Sustained Adoption of Environmentally-Sustainable Practices: Spillovers and Long-Run Impacts

Award Period
Award Amount
Agency Name
Tufts University
Award Number
PI First Name
PI Last Name
Area/s of Research
Climate Change Science
Marine Conservation, Policy and Education

This research seeks to investigate the causes of low adoption rates of rainwater harvesting (RWH)

techniques - specifically demi-lunes - in West Africa, with the Zinder region of Niger as a test site.

RWH techniques are implemented to combat desertification: an irreversible process that leaves land

unsuitable for agriculture. Among the strategies to restore degraded land, techniques that increase

the level and duration of water stored within the soil and replenish soil nutrients, such as RWH

techniques, are among the most promising, but their sustained adoption remains low, despite

decades of investment by governments and NGOs to promote them. This research builds upon a

successful pilot conducted in 2015, as well as a recent larger-scale randomized controlled trial (RCT)

of interventions designed to encourage demi-lune adoption.


This research seeks to measure:

1. the sustained adoption, disadoption and new adoption among farmers in the original study sample;

2. adoption spillovers to neighboring farmers;

3. land abandonment and recruitment of new land into production, and;

4. agricultural production.


We will be supporting research efforts by providing remote sensing expertise to analyze measures of

demi-lune adoption, land degradation, and land abandonment at the Zinder field site. Using

high-resolution satellite imagery from a combination of data sources, we will work with a remote

sensing specialist to:

1. adapt remote sensing algorithm to determine the extent of demi-lune coverage on farmers' fields;

2. measure NDVI (a satellite-based measure of green coloration, ergo primary production) to

approximate the effects of demi-lunes on field productivity, and;

3. use a combination of approaches to identify abandoned (as opposed to fallow) land as a predictor

for future desertification.

These outcomes will be spatially linked to our data using plot coordinates collected during field

verification visits that will allow us to more completely characterize the spatial characteristics of

adoption, productivity and the precursors to desertification.


The expected outputs of this work is a dataset provided to the project PIs that records the outcomes

above at the plot level for the project years, and documentation of the algorithm to continue

measurement into the future.