition suits or hostile acquisition, they established a crucial precedent: that communal management of land, coupled with formal legal tenure, is a viable mechanism for stewardship.

2.4 Modern Landesque Capital Investment and the Funding Gap

Today, a new wave of "Land Huis" has emerged in the form of non-profit community organizations seeking to restore these same landscapes. Groups like Paepae o Heʻeia (O‘ahu), Hui Mālama o Ke Kai (Maui), and Waipā Foundation (Kaua‘i) are actively restoring fishponds and lo‘i systems.

However, these modern organizations face a critical barrier that their 19th-century predecessors did not: they operate primarily on public lands (state or county) without the benefit of ownership capital. While the State of Hawai‘i acts as the legal landowner (often via the Department of Land and Natural Resources), it rarely invests capital funds into the restoration of the "assets" on these lands. Instead, community groups are left to patch together short-term grants for what is essentially public infrastructure maintenance.

For example, the restoration of the 7,000-foot kuapā (wall) at Heʻeia Fishpond involves moving tons of coral and basalt—a capital project in every engineering sense (Paepae o Heʻeia 2025). Yet, because it is classified as "cultural practice" rather than "infrastructure improvement," it largely falls outside the scope of the State’s primary capital financing mechanism, the General Obligation bond market.

3. Methodology

3.1 Study Area and Data Sources

This analysis covers the major Hawaiian Islands (Hawai‘i, Maui, O‘ahu, Kaua‘i, Moloka‘i, Lāna‘i). The spatial framework operates on the NAD83(HARN) / UTM zone 4N projected coordinate system (EPSG:3750) to ensure accurate area calculations.

We integrated four primary datasets to identify suitable sites:

1. Government Land Ownership: Detailed parcel data from the Hawaii Statewide GIS Program, filtered to exclude federal lands and retain only State and County owned parcels. This defines the "public land" universe where state CIP funding is applicable.
2. State Land Use Districts (SLUD): Official district boundaries (Agriculture, Rural, Conservation, Urban) which define the regulatory envelope for each parcel.
3. Indigenous Agricultural Potential: The spatial model developed by Kurashima et al. (2019), which maps the pre-contact extent of intensive agriculture.
4. Current Land Cover: The Hawaii Cropland Data Layer (HCDL) v2.0 (Li et al. 2024), a 10-meter resolution raster product developed through a collaboration between USDA NASS and the University of Hawaii at Manoa. This dataset leverages AlphaEarth Foundations geospatial embeddings to provide crop-specific classification, which we used to identify existing development or conflicts.

3.2 Indigenous Agricultural Modeling

To ground our analysis in traditional ecological knowledge (TEK) and biophysical reality, we utilized the Kurashima et al. (2019) geospatial model. This model reconstructs the "fundamental niche" for traditional Hawaiian agriculture based on three primary systems:

• Lo‘i Kalo (Wetland Taro): Modeled based on the presence of perennial streams, hydraulic soil types, low slope (< 10°), and elevation (< 415m).
• Dryland Field Systems: Modeled for areas with 750–1600mm annual rainfall, slope < 12°, and substrates of sufficient geological age (> 400kyr) to support soil fertility.
• Agroforestry: Defined by broader parameters (slope < 30°, rainfall > 750mm) representing the "colluvial slope" systems that often buffered the core agricultural zones.

Our analysis specifically targets the "High Potential" zones from this model, representing the optimal convergence of climatic and geomorphic conditions.

3.3 Suitability Analysis & Scoring

To identify "restoration-ready" public lands, we performed a spatial intersection of these layers using Python (Geopandas, Rasterstats). We utilized the NAD83(HARN) / UTM zone 4N projected coordinate system (EPSG:3750) for all geometric calculations to minimize distortion.

We developed a composite Suitability Score (0–100) to rank sites based on their viability for ahupua‘a-scale restoration. The scoring algorithm weights three factors:

1. Indigenous Agricultural Potential (40%): Derived from the Kurashima model rank.
   • High = 100 points
   • Medium = 60 points
   • Low = 30 points
2. Land Use Compatibility (35%): Based on the regulatory ease of restoration.
   • Agricultural District = 100 points
   • Rural District = 75 points
   • Conservation District = 50 points (Restoration possible but regulatively burdensome).
   • Urban District = 25 points.
3. Parcel Size (25%): Favoring larger contiguous parcels that allow for system-scale restoration.
   • > 100 acres = 100 points
   • 25–100 acres = 75 points
   • 5–25 acres = 50 points
   • < 5 acres = Scaled linearly down to 0.

Exclusion Criteria: We utilized the HCDL v2.0 to filter for conflict. Using zonal statistics (rasterstats), we assigned each parcel a dominant land cover class. Parcels where the majority pixels were classified as "Commercial/Industrial" or "High Intensity Developed" were flagged for exclusion, ensuring that priority sites are not currently occupied by critical modern infrastructure.

3.4 Contextual Validation

To validate the model, we calculated a Proximity Score based on the distance to known, successful biocultural restoration projects (Reference Projects) such as Paepae o Heʻeia or Waipā Foundation. We employed an inverse linear distance function with a 5km maximum buffer:

• 0 meters (adjacent) = 100 points
• 5,000 meters = 0 points

This score serves as a proxy for social readiness and the potential for knowledge transfer, assuming that proximity to existing practice lowers barriers to entry for new projects.

4. Results

4.1 Distribution of Suitable Lands

The spatial analysis evaluated government-owned parcels across the main Hawaiian Islands (Figure 1). Of these, 1,639 parcels (approximately 313,882 acres) were identified as "Medium Suitability" or higher (Score ≥ 40), representing the core land base for potential biocultural restoration.

Geographic Distribution:
Hawai‘i Island dominates the portfolio in terms of count, while Kaua‘i holds significant large-acreage opportunities. Note: O‘ahu was excluded from this final suitability calculation due to data gaps in the underlying Kurashima et al. (2019) public dataset release; however, high-potential areas are known to exist there and will be added as data becomes available.

Island	Parcel Count	Total Acres	Mean Suitability Score	Primary Opportunity
Hawai‘i	1,327	151,097	64.26	Dryland & Lo‘i
Kaua‘i	165	104,979	49.42	Agroforestry
Maui	76	35,133	46.13	Agroforestry
Moloka‘i	68	22,540	61.50	Agroforestry
Lāna‘i	3	133	50.27	Agroforestry

System Type Analysis:
Breaking down the suitable acreage by traditional agricultural system reveals that Dryland Field Systems represent the largest restoration opportunity by area, particularly on Hawai‘i Island. Agroforestry potential is the dominant system on Kaua‘i, Maui, and Moloka‘i. Lo‘i Kalo potential, while smaller in absolute acreage (approx. 556 acres identified in the core analysis set), represents a high-value cultural asset concentrated on specific hydrological features.

Land Use District Analysis:
The intersection with State Land Use Districts (SLUD) reveals a distinct bifurcation in restoration opportunities:

• Agricultural District: Contains 9,793 government parcels (1.54 million acres) with the highest mean suitability score (43.98). These lands represent the "low-hanging fruit" for restoration—areas where both the ecological potential and the regulatory framework align.
• Conservation District: Contains 2,845 parcels but a massive land area (1.44 million acres). While the mean suitability score is lower (30.32), this is largely due to the scoring penalty applied to the Conservation district. Critically, many of Hawai‘i’s most intact historical systems (e.g., upper valley lo‘i and forest zones) are located here.

4.2 Ownership Patterns

The State of Hawai‘i (DLNR and other agencies) is the primary landowner of suitable sites, holding over 265,000 acres of the high-suitability land base.

• Department of Hawaiian Home Lands (DHHL): DHHL holds the highest number of suitable parcels (961), comprising 33,617 acres. This high parcel count implies a fragmented land base that may benefit from the "Hui" model of aggregated management.
• State (Other/DLNR): Holds the largest contiguous acreage (265,518 acres), including massive forest reserves on Hawai‘i and Kaua‘i suitable for agroforestry.
• County Lands: County governments collectively own 164 suitable parcels (6,059 acres), often in closer proximity to population centers.

Top Suitability Sites:
The analysis identified several "perfect" candidates (Score = 100), including an 8,353-acre State parcel on Hawai‘i Island. These sites represent immediate opportunities for pilot "Landesque Capital" investment projects.

5. Discussion

5.1 Reframing Infrastructure: The "Restoration-Ready" Land Base

The results demonstrate that the State of Hawai‘i possesses a vast inventory of "restoration-ready" land—over 1.5 million acres where public ownership overlaps with indigenous agricultural potential. Currently, these lands are viewed primarily as conservation assets (passive management) or agricultural leases (commercial management).

We propose a Tiered Restoration Strategy based on current land cover, which dictates the level of capital investment required:

• Tier 1: Immediate Opportunity (Fallow/Shrubland): Approximately 14% of the suitable land base (approx. 3,300 acres in the core set) is currently classified as fallow. These lands have minimal conflicting uses and require the lowest barrier to entry—primarily invasive species removal and infrastructure repair. These are prime candidates for immediate "pilot" CIP projects.
• Tier 2: Negotiated Restoration (Pasture): A significant portion (**24%**) is currently used for grazing. Restoring these lands requires a "negotiated" approach, potentially integrating silvopasture models where native trees and dryland crops coexist with managed grazing, rather than displacing current lessees.
• Tier 3: Heavy Lifting (Forest): The majority (**61%**) is currently forested, often with invasive species. Restoring these systems represents a massive capital undertaking—clearing, terracing, and replanting—that fits the definition of "heavy infrastructure" work suited for long-term bond financing.

By reframing these landscapes as public infrastructure, the State could justify the use of CIP bonds to restore them. For example, the Tier 1 parcels represent sites where capital investment in wall repair and irrigation reconstruction would yield direct returns in aquifer recharge and flood mitigation—services that otherwise require expensive gray infrastructure.

5.2 The Conservation Paradox

A critical finding is the vast acreage of high-potential land locked within the Conservation District (1.44 million acres). While intended to protect resources, the strict regulations of the Conservation District often hinder the restoration of the very biocultural systems that created those resources.

Restoring a lo‘i in a Conservation District requires a Conservation District Use Permit (CDUP), a costly and bureaucratic process that treats traditional stewardship akin to development. Our analysis suggests that a significant portion of the state's "landesque capital" remains dormant not because of ecological unsuitability, but because the regulatory framework focuses on preservation (static) rather than restoration (active).

5.3 Future Research: Tenure and Governance

While this analysis identifies where restoration should happen, it leaves open the question of how tenure should be structured. The "Land Hui" history suggests that long-term community stewardship is essential for maintaining these systems.

Future research should explore:

1. Neo-Traditional Tenure: How can modern legal instruments (e.g., long-term stewardship leases, community-based subsistence fishing areas) replicate the functional benefits of the Hui system on public lands?
2. Land Back Mechanisms: How can the identification of these "restoration-ready" public lands inform the social-organizational approaches of #LandBack efforts? Specifically, how can community governance structures be designed to manage these large-scale infrastructure assets effectively?

6. Conclusion

Biocultural restoration is not merely a cultural project; it is a capital improvement strategy for a changing climate. Our analysis confirms that the State of Hawai‘i holds the keys to a vast system of dormant landesque capital—infrastructure that, if restored, could provide food security and climate resilience for generations.

By recognizing these ancient systems as public infrastructure and aligning capital funding (CIP) with their restoration, Hawai‘i can move beyond a model of passive conservation to one of active, abundance-generating stewardship. The land is ready; the policy just needs to catch up.

Data Availability Statement

The code used for the spatial analysis, including scoring algorithms and data processing scripts, is available in the GitHub repository at https://github.com/ainacip/ainacip. The processed datasets, including the "Ideal Sites" shapefiles and prioritization results, are archived within the repository's data/processed/ directory. Raw data sources (State GIS, HCDL) are publicly available from their respective agencies.

7. References