Characteristics of selected watersheds
The study was conducted in the Gumara watershed specifically on the selected four sub-watersheds (Figure 4). The watersheds Girbi and Gena-mechawocha are under the Farta district administrative; Tankua Gebriel in Fogera and Wanzaye in Dera districts. The altitude of the selected watersheds ranging from 1800 to 2846 m a.s.l. (Figure 4). These watersheds cover an area of 581.5 ha for Girbi, 695.5 for Gena-mechawocha, 627.1 for Tankua Gebriel and 555.2 ha for Wanzaye.
The land use land cover of the four watersheds is shown in figure 5 (Gena-mechawocha (a); Tankua Gebriel (b); Wanzaye (c) and Girbi (d). In these watersheds the land use is classified as built-ups, forests, cultivated and grazing land. The built-ups were hard to identify and were the dispersed rural settlements. The area covered under this land use category indicated greater in Girbi watershed since the tip of the watershed includes parts of the Debre Tabor town (about 101 ha) and followed by Gena-mechawocha (7.4 ha, Figure 5). The area under forest cover, which includes trees planted around homesteads and there is an increasing trend over the past years (Halefom et al., 2019). This is attributable to the afforestation program of the government and planting eucalyptus trees at the household level. The area covered with forests in Girbi and Gena-mechawocha were about 277 and 250 ha, respectively (Table 1). However, in Girbi, almost all are eucalyptus plantations in the hillslopes and near the homesteads. The reason is due to its proximity to the urban area (Debre Tabor).
Cultivated land covers the greater percentage of the total watershed area with 28, 62, 72 and 68 percent in the Girbi, Gena-mechawocha, Wanzaye and Tankua Gebriel watersheds, respectively (Table 1). This land-use type increased from time to time due to an increase in population growth and the land redistribution in the Amhara Regional State in 1997 that allocated much of the marginal land to landless farmers. The grazing land area coverage for all the watersheds was minimum. The maximum grazing land coverage from these watersheds was 79 ha from Tankua Gebriel. Therefore, farmers use crop residues as feed sources of livestock.

Data collection and analysis

Data were collected qualitatively through focus group discussions (FGD) with farmers; field observations and district level experts were participated (Figure S1). FGD was used to obtain qualitative data from selected and representative households. The FGD was conducted based on prepared checklists and semi-structured questionnaires. Four FGDs having 15 members per group were used to conduct the research. A total of 6 women participated in the discussion. During this session, respondents expressed their opinions, views, feelings, and perspectives. During field excursions kebele managers and natural resource management (NRM) experts were included based on their experiences in similar works. The data collection considers experts working in the watershed where the administrative boundary of the watershed lies within the four districts of the south Gondar zone. From these four districts namely Farta, Fogera, and Dera, experts were included for data collection. In the south Gondar zone, these districts were relatively food secured areas where watershed programs were not supported with productive safety net programs. Whereas, food-insecure districts were sponsored by watershed development programs based on the labor invested and mandatory for an individual being registered as food in secured. The research focused to understand the approach in these districts to identify whether there exists common interest, external forces and factors, that led to short-term benefits and long-term sustainability in the watersheds. All the data were collected and administered by the researchers for a day-long period at each watershed. The field observation  helped to better understand the various phenomena under investigation. Some of the observed occurrences were included an overview of the whole area of the watershed with transect walk at different slope locations, the level of natural resources degradation, private and communal grazing lands, water sources, and traditional water diversion ditches. The primary information was supported with secondary sources of information through reviewing published and unpublished documents. Thematic analysis was used to analyze the collected qualitative information. Therefore, the analysis was built most cohesively. During the discussion, an action research model was used. For conflicting ideas further detailed discussion and informal questions were forwarded to validate them especially from experts from kebele and district officers.