Figure 1:
Map of church forest study sites in South Gondar, Ethiopia. (A) The Amhara Peoples National Regional State in Ethiopia.
(B) The South Gondar Zone in the Amhara Region. (C) Study sites in the South Gondar Zone.
Table 1
Summary of church forest characteristics.
| Church name | Locationa | Area (ha) | Perimeter (m) | Elevation (m)a | Proximity to paved road (km) | Year established |
|---|---|---|---|---|---|---|
| Hiruy | N11°51′ E38°03′ | 4.4 | 763 | 2611 | 1.8 | 360 |
| Debresena | N11°51′ E37°59′ | 11.8 | 1382 | 2690 | 1.4 | ˜1500 |
| Dengolit | N11°36′ E38°04′ | 20.5 | 2197 | 2500 | 33.4 | ˜1300 |
| Gelawdios | N11°38′ E37°48′ | 70.2 | 5560 | 2549 | 42.8 | ˜1500 |
aWe recorded forest elevation and geographic coordinates using handheld Global Positioning System (GPS) units, and calculated forest area (ha), perimeter (m) and proximity to paved roads (km) using Google Earth Pro and QGIS version 2.0.1. Actual church landholdings may differ substantially from forested area, as not all church land is church forest.
Table 2:
Summary of church forest community respondent and household characteristics.
| 2002 | 2014 | 2002 | 2014 | 2002 | 2014 | 2002 | 2014 | |
|---|---|---|---|---|---|---|---|---|
| Male | 52% | 49% | 93% | 77% | 81% | 88% | 68% | 63% |
| Farmer | 95% | 91% | 100% | 100% | 97% | 100% | 85% | 80% |
| Age | ||||||||
| Under 20 | 14% | 0% | 0% | 0% | 0% | 0% | 8% | 0% |
| 20–39 | 29% | 39% | 40% | 13% | 65% | 19% | 40% | 35% |
| 40–60 | 38% | 52% | 37% | 67% | 26% | 68% | 33% | 48% |
| Over 60 | 19% | 9% | 23% | 20% | 10% | 13% | 20% | 18% |
| Literacy | 33% | 17% | 43% | 27% | 48% | 45% | 45% | 33% |
| Own woodlot | 62% | 52% | 60% | 63% | 53% | 55% | 56% | 43% |
| Own land | – | 78% | – | 97% | – | 94% | – | 85% |
| Mean land size (ha)a | – | 1.26 | – | 0.88 | – | 1.15 | – | 0.71 |
| Mean cropland (ha)a | – | 0.74 | – | 0.78 | – | 0.95 | – | 0.69 |
| Mean woodlot (ha)a | – | 0.18 | – | 0.05 | – | 0.14 | – | 0.10 |
| Mean pasture (ha)a | – | 0.30 | – | 0.08 | – | 0.18 | – | 0.07 |
| Number of respondents | 21 | 21 | 30 | 30 | 31 | 31 | 40 | 40 |
aDetailed land ownership data were collected in 2014; these variables are not available for the 2002 sample.
Figure 2:
Landholding and landlessness across farm communities surrounding four church forest sites.
Figure 3:
Beliefs surrounding why natural forests are declining in the area surrounding 4 church forest sites.
Figure 4:
Extractive benefits (red shades) versus renewable benefits (green shades) sought from four church forests in 2014.
Figure 5:
Changes in extractive benefits (red shades) versus renewable benefits (green shades) sought from church forests, 2002–2014.
Figure 6:
Changes in respect for religious versus legal authority in church communities, 2002–2014.
Figure 7:
Changes in beliefs surrounding why church forests persist, 2002–2014.
Table 3:
Logistic regression results with community and year fixed effects for church forest attitudes and behaviors by year (2002 versus 2014) and using pooled sample clustered by household (2002–2014).
| 2002 | 2014 | 2002–2014 | 2002 | 2014 | 2002–2014 | 2002 | 2014 | 2002–2014 | 2002 | 2014 | 2002–2014 | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Male | –0.14 | 0.37 | 0.03 | –1.37** | 0.42 | –0.29 | 1.16** | –0.25 | 0.21 | –0.21 | 0.19 | –0.12 |
| Over 40 | –0.60 | 0.60 | –0.13 | –0.57 | –0.95* | –0.46 | 0.61 | 0.98* | 0.39 | 0.62 | –0.65 | 0.25 |
| Literate | 0.61 | 0.92* | 0.68** | –0.13 | 0.78* | 0.25 | 0.68 | –0.57 | 0.10 | 0.21 | 1.94*** | 0.44 |
| Own woodlot | –0.06 | –2.87*** | –1.46*** | –0.99 | 0.66 | –0.08 | –0.20 | –1.51** | –0.92** | –0.64 | –0.72 | –0.78*** |
| Landholding | – | 0.31 | – | – | 0.44 | – | – | –0.62*** | – | – | 0.24 | – |
| Hiruy | – | – | – | – | – | – | – | – | – | – | – | – |
| Debresena | 1.16 | 1.37 | 1.19* | 2.01** | 2.06** | 1.16*** | –2.12*** | 1.73** | –0.10 | –0.45 | –0.19 | –0.25 |
| Dengolit | 1.45 | –1.66** | –0.29 | 2.60*** | 1.85** | 2.03*** | 0.41 | 1.36* | 0.64 | –1.54** | –3.12*** | –1.73*** |
| Gelawdios | –0.21 | –2.53*** | –1.23** | 2.33*** | 2.84*** | 2.24*** | –0.25 | 1.25* | 0.49 | –1.35** | –a | 0.37 |
| Year | – | – | –1.21*** | – | – | –1.20*** | – | – | –1.09*** | – | – | –0.04 |
| Constant | 1.41 | 2.70*** | 3.03*** | 1.12 | –3.28*** | –0.57 | –0.19 | –0.24 | 0.59 | 1.44* | 0.30 | 1.00** |
| Pseudo-R2 | 0.13 | 0.30 | 0.17 | 0.16 | 0.16 | 0.16 | 0.17 | 0.15 | 0.08 | 0.08 | 0.16 | 0.16 |
| N | 122 | 122 | 244 | 122 | 122 | 122 | 122 | 122 | 122 | 122 | 82 | 244 |
*P<0.10; **P<0.05; ***P<0.01.aAll respondents in Gelawdios in 2014 reported a preference for native species; hence Gelawdios is dropped from Model 4 in 2014.