The disease assessment was conducted in three faba bean-producing regions; the Oromia region (North Shewa, West Shewa, Southwest Shewa, Arsi, and Finfinne Special Zone), the Amhara (South Gondar, North Shewa, and South Wollo), and Tigray (South Tigray). Faba bean production potential was taken into account when selecting the regions and zones. Mainly the study areas have differed in altitude, and weather conditions) (Rainfall, Temperature, and relative humidity) (Table 1). The rainfall data of the surveyed districts were obtained from meteorological stations from planting to survey periods during the surveyed year. FBG and other diseases have been assessed from three zones in the 2019 and 2020 main cropping seasons in the Amahara region, from three zones in 2019 and, five zones in 2020 in the Oromia region, and one zone in the Tigray region in 2019 and 2020.

The sampling structure consisted of three hierarchal levels of faba bean-growing agroecology in Ethiopia. The Amhara, Oromia, and Tigray regions are Ethiopia’s highlands of potential faba bean production areas. Within these three regions of faba bean-growing highlands, six zones from the Oromia region and three zones from the Amhara region, and one zone from the Tigray region were selected by purposive sampling. The number of districts from each representative zone was also selected by purposively sampling methods based on production potential, accessibility, and area coverage. A total of 999 faba bean fields (283 in 2019 and 716 fields in the 2020 cropping season) across 50 districts within nine zones in the three regions were assessed in main cropping seasons. In N/Shewa (Oromia) 9 districts (n = 137), W/shewa 10 districts (n = 104), S/W/shewa 6 districts (n = 78), Arsi 7 districts (n = 91), Special zone 2 district (n = 16), N/Shewa (Amhara region) 6 districts (n = 160), S/Gondar 4 district (n = 152), S/wollo 4 district (n = 171), and S/Tigray 3 districts (n = 90) were selected during assessments. The FBG and other faba bean disease assessments from each locality in the districts were selected randomly. Generally, from the assessed faba bean fields, 483 fields were across 13 districts, 426 fields across 34 districts, and 90 fields across 3 districts of Amhara, Oromia, and Tigray regional states, respectively (Table 1).

Field surveys were made at two growth stages, i.e., first stage survey (seedling to flowering growth stage) at end of July, and the late-stage survey (podding to maturity growth stage) was conducted at end of September to early October. Each survey site was coordinated with the Global Positioning System (GPS) for mapping disease distribution in Ethiopia. The distribution map of FBG in three regions (Amahara, Oromia, and Tigray) was developed representing the highlands of faba bean-producing areas in Ethiopia.

Identifications were conducted based on previously described and reported symptomology of FBG disease by Hailu et al. (2014) and Bitew et al. (2022) to confirm the typical symptoms of FBG visualized in the faba bean plant and other host species during the assessment periods. The FBG symptoms observed on other host species were confirmed by observing key morphological characteristics such as typical Physoderma zoosporangia, and zoospore structures under a light microscope as described by You et al. (2021).

Disease assessments were studied by observing fields of faba bean and sampling by following available roads and routes in the selected location with assessment stops at 3-5 km intervals based on vehicle odometer. Five points with a 10 m distance among the quadrants from each faba bean farm were received by throwing a quadrant area of 1 m² in a walking “X” pattern on the farm, and the points were averaged and noted on the datasheet for incidence and severity per farm. Disease incidences were computed as diseased faba bean plants with shows symptoms to a total of faba bean plants examined in the quadrants. Disease severity of FBG was measured based on 15 faba bean plants randomly selected using a modified 0-9 scoring scale where, 0 = no symptom, 1 = very small and few green lesions on the leaves, 2 = very small and green sunken lesions, 3 = many green sunken small lesions, 4 = many small lesions and few large lesions turning into brown color, 5 = many brown color and large lesions, 6 = brown lesions coalescing, 7 = brown large lesions coalescing, 8 = plants darkened and stem collapsed, and 9 = dead plants (Ding et al., 1993).

Ascochyta blight disease symptoms on faba bean plants using the 1-9 disease scoring scales as stated by Hanounik et al. (1987), Chocolate spot severity disease using standard disease scales of 0-9 by Bernier, and Faba bean rust disease using the 1-9 disease scoring scale as stated by International Center for Agricultural Research in the Dry Areas (1986), and faba bean root rot disease as an area of root tissue damaged using a 0-5 rating scale as McFadden et al. (1989) was used to score disease severity on 15 faba bean plant selected in the quadrants.

Disease prevalence (DP), DI, and Percent Severity Index of the FBG and other faba bean diseases were computed based on the following formulas expressed in three separate percent disease indices.

The incidence and severity of FBG were classified into discrete binomial qualitative data groups as previously described by Nayak et al. (2018). As a result, several fields recorded with FBG showing ≤35% and >35% incidence, and ≤25% and >25% severity were used as class boundaries to generate a binary dependent variable (Table 2). To depict the bivariate distribution of the fields, contingency tables of DIs, prevalence, severity, and independent variables were created (Table 2). The relationships of Ethiopia’s highlands of potential faba bean production areas, zones, crop growth stage, the density of crop, cropping system, field slope, cropping pattern, altitude, the density of weeds, type of soil, application of fertilizer, method of drainage, nearby crop type, and ±fungicide spray(s) were investigated using logistic regression by the SAS procedure of GENMOD (Yuen et al., 1996).

The logistic regression model enables the assessment of the significance of several independent variables that influence the response variable (Fininsa and Yuen, 2001). The response variable was the likelihood of FBG incidence exceeding 35% and severity exceeding 25% in a given faba bean field.

The significance of the independent variables was investigated in two methods (Belete et al., 2013). First, the relationship between all independent variables with incidence and severity in a single variable model was considered. This entailed testing the deviation reduction assigned to a variable when that variable was originally introduced into the model. Second, the relationship of an independent variable with FBG incidence or severity was assessed when it was added to the model with all other independent variables. Finally, in the reduced multiple variable models, selected independent variables that have a significant association with FBG incidence or severity when entered first and last into a model were incorporated.

The parameter estimates and standard errors were examined in both single and multiple models using the GENMOD function in SAS. Fininsa and Yuen (2001) demonstrated how to create a deviation table for the final reduced multiple variable models. The likelihood ratio test was used to evaluate the importance of the variables and was compared to the Chi-square value when variables were added to the reduced model (McCullagh and Nelder, 1989). Analysis of deviation, odd ratio, and standard error of new variables in a reduced model demonstrated the role of independent variables and variable classes within the independent variables in determining the importance of FBG incidence.

The gall-like symptoms were detected during the assessment periods on above-ground faba bean plant parts on leaves, petioles, stems, and pods as described in Fig. 2A-F. The early infection of FBG symptoms with very small tumor-like swelling (normal green) symptoms was observed on newly emerged faba bean leaves (Fig. 2A). The medium-sized tumor-like swelling (brownish color) on the underside leaves and with depression on the upper leaf surface with mostly uninfected leaf green areas was observed during assessment in the faba bean field (Fig. 2B). The most of faba bean fields assessed were severely infected with irregular, coalesce, and large necrotic galled areas on petiole, leaves, stems, and pods (Fig. 2C-F).

While the faba bean is the major host most severely attacked by FBG, the gall symptoms were also found on other legume crops such as field pea (Pissum sativum) on leaves and stems (Fig. 3). The FBG was also observed attacking the Trifolium spp. and other legume plant species like Vicia sativa (Vetch) and Medicago scutellata.

The typical symptoms of FBG were observed as similar as indicated in Fig. 2. The FBG symptoms observed on other host species were confirmed by observing key morphological characteristics such as typical Physoderma zoosporangia, and zoospore structures under a light microscope.

The overall FBG disease across regions (Amhara, Oromia, and Tigray) and cropping seasons (2019 and 2020) recorded 76.2% prevalence. The prevalence of FBG for the 2019 and 2020 surveyed seasons was recorded at 72.8% and 78.5% respectively (Table 3). The FBG disease was most prevalent in the Amhara region (95.7%), followed by Tigray (83.3%), and less in the Oromia region (54%). The highest prevalence of FBG disease was recorded from the South Gondar zone (98%) and North Shewa (95%), followed by South Wollo (94.2%) zones. The 83.9% and 83.3% of FBG prevalence were obtained from the North Shewa zone of Oromia and the South Tigray zone respectively. The minimum FBG DP was observed in Arsi (33%), West Shewa (14.5%), and Southwest Shewa (10.8%) zones in the Oromia region.

In the faba bean fields where the FBG was assessed, DI and severity were influenced by different biophysical factors such as crop growth stage, crop density, cropping system, altitude, weed density, application of fertilizer, method of drainage, nearby crop, and application of fungicide (Table 3). The lowest incidence (8.6%) and severity (3.6%) were obtained from the faba bean field practiced crop rotation with non-leguminous crops (cereal and vegetable crops). In contrast, 58.2% incidences and 22.07 severity percentage of FBG disease were recorded from the faba bean field with a mono-cropping system. The highest DI (81.3%) and severity (29.8%) were recorded from an independent variable class of maturity growth stage, whereas the lowest incidence (45.5%) and severity (15.07%) were received from the seedling growth stage. The higher mean incidence (69.3%) and severity (25.7%) were observed in faba bean fields with infested with higher weed density as compared to fields with low weed density of incidence (44.7%) and severity (17.6%). Faba bean fields situated at high altitudes (>2,400 m.a.s.l) had the highest FBG incidence (74.1%) and severity (28.5%) as compared to areas with medium altitudes (2,201-2,399 m.a.s.l) and lower altitudes ranged (1,900-2,200 m.a.s.l.) (Table 4). Inorganic (NPS: nitrogen, phosphate, and sulfur with the ratio of 19% N, 38% P₂O₅, and 7% S) fertilizer application to faba bean fields at sowing time was found associated with the lowest incidence (51.4%) and severity (16%) of FBG disease as compared to faba bean fields that did not receive any type of fertilizer. Faba bean fields near legume crops had the highest DI (74.1%) and severity (26.1%), whereas faba bean fields near cereal crops had the lowest incidence and severity (55.06% and 21.3%, respectively). Faba bean fields sprayed with fungicide had the lowest incidence (46.7%) and severity (9.3%) of FBG compared to faba bean fields that did not receive any fungicide (67.6% incidence, 33.2% severity) (Table 3).

The distribution map of FBG disease in Ethiopia through the Amahara, Oromia, and Tigray regions was created to depict the highlands of faba bean-producing areas. In general, the total FBG DI varied from 0% to 100% in all examined faba bean fields with the highest distribution range from 61% to 100% (Fig. 4).

The combined means for years 2019 and 2020 assessments for prevalence, incidence, and severity of chocolate spot, Ascochyta blight, faba bean rust, and root rot diseases found in various zones in the three surveyed regions are summarized in Table 4. Chocolate spot is more prevalent and was more common in the West Shewa (100%), Finfinne Special Zone (100%), and North Shewa (99%) of the Oromia region, followed by South Gonder (98%) and North Shewa (95%) of Amahara region. Ascochyta blight prevalence was found in Finfinne Special Zone (100%), North Shewa (98%), West Shewa (98%), and Southwest Shewa (97%), followed by the Arsi zone (91%) of the Oromia region. Except for faba bean rust in South Wollo, no faba bean rust or root rot disease was detected in the Amhara region. The incidence of major faba bean diseases tended to follow a similar pattern as found for DP (Table 4).

When independent variables such as the assessed region, zone, year, crop growth stage, cropping system, altitude, weed density, and fungicide application are entered first into the logistic regression model, they are strongly associated with FBG incidence and severity. When slope, soil type, drainage methods, and crop nearby were entered first into the logistic model, they exhibited no association with FBG incidence and severity (Table 5).

The independent variables of surveyed regions, surveyed year, zones, crop growth stage, altitude, crop density, weed densities, cropping pattern, cropping system, fertilizer application, crop nearby, and fungicide spray, when variable entered first (VEF) into the model for FBG incidence, the highest deviance reduction, and best probability values, were observed. Similarly, when the variable entered last (VEL) into the model for FBG incidence the surveyed region, surveyed year, zones, altitude, weed density, cropping system, and soil type, fungicide application showed the greatest deviance reduction and best probability values. The surveyed region, surveyed year, zones, crop growth stage, crop density, cropping system, the slope of the field, cropping pattern, altitude, weed density, fertilizer application, and fungicide application when VEF into the model for FBG severity showed the greatest deviance reduction and best probability values. In addition, the surveyed region, surveyed year, zones, crop growth stage, crop density, cropping system, the slope of the field, cropping pattern, altitude, weed density, and fungicide application when entered last into the model for FBG severity showed the greatest deviance reduction and best probability values (Table 5).

In reduced multiple variable models, clustered variables of surveyed regions, year, zones, crop growth stage, crop density, cropping system, field slope, cropping pattern, altitude, weed density, soil type, fertilizer application, and fungicide were statistically tested as presented in Table 6.

The analysis results of deviance for the variables added one by one to the reduced model showed the importance of each variable and variable class. The deviance value, parameter estimates, and standard error of new variable classes in a reduced model have resulted in FBG DI and severity are given in Tables 6 and 7. The highest FBG incidence recorded from the assessed faba bean field has a high probability of association with the surveyed year 2020, Amhara region, South Gondar and North Shewa zone (Amhara), maturity growth stage, mono-cropping, high altitude (>2,400 m.a.s.l), high weed infestation, and unsprayed fungicide (Table 6). Whereas, the surveyed year 2019, Oromia region, West Shew, S/W/Shewa zone, seedling growth stage, lower-altitude (1,900-2,200 m.a.s.l.), crop rotation, low weed infestation (>100/m²) and sprayed fungicide had associated with the lowest FBG incidence (Table 6). Tigray region, 2020 surveyed year, North Shewa (Oromia), maturity growth stage, mono-cropping system, high altitude (>2,400 m.a.s.l.), high weed density (>100/m²), and unsprayed fungicide were associated with the highest FBG severity (Table 7). The lowest FBG severity assessed in the surveyed year of 2019, Oromia region, Southwest Shewa, and Arsi zone, seedling growth stage, crop-rotation system, low altitude ranges (1,900-2,200), low weed densities, sprayed fungicide of the faba bean field had shown a high likelihood of association (Table 7).

FBG was the most prevalent in faba bean production areas, particularly in the South Gondar, South Wollo, and North Shewa of the Amhara region, North Shewa zone of the Oromia region, and South Tigray zone of Tigray region. Furthermore, in Ambo, Jeldu, Dendi, Chelia, and Midakegn districts from the West Shewa zone, and Chitu district from the Southwest Shewa zone of the Oromia region highland areas was reported from present study as newly infested faba bean production fields with FBG disease. A logistic regression model highlighted the relationship between the different independent factors and FBG incidence and severity during the assessment periods of 2019 and 2020. The independent variable classes of the crop at podding and maturity stages, mono-cropping, altitudes greater than 2,400 m.a.s.l., high weed density, and unsprayed faba bean fields with fungicide as being significantly associated with the highest disease intensity, indicating that these are likely important factors for FBG epidemics. Crop rotation; faba bean production at altitude ranges from 1,900 to 2,200 m.a.s.l., and from 2,201 to 2,399 m.a.s.l.; low weed infestation in faba bean field; fungicide application can all be considered as management opportunities towards limiting the adverse FBG impact on faba bean. Whereas FBG may overriding cause of devastating faba bean yield losses, it remains important to also monitor and manage the other faba bean diseases to minimize overall losses from diseases. Finally, further investigation of disease biology and ecology, active and passive FBG pathogen inoculum transmissions, defining the diversity of Ethiopian FBG pathogen isolates, and FBG interactions with other co-occurring pathogens could be important foci for future studies.