[Retracted] Climatic, Regional Land-Use Intensity, Landscape, and Local Variables Predicting Best the Occurrence and Distribution of Bee Community Diversity in Various Farmland Habitats in Uganda

Table 3

Richness and abundance of bees attracted to different habitats (land uses) frequently observed in the coffee-banana farming systems of central Uganda. Most attractive habitats were those with > 20% of weeds/crops/grass/wild plant species blooming at the time of visit.

Foraging habitats: seminatural habitat types/crop-field habitat types	Habitat frequency	Habitat size range	𝑁	Bee species; (X ± SE)	Bee abundance; (X ± SE)
Undisturbed field margins associated with or without termite mounds and shrubs/trees/grass species	125	0.05–0.5	159	22.45 ±3.45 a	223.80 ± 35.67 d
Field boundaries/hedgerows/track-sides	100	0.01–0.05	150	24.32 ±5.23 a	256.56 ± 25.78 d
Forest remnants/forest patches	35	0.5–40	45	16.54 ± 2.41 b	169.65 ± 19.21 d
Woodlots (pines, etc.)	20	0.5–40	30	14.22 ± 2.17 b	87.981 ± 11.23 f
Woodlots (Eucalyptus blooming)	15	0.5–30	22	5.291 ± 1.45 d	876.12 ± 125.7 b
Agroforestry tree species (Moringa sp., Sesbania sp., Ricinus communis,Leucaena leucocephala,Calliandra sp, Cassia sp., Sena spectabilis, etc.)	35	0.05–2	45	14.12 ± 2.31 b	259.54 ± 35.78 c
Grazed fields (Pasturelands), grasslands	115	0.05–10	190	13.76 ± 1.21 b	47.891 ± 7.782 f
Swampy (marshland) habitats, and (streams-sides)	54	0.3–5	89	21.45 ±2.31 a	211.45 ± 34.51 d
Abandoned gardens	45	0.02–1	258	14.23 ± 1.13 c	233.67 ± 15.67 d
Fallows (weedy, grassy, bushy, woody, grassy-bushy, bush-woody, scrubby)	122	0.03–3	356	15.21 ± 1.16 b	176.56 ± 13.76 d
Herbaceous crop-habitats(unweeded fields) and weedy harvested fields	135	0.02–3	367	19.23 ±1.27 a	284.56 ± 22.67 c
Perennial crops blooming (coffee)	106	0.05–12	600	19.54 ±1.61 a	876.23 ± 68.98 b
Perennial crops blooming (banana)	112	0.05–15	789	11.12 ± 1.93 c	251.12 ± 31.76 c
Perennial fruit crops blooming (avocado, mango, citrus, orange, lemon, tangerines, guava, papaya, etc.)	62	0.03–10	98	4.783 ± 1.32 e	1986.5 ±398.6 a
Perennial fruit crops blooming (passion fruits)	15	0.01–3	22	12.76 ± 1.93 c	76.456 ± 10.43 f
Home gardens of annual vegetable crop species blooming (pumpkin, watermelon, Amaranthus, Cleome, bitter Solanum, onion, cucumber, garlic, lettuce, etc.)	59	0.01–0.5	79	12.56 ± 1.35 c	367.65 ± 51.73 c
Annual commercial/cash vegetable crops (tomato, eggplant, chiles, pepper)	88	0.02–6	121	8.514 ± 1.25 d	34.125 ± 8.911 f
Annual commercial/cash crops (sim-sim, sun flower)	25	0.02–10	48	11.34 ± 2.11 c	113.32 ± 11.74 e
Annual cereals (maize) mixed with pulse crops (beans, cowpea, greengram, soybean, groundnut): maize blooming	91	0.04–7	421	2.125 ± 2.11 f	198.23 ± 23.56 d
Annual cereals (maize, sorghum, millet) mixed with pulse crops (beans, cowpea, greengram, soybean, groundnut): sorghum-millet blooming mixed crops	33	0.03–6	259	5.126 ± 1.45 d	59.546 ± 14.65 f
Annual cereals (maize, sorghum, millet) mixed with pulse crops (beans, soybean, groundnut): beans-cowpea blooming mixed crops	45	0.02–5	342	4.211 ± 1.13 e	19.176 ± 1.542 f
Annual cereals (maize, sorghum, millet) mixed with pulse crops (beans, soybean, groundnut): groundnut blooming	65	0.02–6	403	3.214 ± 0.91 e	14.548 ± 3.541 f
Bi-annual crops (cassava blooming)	71	0.02–8	198	1.897 ± 0.25 f	57.459 ± 15.21 f
Annual crops (sweet-potato blooming)	78	0.02–3	400	5.128 ± 1.27 d	157.67 ± 12.67 d
Annual cereals (rice blooming)	38	0.05–15	55	3.459 ± 0.87 e	21.235 ± 3.563 f
Annual crops (Irish potato blooming)		0.05–12	43	7.126 ± 1.41 d	28.659 ± 6.124 f

Habitat frequency = number of observation cases or number of times the habitat type was encountered across all 26 study sites and all sampling rounds.
Habitat size range (ha) = the data show the minimum and the maximum size of the type of habitat encountered during butterfly faunistic surveys.
𝑁 = number of samples (bees species and individuals) recorded in five sampling rounds across the 26 study sites in 2006.
Within columns, different letters show significant differences of the means at 𝑃=0.05 according to Tukey test performed after Kruskall-Walis ANOVA test indicating that the habitat type was significant (𝑃<0.01) for the number of species and individuals attracted.