📄 agriculture

A Terrain-Thermal Convergence Framework for Agricultural Climate Suitability: Evidence from Hawaiian Specialty Coffee

This study introduces a terrain-thermal convergence framework demonstrating that traditional thermal-only climate suitability models significantly overestimate agricultural adaptation potential for Hawaiian specialty coffee by ignoring critical topographic constraints, which reveal that land availability is finite and discontinuous rather than a continuous surface.

Original authors: Simon-Hans Edasi

Published 2026-07-06

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Simon-Hans Edasi

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The Big Idea: The "Thermostat" vs. The "Floor Plan"

Imagine you are trying to find the perfect spot to grow a very picky plant, like Hawaiian specialty coffee.

Most climate scientists look at the weather forecast and say, "Hey, if it gets warmer, the perfect temperature for this plant will move up the mountain. So, farmers should just move their crops higher up!" They draw a map showing a huge new area of land that looks perfect for coffee because the temperature is right.

This paper says: "Wait a minute. That map is lying to you."

The author, Simon-Hans Edasi, argues that looking at temperature alone is like looking at a thermostat and assuming you can build a house anywhere the temperature is comfortable. But you can't just build a house on a cliff, in a swamp, or on a steep lava rock face. You need a floor plan that actually exists.

This paper introduces a new way of thinking called the "Terrain-Thermal Convergence Framework." In plain English, it means: Don't just ask, "Is it warm enough?" Ask, "Is it warm enough AND is the ground actually flat and safe enough to farm?"

The Two Coffee Towns: Kona and Ka'u

The study looks at two specific coffee-growing areas on the Big Island of Hawaii: Kona and Ka'u. They are like two different neighborhoods with very different personalities.

Kona (The Picky Neighbor):
- The Climate: Kona coffee likes it cool and specific. It has a very narrow "comfort zone."
- The Problem: As the planet warms, the "cool zone" moves up the mountain.
- The Old Mistake: Old models said, "Great! The coffee can just move up the mountain to stay cool."
- The Reality Check: The author looked at the actual ground. He found that the "Kona-style" land (gentle slopes, specific soil) stops at a certain height. Above that height, the land becomes too steep, rocky, or wild.
- The Result: The "upslope refuge" doesn't exist. The coffee can't move up because there is no "Kona-style" ground up there. The old models overestimated the available land by making it look like there was a whole new neighborhood to move into, when in reality, the neighborhood just ends.
Ka'u (The Chill Neighbor):
- The Climate: Ka'u coffee is tougher. It likes it a bit warmer and can handle a wider range of temperatures.
- The Problem: As the planet warms, the "perfect temperature" moves into Ka'u's current zone.
- The Old Mistake: Old models said, "Wow! Now that it's warmer, we can grow coffee on almost the entire island!" They showed a massive map of red zones indicating huge expansion.
- The Reality Check: The author applied the "floor plan" check. He found that while the temperature is right for coffee all over the island, the terrain (the shape of the land) is not. Most of the island is too steep or has the wrong soil.
- The Result: The "expansion" is an illusion. The land that looks good on a temperature map is actually a steep cliff or a lava field.

The "20x" Mistake

The paper does some math to show just how wrong the old models are.

For Ka'u: The old temperature-only models suggested there was 20 times more land available for expansion than there actually is. It's like a real estate agent telling you, "You can build a mansion anywhere in this city!" when really, you can only build on 5% of the lots because the rest are swamps or mountains.
For Kona: The old models suggested there was a safe place to move to. The new model says, "Nope. That place doesn't exist."

The New Map: The "Decision Surface"

The author created a new map (Figure 4 in the paper) that acts like a truth filter.

Blue areas: These are spots where the temperature is getting worse for Kona coffee, and there is no safe place to move.
Red areas: These are spots in Ka'u where the temperature is getting better, but only if the ground is actually farmable.
Grey areas: These are the "fake" spots. The temperature might be perfect, but the ground is too steep or rocky. You can't farm there.

Why This Matters (In Simple Terms)

If you are a coffee farmer or an investor:

Don't trust the temperature map alone. It's like trusting a weather app that tells you it's a nice day for a picnic, but doesn't tell you that the park is closed for construction.
For Kona farmers: You can't just "move up the hill." You have to stay put and figure out how to keep your current trees cool (using shade trees, for example) because the "new land" doesn't exist.
For Ka'u farmers: There is room to grow, but you have to be careful. Don't buy land just because the temperature is right. Check if the ground is actually flat enough to plant on.

The Bottom Line

The paper concludes that climate suitability is not a smooth, continuous surface. It is a "constrained manifold."

Think of it this way:

Temperature is the ceiling (how high the heat can go).
Terrain is the floor (what you can actually stand on).
Real Farming only happens where the ceiling and the floor meet.

If you only look at the ceiling, you think you have a huge room to live in. But if you look at the floor, you realize you're actually standing on a tiny, specific ledge. This paper teaches us to always check the floor before we celebrate the ceiling.

Technical Summary: A Terrain-Thermal Convergence Framework for Agricultural Climate Suitability

Problem Statement
Current models of agricultural climate suitability typically treat land as a continuous surface where any thermally appropriate area is implicitly available for production. This study argues that this assumption fails for specialty agriculture, particularly in geographically specific systems like Hawaiian coffee, where cultivable terrain is spatially structured, finite, and discontinuous. The paper posits that treating climate suitability as a continuous surface systematically overestimates adaptation opportunities by ignoring the "terrain floor"—the specific topographic, soil, and aspect constraints that define the realized agricultural niche. In volcanic, orographic systems, thermal projections often include steep lava fields, isolated ridgelines, and hydrologically disconnected terrain that satisfy temperature criteria but cannot support agriculture.

Methodology
The study introduces a "Terrain-Thermal Convergence Framework" that models suitability as the intersection of a Gaussian thermal niche and a topographic feasibility screen. The analysis focuses on the Kona and Ka'u districts of the Big Island of Hawai'i, utilizing a 500 m island-wide grid ( $n = 385$ labeled farm cells).

Data Integration: The framework integrates geospatial data from the 2020 Hawaii Agricultural Land Use Baseline, USGS 3D Elevation Program (3DEP), Sentinel-2 NDVI, ERA5-Land climate reanalysis, and SSURGO soil properties.
Topographic Feasibility Screen ( $T_r$ ): Eight standardized topographic features (mean elevation, elevation SD, max slope, aspect sine/cosine, total/local relief, distance to coast) are used to define a "topographic feasibility set." Cells are scored based on their Euclidean distance to regional centroids (Kona and Ka'u). Only cells within the 95th-percentile distance threshold of observed farm terrain are considered "feasible."
Thermal Suitability Set ( $C_r$ ): Separate Gaussian thermal envelopes are fitted to ERA5-Land temperatures for each district ( $Kona: \mu=18.4^\circ C, \sigma=1.4^\circ C$ ; $Ka'u: \mu=20.0^\circ C, \sigma=2.1^\circ C$ ). Future suitability is projected using ensemble CMIP6 warming scenarios ( $\Delta T \approx +1.00^\circ C$ by 2035; $+1.35^\circ C$ by 2045) applied via a lapse rate of $-0.89^\circ C \text{ km}^{-1}$ .
Convergence Analysis: The final suitability metric is the intersection $F_r(\Delta T) = T_r \cap C_r(\Delta T)$ . The study quantifies the "overestimation factor" as the ratio of the thermal-only suitable area ( $|C_r|$ ) to the terrain-constrained suitable area ( $|F_r|$ ).

Key Results

Magnitude of Overestimation: Climate-only models overestimate terrain-accessible adaptation opportunities by approximately one to two orders of magnitude. For the Ka'u district, thermal-only projections imply a vast island-wide expansion, but the terrain screen reduces the gaining area to only 5.1% of farmable terrain, resulting in an overestimation factor of 19.4× (approx. 20×).
The Illusion of the Upslope Refuge: For the Kona district, thermal-only analysis suggests a 36.5% gain in suitability by 2035, implying an upslope migration refuge. However, the terrain screen reveals that the Kona agro-ecological profile (specific slope, aspect, and soil structure) does not extend into the higher elevation bands that become thermally suitable. Consequently, the "upslope refuge" disappears entirely when terrain constraints are applied; zero Kona-profile cells gain thermal suitability.
Divergent Trajectories:
- Kona: Faces a "terrain-locked thermal contraction." Its narrow thermal niche contracts within a manifold that has no upslope continuation.
- Ka'u: Faces a "spatial constraint." While its broader thermal niche expands and aligns with warming, the expansion is bounded by a fixed terrain envelope that does not grow with the thermal signal.
Spatial Distribution: Only 3.8% of the island's land matches the Kona topographic profile, and 4.8% matches the Ka'u profile. The remaining 91.4% lacks the necessary agro-ecological signature, regardless of thermal conditions.

Significance and Claims
The paper claims that agricultural climate suitability is not a continuous surface but a "constrained feasibility manifold." The primary significance of this work is the demonstration that terrain constraints are a first-order control on adaptation space, not a secondary refinement.

Structural Asymmetry: The study highlights that different districts face structurally different risks. Kona's risk is thermal contraction within a fixed terrain belt, while Ka'u's risk is the misallocation of investment based on the false appearance of vast expansion potential.
Decision Support: The "convergence map" (Figure 4) serves as a decision surface that distinguishes between theoretical thermal opportunity and actionable adaptation space. For Kona, it identifies specific fringe cells for shade-management and variety trials. For Ka'u, it identifies specific non-farmed cells for expansion feasibility studies, emphasizing that terrain screening must precede thermal analysis.
Generalizability: The authors argue this is not unique to Hawai'i. The framework applies to any specialty crop system where the realized niche is defined by a spatially restricted terrain configuration (e.g., Champagne geology, Colombian coffee highlands). In such systems, climate-only projections will systematically overstate accessible adaptation space, leading to potential mispricing of supply risk and misdirected investment.

The paper concludes that for geographically specific specialty crops, a feasibility screen is a prerequisite for any projection intended to inform land-use or investment decisions. Without this correction, the gap between climatically suitable land and actually cultivable land remains unquantified, leading to systematic errors in adaptation planning.