Even low levels of tree cover improve dietary quality in West Africa

Bowy den Braber, Charlotte Hall, Martin Brandt, Florian Reiner, Maurice Mugabowindekwe, Laura Vang RasmussenPNAS 2024 · DOI

TL;DR

Even low levels of tree cover, particularly outside of dense forests, significantly improve children's dietary quality in West Africa by increasing the consumption of vitamin A-rich foods. This effect is more pronounced for poorer households in certain regions, highlighting the crucial role of trees in sustainable food systems, even in dryland environments.

What Problem Does This Paper Address?

The global food system faces a dual crisis: widespread undernourishment and the consumption of low-quality diets leading to micronutrient deficiencies and noncommunicable diseases. While forests are recognized for their potential to provide nutritious food, most research has focused on areas with dense tree cover. There's a significant knowledge gap regarding the impact of sparse tree cover, common in regions like West Africa, on dietary quality. Existing remote sensing methods have struggled to accurately quantify tree cover in drylands, hindering large-scale assessments of this crucial link. This paper aims to bridge this gap by investigating how varying levels of tree cover, including scattered trees outside of forests, influence the consumption of vitamin A-rich foods among young children in West Africa.

Methods

The study employs a novel combination of advanced remote sensing and detailed household survey data.

  • Tree Cover Estimation: The researchers utilized high-resolution (3m) PlanetScope imagery to create detailed tree cover estimates across West Africa. A key innovation here is the ability to quantify tree cover in dryland areas, which has been a limitation of previous methods.
  • Dietary Data: Data on children's (aged 12-59 months) consumption of vitamin A-rich foods was sourced from Demographic Health Surveys (DHS) across 10 West African countries.
  • Statistical Analysis: To address potential selection bias (i.e., why certain households might be located in areas with more or less tree cover), the study used a sophisticated quasi-experimental matching technique called covariate balancing generalized propensity scores (CBGPS). This method adjusts for confounding factors. Following matching, generalized additive models (GAMs) were used to analyze the relationship between tree cover and dietary quality. GAMs are crucial because they allow for the detection of nonlinear relationships, meaning the effect of tree cover isn't necessarily constant across all levels.

Datasets

  • Demographic Health Survey (DHS) Data: Data from over 80,000 households across 10 West African countries (Senegal, Mali, Nigeria, Gambia, Benin, Guinea, Chad, Liberia, Sierra Leone, and Cameroon) were used. Specifically, data from 15,875 households with children aged 12-59 months were analyzed for dietary consumption.
  • Tree Cover Data: A recent continental map of African tree cover in 2019, derived from 3m PlanetScope imagery using a deep-learning model, was employed. This dataset includes tree cover from both forest and non-forest areas.
  • Ecoregion Data: Data defining terrestrial ecoregions (Olson et al., 2001) was used to categorize the study area into distinct ecological zones (Sahel-Sudan and lowland forest-savanna).
  • Poverty Data: The living standards dimension of the Multidimensional Poverty Index (MPI-LS) was used to assess household poverty levels.

Key Insights and Ideas

  • The Power of Sparse Trees: The most striking finding is that even low levels of tree cover, often found outside of dense forests, significantly boost the consumption of vitamin A-rich foods. This challenges the notion that only large, intact forests provide substantial dietary benefits.
  • Non-Linearity Matters: The relationship between tree cover and dietary quality is not linear. The benefits plateau and even slightly decline after a certain point (around 18% tree cover), suggesting that an optimal level exists, and simply maximizing tree cover isn't always the most effective strategy for dietary improvement.
  • Context is Crucial: Ecoregions and Poverty: The impact of tree cover is not uniform. It interacts with both the ecological region and the socioeconomic status of households, revealing nuanced pathways to improved nutrition.
  • "Free" Food Pathways: The study implicitly supports the idea that trees provide "free" food resources, particularly for poorer populations who may rely more heavily on wild edibles.

Key Results

  • An increase in tree cover from the median of 8.8% to 16.8% was associated with an increase in the probability of consuming vitamin A-rich foods from 0.45 to 0.53.
  • The positive effect of tree cover on vitamin A-rich food consumption was observed to flatten out when tree cover exceeded 30%.
  • In the lowland forest-savanna ecoregion, poorer households were significantly more likely to consume vitamin A-rich foods at low levels of tree cover (P<0.001).
  • In the Sahel-Sudan ecoregion, the effects of tree cover were more equitable across poverty levels, with non-poor households benefiting marginally more at moderate tree cover levels (15-25%).

Limitations and Future Work

  • Causality: The study acknowledges that the data is cross-sectional, meaning it shows associations rather than direct causal links. While sophisticated statistical methods were used to mitigate bias, experimental manipulation is needed for definitive causal claims.
  • Pathway Specificity: The study could not disentangle the exact pathways through which trees influence diets (e.g., direct provision of wild foods, income from NTFPs, ecosystem services, or fuelwood for cooking).
  • Future Directions:
    • Investigate how tree presence influences agricultural diversity, yields, and crop nutritional content.
  • Develop remote sensing techniques that can identify specific tree species to understand which ones contribute most to dietary quality.
  • Gather more detailed food consumption data to quantify micronutrient intake.
  • Trace the origin of consumed foods to differentiate between direct collection from trees and market purchases.
  • Explore the specific mechanisms driving the observed differences across ecoregions and poverty levels.

Summary updated

Abstract

Forests are attracting attention as a promising avenue to provide nutritious and "free" food without damaging the environment. Yet, we lack knowledge on the extent to which this holds in areas with sparse tree cover, such as in West Africa. This is largely due to the fact that existing methods are poorly designed to quantify tree cover in drylands. In this study, we estimate how various levels of tree cover across West Africa affect children's (aged 12-59 months) consumption of vitamin A-rich foods. We do so by combining detailed tree cover estimates based on PlanetScope imagery (3 m resolution) with Demographic Health Survey data from >15,000 households. We find that the probability of consuming vitamin A-rich foods increases from 0.45 to 0.53 with an increase in tree cover from the median value of 8.8 to 16.8% (which is the tree cover level at which the predicted probability of consuming vitamin A-rich foods is the highest). Moreover, we observe that the effects of tree cover vary across poverty levels and ecoregions. The poor are more likely than the non-poor to consume vitamin A-rich foods at low levels of tree cover in the lowland forest-savanna ecoregions, whereas the difference between poor and non-poor is less pronounced in the Sahel-Sudan. These results highlight the importance of trees and forests in sustainable food system transformation, even in areas with sparse tree cover.
BibTeX
@inproceedings{braber2024even,
  author = {Bowy den Braber and Charlotte Hall and Martin Brandt and Florian Reiner and Maurice Mugabowindekwe and Laura Vang Rasmussen},
  title = {{Even low levels of tree cover improve dietary quality in West Africa}},
  booktitle = {PNAS},
  year = {2024},
  doi = {10.1093/pnasnexus/pgae067},
}
Figure