Environmental changes in biodiversity hotspot ecosystems of
South Ecuador: RESPonse and feedback effECTs (FOR2730)

Abstract:

Tropical dry forests are among the most threatened terrestrial ecosystems worldwide. This study applies robust multi-objective optimization and Pareto frontier analysis to support sustainable land-use planning in dry forest ecosystems, taking as an example the drylands of southern Ecuador. By integrating ecological and socioeconomic indicator bundles, we modeled optimal land-use compositions under uncertainty and compared them to observed allocations derived from GIS, field data, and farmer input. The observed landscape, dominated by silvopasture (57%) and maize (32%), contrasts with the model’s optimal allocation, which prioritizes shaded cocoa (25%) and coffee (23%), reduces silvopasture (15%), and modestly increases maize (37%). The model enhanced a land-use performance index across different levels of considered uncertainty (low: 22–48%; moderate: 10–32%; high: 16–32%), revealing the method’s strength in generating valuable farm-level insights. The Pareto frontier analysis indicated trade-offs between bundles of ecological and economic indicators, mirroring real-world tensions. While observed land use aligns closely with optimized socioeconomic objectives, it underperforms ecologically. Agroforestry emerges as a promising compromise, though incentives and policy support will be key for adoption. Our findings illustrate how robust multi-objective optimization can strengthen intuitive diversification strategies, balance short- and long-term goals, and guide transitions to more resilient land uses. This is especially critical in vulnerable, data-scarce dry ecosystems increasingly affected by environmental and socioeconomic stressors.

Abstract:

Increasing land-use conflicts call for the development of land-use systems that reconcile agricultural production with the provisioning of multiple ecosystem services, including climate change mitigation. Agroforestry has been suggested as a global solution to increase land-use efficiency, while reducing environmental impacts and economic risks for farmers. Past research has often focused on comparing tree-crop combinations with agricultural monocultures, but agroforestry has seldom been systematically compared to other forms of land-use diversification, including a farm mosaic. This form of diversification mixes separate parcels of different land uses within the farm. The objective of this study was to develop a modelling approach to compare the performance of the agroforestry and farm mosaic diversification strategies, accounting for tree-crop interaction effects and economic and climate uncertainty. For this purpose, Modern Portfolio Theory and risk simulation were coupled with the process-based biophysical simulation model WaNuLCAS 4.0. For an example application, we used data from a field trial in Panama. The results show that the simulated agroforestry systems (Taungya, alley cropping and border planting) could outperform a farm mosaic approach in terms of cumulative production and return. Considering market and climate uncertainty, agroforestry showed an up to 21% higher economic return at the same risk level (i.e. standard deviation of economic returns). Farm compositions with large shares of land allocated to maize cultivation were also more severely affected by an increasing drought frequency in terms of both risks and returns. Our study demonstrates that agroforestry can be an economically efficient diversification strategy, but only if the design allows for economies of scope, beneficial interactions between trees and crops and higher income diversification compared to a farm mosaic. The modelling approach can make an important contribution to support land-use decisions at the farm level and reduce land-use conflicts at the landscape level.

Abstract:

We usually have only limited knowledge about the economic consequences of land-use decisions, thus they are uncertain. We analyze the implications of this uncertainty on conservation payments (CP) to preserve wildlife-friendly shade coffee production in southwest Ecuador, when conversion to maize is the most profitable alternative. Our objective is twofold: First, we analyze the consequences of applying Stochastic Dominance (SD) to derive CP, an approach making only minimal assumptions about the preferences of farmers. Second, we investigate the effects of land-use diversification to reduce CP by allowing for shade coffee on part of a landholding, and maize production on what remains. CP derived by SD turned out to be at least twice the amount calculated by an alternative method which maximizes a concave utility function?US$ 166 to US$ 294 ha-1 year-1 instead of US$ 86 ha-1 year-1. Given this result, we doubt that the assumptions underlying SD are reasonable for farmers, who are known to be riskaverse. Allowing for land-use diversification has a significant impact on CP. The optimal portfolio share of shade coffee is 27 % and for maize 73 % for moderately risk-averse farmers?without any CP. A larger share of shade coffee is preferable for strongly risk-averse farmers?51 and 49 % maize. The amount of CP necessary to encourage the expansion of shade coffee to 75 %is US$ 40 ha-1 year-1 (for moderately risk-averse) and US$ 19 ha-1 year-1 (for strongly risk-averse farmers). Stimulating diversification may thus help to significantly reduce CP necessary to preserve less profitable agroforestry options.