Flowering agricultural landscapes enhance parasitoid biological control to Bemisia tabaci on tomato in south China.

Agricultural landscape pattern may enhance biocontrol services by supporting parasitoid populations, including parasitoids of Bemisia tabaci Gennadius (Hemiptera: Aleyrodidae). In this study, we selected four landscape types in Yunnan province, in south China, which were characterized by flower fields, mountain, river, and urban areas as their main elements. We then carried out systematic surveys in tomato fields found within each landscape type, to determine the diversity, occurrence, and parasitism rate of parasitoids. We found that parasitoids from the genus Encarsia and Eretmocerus were the main natural enemies present, and the most abundant species were recorded in the flower and the mountain landscapes. Also, Encarsia formosa (Hymenoptera: Chalcidoidea) formed the highest relative abundance regardless of the landscape type. We also found that the population density of B. tabaci in flower landscapes was the lowest than that in other landscape types. During the main activity period of B. tabaci, the parasitism rate in the flower landscape was higher than that in other landscape types. Our findings showed that the occurrence of B. tabaci was relatively mild in the flower landscapes. The landscape type was also beneficial to parasitoids as it enhanced their parasitism rate. Therefore, agriculture environmental schemes should consider increasing the size of flower fields in the surrounding landscape to enhance the sustainable control of B. tabaci by the natural agricultural ecosystem.

Introduction

The whitefly, Bemisia tabaci Gennadius (Hemiptera: Aleyrodidae) is one of the most economically and agriculturally important insect pest worldwide [1]. It is a polyphagous species, with more than 500 plant species including tomato, cucumber, and other vegetables identified as it hosts [2, 3]. The pest cause serious economic damage not only by direct feeding but also by transmitting plant viruses such as Begomovirus, Carlavirus, Crinivirus, Ipomovirus, and Torradovirus [4]. The rapid expansion of whitefly populations promotes fast and efficient viral transmissions [5-7].

As an environmental-friendly control method, the use of parasitoids to control B. tabaci is not uncommon, in areas where consumers desire reduced use or complete elimination of pesticides from their food [8]. Aphelinid parasitoids, particularly Encarsia formosa and Eretmocerus mundus (Hymenoptera: Chalcidoidea), have outstanding records of successes in biological control against B. tabaci in many countries [9-11]. However, agricultural landscapes in China have changed from a complex pattern with a large proportion of natural habitats, to a simple landscape with a large proportion of arable land, which has dramatically changed the arrangement of arable and non-crop habitats [12, 13]. Agricultural landscape pattern can affect the interaction between pests and parasitoids [14]. First, the parasitism rate of parasitoids in agricultural landscape decreases with the increase in the proportion of crop farmland, because management does not aim to ensure overwintering and reproduction of the parasitoids [15]. Second, an increase in the proportion of non-crop habitats enhances the parasitism rate, because non-crop habitats provide wintering habitat and alternative hosts and food for parasitoids [16]. Third, in a diversified landscape, flowering plants in non-crop habitats can provide food sources such as pollen and nectar for parasitoids [17]. Structurally-complex landscapes can enhance biological control as a result of the high proportion of non-crop habitats, such as wooded mountain, grassland, hedgerows, and rivers in these landscapes [18]. These non-crop habitats may enhance species diversity or the abundance of insect parasitoids in nearby arable lands, thus improving the natural control of pests by effectively reducing their population [19, 20].

In our previous study, we found that although the parasitoids of B. tabaci could be observed in the tomato planting fields, the number of species and black pupae of the wasps were significantly difference under different agricultural landscapes around Kunming, Yunnan Province (S1 Data). Therefore, we hypothesized that the different agricultural landscapes may affect the species and control effect of parasitoids against whitefly. Because of the supply of food, shelter, alternative hosts, and favorable microclimates, parasitoids could benefit from natural field habitats around agricultural landscapes [21, 22]. Yunnan province in China is one of the regions with the richest biodiversity in the world [23, 24]. However, due to the intensification of human activities, the landscape in this province has changed significantly, which has tilted the original ecological balance and promoted the outbreak of the whitefly [25]. In this 2-year study, we selected four typical landscape types characterized with flowers, mountains, rivers, and urban areas as their main elements, in central Yunnan, where agricultural activities are most frequent. The focus of this study on the colonization of whitefly parasitoids in the different landscape types is considered critical for increasing the diversity of parasitoids to effectively suppress the pest population [20]. The study had two aims; the first was to assess the landscape types that best protect the abundance, occurrence, and enhance parasitism rate of parasitoids in these four typical landscape types. The second was to assess the variability of biocontrol services associated with the relationship between tomato growth stage, B. tabaci occurrence periods and parasitism efficiency (plot and landscape level) and also between time periods. Therefore, the overall aim was to identify which landscape type was most favorable to protect parasitoids and enhance their effective control to B. tabaci.

Study area

The study was conducted within a radius of 0.5 km in agriculture landscapes around each of 12 tomato field plots (20 m×40 m), which were located in the surroundings of Kunming, south China (24°42’45’’N-25°22’43’’N, 102°22’18’’E-103°10’90’’E). The use of Google Earth Profession and field inspections (ground-truthing) were used to determine the land cover types [20, 26, 27]. A principal component analysis (PCA) was performed to reduce the dimensions of the landscape data. Ten land cover types were identified for the PCA analysis, the land cover type with the largest area and the absolute value of first principal component greater than 0.9 was selected as the landscape type: (1) flower fields, (2) river (rivers, lakes, reservoirs etc.), (3) mountain (forest with altitude difference more than 150 m), (4) urban areas, (5) vegetable fields, (6) fruit trees, (7) trees (windbreaks, border trees or ornamental trees), (8) bushes, (9) grasslands, and (10) wastelands. The altitude difference among the landscape types was within 20 m, except for mountains. Principal component axes were extracted using correlations among the landscape types, and the resulting factors were not rotated.

Flower landscape type: three of them were divided into flower landscape type, their main landscape cover types were flower fields, which evenly distributed in the agricultural landscape types, and the main types of flowers were Rosa chinensis, Dianthus caryophyllus, Myosotis sylvatica and Eustoma grandiflorum.

River landscape type: three of them were divided into river landscape type, their main landscape cover types were rivers. The Panlong river across these three landscapes, the main type of tree plant was Eucalyptus robusta, and some fruits such as Vitis vinifera and Malus domestica and vegetables such as Cucumis sativus were planted here. The purpose of setting up the river landscape type was to pay attention to the high humidity environment caused by the river.

Mountain landscape type: three of them were divided into mountain landscape type, their main landscape cover types were cypress forest with altitude difference more than 150 m. There was abundant vegetation, the main types of tree plants were Sabina chinensis and Pinus yunnanensis, the main types of shrubby plants were Cotoneaster microphyllus and Pyracantha fortuneana, and the main types of herbaceous plants were Imperata cylindrica, Polystichum acutidens and Cymbidium elegans.

Urban landscape type: three of them were divided into urban landscape type. These landscapes were close to the town and their main landscape cover types were buildings. There were few vegetation species in this landscape type, and only some fruits such as V. vinifera and M. domestica and vegetables such as C. sativus and other vegetation are scattered.

Each plot was at least 5 km apart to avoid potential interactions of the insect populations from different sites. The study was carried out in tomato planting fields both in 2018 and 2019 under the different landscape types. The cultivar planted was tomato cv. ‘Zhongyan TV1’ (Lycopersicon esculentum Mill.). Tomato seedlings were first nursed in June and harvested in October. No plot was treated against pests or diseases in our experiments. Plots were kept weed-free by manual weeding when necessary.

Sampling

Population dynamics

The population dynamics of B. tabaci and its parasitoids was determined by sampling in each 800m2 plot (20 m×40 m) of tomato planting field in each landscape. The first sampling started 10 days after tomato transplanting. Five yellow sticky traps (20 cm×25 cm) at each sampling point were hanged in the fields and collected every 10 days using the same five point sampling method until the end of the growing season, the five yellow sticky traps were positioned following the planting line of tomato, and the height increased with the growth of tomato. The yellow sticky traps were taken back to the laboratory to count the number of individuals of B. tabaci and its parasitoids. The averages of the total numbers of B. tabaci and its parasitoids on each yellow sticky trap were calculated as their respective population densities. The adult parasitoids were collected for species identification. The growth period of tomato was recorded at each survey. This was divided into seedling, anthesis, fruit expansion, and harvest periods [28, 29].

Relative abundance of parasitoids

The number of individual species in the community obtained from each yellow sticky trap was used as the basis for data analysis, the relative abundance of each parasitoid species was calculated as the proportion [30].

Parasitism rate

Parasitism rates were evaluated in the same tomato planting field alongside the population dynamics. The first sampling started 10 days after tomato transplanting. Leaves with nymphs of B. tabaci were collected every 10 days using the five point sampling method until the end of the growing season. In each sampling point, five tomato plants were sampled (avoiding the plants closest to any edge to minimise edge effects). On each tomato plant, five leaves of similar age at the upper, middle, and lower positions were examined, giving a total of 375 monitored leaves per field. Leaves were transported to the laboratory to count nymphs, after which they were placed in a Petri dish with agar and kept in an artificial climate box (Shanghai Boxun, BIC-400, T = 25°C, rH = 65%, L/D = 14h/10h). The parasitism rate was calculated after eclosion of B. tabaci and parasitoids based on the formula [31, 32]: where P is the percentage of parasitism, P is the number of parasitoids, and W is the number of B. tabaci.

Description of seasonal activity

The seasonal activity curve was standardized following the quartile method of Fazekas et al. and Zhang et al. [25, 33]. This method divides the seasonal activity into three periods: early, main, and late, and formally identifies the start and end of each of these, as well as the date of the seasonal activity peak. First, the numbers observed are summed and the three cardinal points are the dates when 25, 50 and 75% of the total densities are reached. These also divide the curve into four segments. The start of the main activity period corresponds to the date when the cumulative densities reach 25% of the total (the start of the second quartile on the vertical axis), and the end corresponds to the date when 75% (the end of the third quartile on the vertical axis) is reached. The early activity period was defined formally as from the start of the census to the beginning of the main activity period, and the likewise formalized late activity period was defined formally as from the end of the main activity period until the end of the observations, when activity stopped.

Data analysis

The population densities of B. tabaci and E. formosa adults (the average quantity of adults in each yellow sticky trap), as well as the relative abundance of parasitoids and the parasitism rate of E. formosa on B. tabaci, were analyzed by using a one-way analysis of variance (ANOVA) after tests of normality (Shapiro–Wilk) and homoscedasticity (Bartlett), with agriculture landscape types as the main effect. The data have been logarithmic transformed if it did not follow a normal distribution. To reduce the impact of occurrence time on the population densities of B. tabaci and its parasitoids, the activity period of B. tabaci and its parasitoids were divided into early, main, and late activity period by quartile method. The least significant differences (LSD) were determined using data from one sampling conducted in the same activity periods, in the same agriculture landscape types as one replicate. A significance level of P = 0.05 was used for all tests. Data analyses were performed using SPSS 20.0. The figures of cumulative seasonal activity curves and population dynamics were made with Origin 2018.

Results

Interpretation of principal components

At each landscape, we were able to interpret the first three principal components which accounted for about 85% of the variation in the landscape variables. We divided the 12 landscapes into four types according to the interpretation of principal components. Three of these which had the highest eigenvector of principal component 1 of flower fields (2018: 0.963, 0.950, and 0.913, respectively; 2019: 0.940, 0.985, and 0.944, respectively), were grouped as flower landscape type. Another three which had the highest eigenvector of principal component 1 of river (2018: -0.956, -0.992, and 0.938, respectively; 2019: -0.970, -0.992, and -0.921, respectively), were grouped as river landscape type. The next three which had the highest eigenvector of principal component 1 of mountain (2018: -0.945, 0.966, and 0.915, respectively; 2019: 0.915, 0.900, and -0.920, respectively) were grouped as mountain landscape type. The last three which had the highest eigenvector of principal component 1 of urban (2018: 0.946, -0.906, and 0.927, respectively; 2019: -0.907, 0.911, and -0.913, respectively) were grouped as urban landscape type (Table 1).

Table 1

Principal component loading diagrams examining the landscape variables at four landscape types.

Planting years	Landscapes	Eigenvector of Principal Component 1
		flower fields	river	mountains	urbans	vegetable fields	fruit trees	trees	bushes	grasslands	wastelands
2018	1	0.96	0.23	0.04	0.78	-0.67	0.32	-0.40	-0.41	0.80	-0.86
	2	0.95	-0.31	-0.55	0.87	0.58	0.04	0.61	-0.38	-0.06	-0.85
	3	0.91	-0.33	0.34	0.07	0.24	-0.80	0.89	0.57	0.74	-0.84
	4	0.77	-0.96	0.42	0.36	-0.82	0.71	0.29	0.48	-0.83	0.20
	5	0.78	-0.99	-0.44	-0.46	-0.62	0.80	0.74	0.38	-0.48	0.89
	6	-0.89	0.94	0.66	0.73	-0.86	0.39	0.29	-0.05	0.53	-0.79
	7	-0.43	-0.43	-0.95	0.78	0.86	0.88	-0.06	0.27	0.29	-0.51
	8	0.64	0.82	0.97	-0.31	0.49	0.18	-0.46	0.87	-0.10	-0.49
	9	0.43	-0.46	0.92	0.16	0.80	0.75	0.22	-0.55	-0.22	-0.88
	10	0.74	0.89	-0.52	0.95	-0.52	-0.68	0.83	0.63	-0.34	-0.87
	11	0.24	-0.60	0.76	-0.91	0.72	-0.80	0.83	-0.87	-0.22	0.89
	12	0.81	-0.41	0.77	0.93	-0.33	0.20	0.61	-0.71	0.19	-0.89
2019	1	0.94	0.16	0.64	0.60	-0.14	0.87	0.87	0.74	-0.73	-0.84
	2	0.99	0.40	0.25	0.49	-0.40	-0.22	0.68	0.76	-0.74	-0.79
	3	0.94	0.71	-0.53	0.14	-0.53	-0.53	0.41	0.70	-0.77	-0.53
	4	0.87	-0.97	0.82	-0.70	-0.48	0.73	0.62	0.76	-0.75	0.88
	5	0.89	-0.99	-0.63	-0.54	-0.50	0.80	0.68	0.84	-0.62	0.86
	6	0.87	-0.92	0.71	-0.67	-0.75	0.81	0.73	0.72	-0.70	0.86
	7	-0.39	0.32	0.92	0.86	0.69	0.50	0.22	0.33	-0.72	-0.86
	8	-0.45	-0.84	0.90	0.70	0.49	0.46	0.66	0.35	-0.86	-0.71
	9	-0.49	-0.32	-0.92	0.81	0.47	0.21	0.42	-0.16	-0.75	0.88
	10	-0.39	-0.54	0.65	-0.91	0.65	-0.87	0.83	-0.76	0.88	0.89
	11	0.26	0.86	-0.68	0.91	-0.66	0.86	-0.83	0.74	-0.32	-0.89
	12	0.85	-0.62	0.72	-0.91	0.73	-0.83	0.81	-0.87	-0.25	0.89

Parasitoid species of Bemisia tabaci in different types of landscapes

More than 150,000 parasitoids were collected in the sampled landscapes; they belonged to the genus Encarsia and Eretmocerus. The number of parasitoid species in the flower and the mountain landscapes were higher than those in the river and the urban landscapes. E. formosa accounted for the highest relative abundance in the four landscape types regardless of the tomato planting years (flower landscape in 2018: F = 43.0190; df = 5, 17; P = 0.0001. river landscape in 2018: F = 266.1050; df = 5, 17; P = 0.0001. mountain landscape in 2018: F = 63.4520; df = 5, 17; P = 0.0001. urban landscape in 2018: F = 151.6260; df = 5, 17; P = 0.0001. flower landscape in 2019: F = 144.3720; df = 5, 17; P = 0.0001. river landscape in 2019: F = 901.3930; df = 5, 17; P = 0.0001. mountain landscape in 2019: F = 303.2620; df = 5, 17; P = 0.0001. urban landscape in 2019: F = 913.9680; df = 5, 17; P = 0.0001), indicating that E. formosa was the main parasitoid in the four landscape types (Table 2).

Table 2

Parasitoid species of Bemisia tabaci in different types of landscapes in Kunming, south China.

Planting years	Parasitoid species	Relative abundance in different landscape types (%)
		Flower landscape	River landscape	Mountain landscape	Urban landscape
2018	Encarsia formosa	54.58±3.11a	75.40±3.14a	63.57±6.05a	76.84±4.25a
	Encarsia sophia	12.80±2.36b	13.49±1.51b	10.97±1.99b	13.97±3.46b
	Encarsia sp1	11.71±5.45b	4.40±2.56cd	8.97±2.45b	8.40±2.33b
	Eretmocerus hayati	4.60±0.69b	5.63±0.44c	4.69±0.96b	0.00±0.00c
	Eretmocerus sp1	7.04±0.70b	0.00±0.00d	7.00±1.62b	0.00±0.00c
	Other species	9.27±1.86b	1.09±0.56cd	4.80±0.63b	0.79±0.48c
2019	Encarsia formosa	57.58±2.41a	79.37±1.41a	71.92±1.15a	84.30±2.08a
	Encarsia sophia	12.94±1.67b	12.87±1.22b	10.37±2.67b	10.68±1.36b
	Encarsia sp1	11.61±2.49b	4.09±1.46c	5.53±1.19c	3.76±0.88c
	Eretmocerus hayati	8.27±0.78bc	1.41±0.70cd	2.89±0.55c	0.00±0.00d
	Eretmocerus sp1	3.32±0.78c	0.00±0.00d	4.14±1.41c	0.00±0.00d
	Other species	6.29±1.10c	2.26±0.55cd	5.15±1.57c	1.25±0.59cd

Data in the table are mean ± SE. The different lowercases indicate significantly different at the 0.05 level with the different parasitoid species in the same type of landscapes during the same tomato’s planting year.

Seasonal activity of Bemisia tabaci and Encarsia formosa

The length of the main activity periods of B. tabaci in 2018 ranged from 21 days (flower landscape and river landscape) to 23 days (mountain landscape and urban landscape) and the peak activity was reached on 3rd September (flower landscape), 28th August (river landscape), 18th September (mountain landscape) and 19th October (urban landscape). The length of the main activity periods of B. tabaci was not significantly different between years, and it ranged from 21 days (urban landscape) to 31 days (mountain landscape) in 2019. The peak activity was reached on early September to mid-September. For E. formosa, the length of the main activity periods in 2018 was 47 days in the urban landscape and about 30 days in other landscape types. The peak activity was reached on 9th September (flower landscape), 12th September (river landscape), 23rd September (mountain landscape) and 19th October (urban landscape). The length of the main activity periods in 2019 was 22 days in the flower landscape and about 30 days in the other landscape types. The peak activity was reached on 10th September (flower landscape and river landscape) and 20th September (mountain landscape and urban landscape). The main activity period of B. tabaci and E. formosa encompassed the anthesis to fruit expansion of tomato in both years of study (Table 3).

Table 3

Main activity periods and peak activity dates of Bemisia tabaci adults and Encarsia formosa in different types of landscapes in Kunming, south China.

Planting years	Landscape types	Bemisia tabaci		Encarsia formosa
		Main activity period (duration in days)	Peak activity date	Main activity period (duration in days)	Peak activity date
2018	Flower landscape	21 Aug.-10 Sep. (21)	03 Sep.	21 Aug.-19 Sep. (30)	09 Sep.
	River landscape	21 Aug.-10 Sep. (21)	28 Aug.	30 Aug.-28 Sep. (30)	12 Sep.
	Mountain landscape	07–29 Sep. (23)	18 Sep.	07 Sep.-09 Oct. (31)	23 Sep.
	Urban landscape	08–30 Oct. (23)	19 Oct.	27 Sep.-13 Nov. (47)	19 Oct.
2019	Flower landscape	20 Aug.-10 Sep. (22)	03 Sep.	30 Aug.-20 Sep. (22)	10 Sep.
	River landscape	20 Aug.-10 Sep. (22)	01 Sep.	30 Aug.-30 Sep. (32)	10 Sep.
	Mountain landscape	30 Aug.-20 Sep. (31)	10 Sep.	30 Aug.-30 Sep. (32)	20 Sep.
	Urban landscape	10–30 Sep. (21)	16 Sep.	30 Aug.-30 Sep. (32)	20 Sep.

Population dynamics of Bemisia tabaci and Encarsia formosa

In the flower landscape, the population densities of B. tabaci and E. formosa were relatively low in the first survey in 2018, then they all gradually increased. The population density of B. tabaci fluctuated after late July, increased at a sharp rate in late August and peaked in early September (female: 70.80 per. yellow sticky trap; male: 32.40 per. yellow sticky trap). The population density of E. formosa increased from the beginning of the survey to late August, then increased to the maximum population density in late September after a slight decline (39.40 per. yellow sticky trap). During the survey of 2019, the population of B. tabaci maintained a relatively low trend till the mid-August, and then the density of female adults began to increase gradually and peaked to 68.20 per. yellow sticky trap in early September. The density of male adults maintained a relatively flat trend till the end of the sample date. During the same sampling period, the population of E. formosa also maintained a relatively low trend till the middle august, then increased gradually and peaked to 35.73 per. yellow sticky trap in late September and then decreased till the end of the sample date (Fig 1).

Fig 1

Population dynamics of Bemisia tabaci adults and Encarsia formosa (mean±SE) in different types of landscapes in Kunming, south China.

In the river landscape, the population density of B. tabaci increased from early July to mid-July and then increased sharply from late July after a brief decrease. The density of female adults peaked to 334.13 per. yellow sticky trap in mid-September, and the density of male adults peaked to 231.20 per. yellow sticky trap in late August. After that, the population declined sharply till the end of the investigation in 2018. From the beginning of the survey, the population of E. formosa kept a slow growth trend till mid-early September, then decreased a little till late September. It began to rise again, peaked to 36.07 per. yellow sticky trap in early October, and then declined till the end of the sample date in 2018. During the survey in 2019, the population density of B. tabaci began to increase gradually from early July to late July, then they increased sharply and peaked in early September (female: 301.60 per. yellow sticky trap; male: 182.13 per. yellow sticky trap). After that, the population density of B. tabaci decreased sharply till the end of the sample date. The population of E. formosa began to increase gradually from early July and peaked to 38.27 per. yellow sticky trap in mid-August, then they slowly declined to mid-September and rise again to 34.33 per. yellow sticky trap in mid-October and then slowly declined till the end of the sample date (Fig 1).

In the mountain landscape, the population density of B. tabaci increased slowly from mid-July till mid-late August, reached its first peak in early September (female: 118.53 per. yellow sticky trap; male: 60.13 per. yellow sticky trap), then declined briefly. It increased to its second peak in late September (female: 166.13 per. yellow sticky trap; male: 71.47 per. yellow sticky trap), then it declined sharply till the end of the sample date in 2018. The population of E. formosa increased steadily from the beginning of the survey, and then slowly increased to the maximum population density (36.53 per. yellow sticky trap) in mid-late October after a small decline in early September. After that, it declined till the end of the sample date in 2018. During the survey in 2019, the population density of B. tabaci began to increase gradually from early July. The density of female adults increased sharply from early August and peaked to 190.40 per. yellow sticky trap in late September. The density of male adults increased sharply from late August to early September, then increased slowly and peaked to 69.67 per. yellow sticky trap in late September. After that, the population density of B. tabaci decreased till the end of the sample date. The population of E. formosa maintained a relatively low trend till the early October, then increased and peaked to 40.67 per. yellow sticky trap in mid-October. After that, it slowly declined till the end of the sample date (Fig 1).

In the urban landscape, the population density of B. tabaci and E. formosa were relatively low in July. In early August, the population of B. tabaci began to increase, and both female and male adult densities peaked in late October (female: 150.40 per. yellow sticky trap; male: 80.87 per. yellow sticky trap). The population then declined slowly till the end of the sample date in 2018. The population of E. formosa gradually increased from mid-late August to late September, decreased to late October and then began to rise again, peaking to 40.27 per. yellow sticky trap in mid-late November. The population then declined till the end of the sample date in 2018. During the survey in 2019, the population density of B. tabaci maintain a relatively flat and low trend till to mid-August. The density of female adults began to increase sharply from mid-August to late August, then fluctuated and peaked to 152.67 per. yellow sticky trap in mid-October. The density of male adults began to increase sharply from late August and peaked to 66.47 per. yellow sticky trap in mid-September, then maintain a relatively flat trend till to mid-October. After that, the population density decreased till the end of the sample date. The population of E. formosa gradually increased to late August and maintain a relatively flat trend till to late September, then increased and peaked to 31.00 per. yellow sticky trap in mid-October. The population then declined till the end of the sample date (Fig 1).

There were variations between the population dynamics of B. tabaci and E. formosa in different types of landscapes. That of B. tabaci in flower landscapes was lower than in other landscape types, followed by the urban and the mountain landscapes in which the population dynamics changed significantly. The population dynamics in river landscape was more evident. The growth trend was faster, with a clear peak. The population dynamics trend of E. formosa was generally the same in different landscape types (Fig 1).

The highest population density of B. tabaci during the early and main activity periods, was recorded in the river landscape compared to the other landscape types. The population density in the flower landscape was significantly lower than that in the other landscapes during the main activity period (early activity period in 2018: F = 9.9230; df = 3, 56; P = 0.0001. early activity period in 2019: F = 6.3000; df = 3, 56; P = 0.0010. main activity period in 2018: F = 43.1660; df = 3, 35; P = 0.0001. main activity period in 2019: F = 42.8840; df = 3, 35; P = 0.0001). During the late activity period, the population density in the flower landscape was significantly lower than that in the urban and the river landscapes in 2018 (F = 3.5920; df = 3, 47; P = 0.0209), and was significantly lower than that in the river landscapes in 2019 (F = 2.1200; df = 3, 53; P = 0.1094). During the early and late activity periods of E. formosa, there was no significant difference in the population density among the four types of landscapes (early activity period in 2018: F = 1.2630; df = 3, 56; P = 0.2964. late activity period in 2018: F = 0.5460; df = 3, 32; P = 0.6550. early activity period in 2019: F = 0.4260; df = 3, 59; P = 0.7351. late activity period in 2019: F = 1.0900; df = 3, 41; P = 0.3649). During the main activity period, the density of E. formosa in the flower landscape was significantly higher than that in the other three types of landscapes (2018: F = 3.1700; df = 3, 50; P = 0.0328. 2019: F = 9.9140; df = 3, 44; P = 0.0001) (Table 4).

Table 4

Population density comparison of Bemisia tabaci and Encarsia formosa in different types of landscapes in Kunming, south China.

Planting years	Species	Landscape types	Population density (per. yellow sticky trap)
			Early activity period	Main activity period	Late activity period
2018	Bemisia tabaci	Flower landscape	18.84±3.43c	60.47±11.68c	39.30±10.47b
		River landscape	118.45±26.10a	415.16±39.44a	132.41±30.94a
		Mountain landscape	38.29±7.00bc	185.16±15.35b	94.43±20.92ab
		Urban landscape	54.17±11.05b	196.31±9.27b	137.40±18.42a
	Encarsia formosa	Flower landscape	6.64±1.51a	26.80±1.42a	24.73±4.48a
		River landscape	6.07±1.46a	21.60±2.46b	19.82±4.39a
		Mountain landscape	9.52±1.37a	20.63±1.08b	21.73±4.33a
		Urban landscape	6.45±1.36a	20.36±1.47b	28.17±5.69a
2019	Bemisia tabaci	Flower landscape	13.84±2.47c	65.40±7.91c	39.28±9.70b
		River landscape	90.33±22.21a	388.53±36.50a	121.40±29.92a
		Mountain landscape	41.15±9.07bc	205.89±15.13b	86.78±17.17ab
		Urban landscape	51.48±11.57b	170.96±8.04b	97.47±21.87ab
	Encarsia formosa	Flower landscape	7.77±1.74a	30.47±1.65a	19.27±3.28a
		River landscape	8.28±1.44a	24.30±2.79b	14.58±3.84a
		Mountain landscape	8.53±1.37a	19.23±0.67bc	24.24±5.22a
		Urban landscape	6.10±1.39a	16.77±1.44c	16.53±3.34a

Data in the table are mean ± SE. The different lowercases indicate significantly different at the 0.05 level in different types of landscapes with the same insect during the same tomato’s planting year.

Parasitism rates in different types of landscapes

There was no significant difference in parasitism rate of B. tabaci in different types of landscapes during the early activity period in 2018 (F = 0.3720; df = 3, 62; P = 0.7732). Parasitism rate was higher in flower landscape than that in urban landscape in 2019 (F = 2.1650; df = 3, 56; P = 0.1030). During the main activity period, the parasitism rate was highest in flower landscape and lowest in urban landscape in 2018 (F = 46.8010; df = 3, 35; P = 0.0001). The parasitism rate was highest in flower landscape and lowest in urban landscape in 2019 (F = 9.1890; df = 3, 35; P = 0.0002). During the late activity period, the parasitism rates in mountain landscape was higher than that in river and in urban landscapes in 2018 (F = 2.8360; df = 3, 44; P = 0.0498). There was no significant difference among the four landscape types in 2019 (F = 1.2200; df = 3, 53; P = 0.3123) (Table 5).

Table 5

Parasitism rates in different types of landscapes in Kunming, south China.

Planting years	Landscape types	Parasitism rate (%)
		Early activity period	Main activity period	Late activity period
2018	Flower landscape	19.78±4.96a	58.43±3.17a	21.68±6.20a
	River landscape	18.78±3.65a	21.38±1.87c	9.80±1.31b
	Mountain landscape	18.45±3.15a	37.50±4.23b	23.79±2.70a
	Urban landscape	15.18±2.51a	13.88±1.34c	14.57±4.25ab
2019	Flower landscape	20.94±2.23a	52.66±7.11a	22.60±3.73a
	River landscape	16.34±3.37ab	23.34±1.91bc	17.46±3.65a
	Mountain landscape	16.03±2.25ab	33.26±4.62b	19.94±3.25a
	Urban landscape	11.94±2.55b	18.74±4.80c	12.87±2.98a

Data in the table are mean ± SE. The different lowercases indicate significantly different at the 0.05 level in different types of landscapes during the same activity period and the same tomato’s planting year.

Discussion

Parasitoid species of Bemisia tabaci in different types of landscapes

Our results showed that the main natural enemies of Bemisia tabaci belonged to Encarsia and Eretmocerus except for those in the urban landscape. Qiu et al. and Li et al. also reported that the parasitoids of B. tabaci mainly belonged to Encarsia and Eretmocerus [31, 34]. In our study, the abundance of parasitoids was high in the flower and the mountain landscapes. Many studies have reported that flowering plants and abundant vegetation are beneficial to the diversity of natural enemies. Flowers in the flower landscape provided enough additional food sources for parasitoids, such as nectar. The rich vegetation in the mountain landscape also likely provided good shelter from agricultural disturbance and overwintering sites [35-37].

In the flower landscape, the population of B. tabaci was always significantly lower than that in the other landscape types. The presence of non-host plant plants species of B. tabaci such as Rosa chinensis, Dianthus caryophyllus, Myosotis sylvatica and Eustoma grandiflorum in the landscape could be the influencing factor [34]. On the other hand, the presence of abundant flowering plants in the landscape, may have improved the control ability of the parasitic natural enemies [35, 38, 39]. The combined action of the lots of non-host plant species and parasitic natural enemies kept the population of B. tabaci at a low density. Xiao et al. planted papaya (Carica papaya L.) and other banker plants beside a field to enhance the efficacy of biocontrol, and this had an enhancing effect on the population of Encarsia sophia [11]. Therefore, the use of flowering plants as banker plants beside tomato fields, can provide energy substrates for parasitoids to increase their parasitism rate and population densities, for effective biological control.

Our results showed that the growth trend of B. tabaci in tomato plantings in the river landscape was faster than that in other landscape types, and the population density was higher. The existence of the river created a higher humidity over the landscape. The flight activities of some pests under high humidity conditions are significantly higher than that under low humidity conditions, which may be related to reduced water evaporation under low humidity conditions [40-42]. Similarly, humidity also significantly affects the growth, development, survival, and longevity of pests [43, 44]. Therefore, high humidity may be one of the main reasons for the outbreak of B. tabaci in the landscape.

Parasitism rate in different types of landscapes

During the main activities period, the parasitism rate in tomato plantings in the flower landscape was higher, while that in the urban and the river landscapes were lower. There were abundant flowering plants in the flower landscape, which may have provided abundant nectar. Most parasitic natural enemies need to supplement their nutrition by feeding on pollen and nectar to promote the development of their reproductive system, especially the ovary [38, 45, 46], thereby increasing the number of eggs laid [35, 47] and parasitism rate [39, 48]. Similarly, carbohydrates in pollen and nectar can also provide energy for the survival and activity of natural enemies and prolong their life span [49]. Flower landscapes provide these energy substances. Therefore, it is necessary to set up flowering plants in farmland landscape. However, different flowering plants have different structures and signals towards natural enemies, which results in different levels of attraction. Similarly, different natural enemies show selectivity to different flowering plants due to their different nutritional needs [50-52]. For instance, the presence of different plants around a field had different effects on the parasitism rate of whitefly parasitoids [53]. In our study, we recorded many kinds of flowering plants in the flower landscape. It is therefore necessary to conduct a further study, to identify which of these plants is most beneficial to the fitness of the parasitoids.

In summary, this study aimed to study the impact of farmland landscape on parasitic natural enemies and pests. Together, these can be considered as an agricultural ecosystem. In addition to parasitic natural enemies, predatory natural enemies and entomopathogenic fungi also control B. tabaci [34]. Therefore, this requires that consideration is given to them in the analysis of the natural enemy complex. Furthermore, the role of neutral insects in farmland ecosystems should also be considered, as they serve as alternative food for natural enemies and play important roles in maintaining ecosystem stability [54]. The need to establish a sustainable green eco-agricultural landscape requires that studies are continually carried out to assess the feasibility of green development of different types of landscape.

Previous study, parasitoid species of Bemisia tabaci and its population densities.

Data were initially subjected to a one-way ANOVA with agriculture landscape variables as the main effect. Difference in population densities was compared among agriculture landscape types by Least Significant Difference. The significance threshold was P = 0.05 in all tests. Data analyses were performed using SPSS 20.0.

(XLSX)

Click here for additional data file.

Metadata, covariance matrix of PCA.

The 12 agriculture landscapes located in the surroundings of Kunming, south China (24°42’45’’N-25°22’43’’N, 102°22’18’’E-103°10’90’’E). it was selected by use of Google Earth Profession and field inspections (ground-truthing) once a month during the tomato growing seasons in 2018 and 2019. The cover types in each landscape were divided into 10 types according to vegetation type, human factor interference and land type characteristics. A Principal Components Analysis (PCA) was performed to reduce the dimensions of the landscape data. These ten land cover types were divided for the PCA analysis, the land cover type with the largest area in one landscape and the absolute value of first principal component greater than 0.9 was selected as the landscape type. Principal component axes were extracted using correlations among variables, and the resulting factors were not rotated.

(XLSX)

Click here for additional data file.

Parasitoid species of Bemisia tabaci and its relative abundance in different types of landscapes in Kunming, south China.

Census data were initially subjected to a one-way ANOVA with agriculture landscape types as the main effect. Differences in relative abundance were compared among different parasitoid species of B. tabaci in the same agriculture landscape types by Least Significant Difference. The significance threshold was P = 0.05 in all tests. Data analyses were performed using SPSS 20.0.

(XLSX)

Click here for additional data file.

Population densities of Encarsia formosa and Bemisia tabaci.

Census data were initially subjected to a one-way ANOVA with agriculture landscape types as the main effect. To reduce the impact of occurrence time on the population densities of E. formosa and B. tabaci, the activity period of E. formosa and B. tabaci densities were divided into early, main and late activity period by quartile method. Differences in E. formosa and B. tabaci densities and parasitism rate were compared among agriculture landscape types in the same activity period by Least Significant Difference. The significance threshold was P = 0.05 in all tests. Data analyses were performed using SPSS 20.0.

(XLSX)

Click here for additional data file.

19 Nov 2021

Submitted filename: Rebuttal letter.docx

Click here for additional data file.

28 Mar 2022

10 May 2022

Journal Requirements:

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming.

Response: We have revised our manuscript carefully and we ensure that our manuscript meets PLOS ONE's style requirements, including file naming.

2. Thank you for stating the following financial disclosure: Please state what role the funders took in the study. If the funders had no role, please state: "The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript." If this statement is not correct you must amend it as needed. Please include this amended Role of Funder statement in your cover letter; we will change the online submission form on your behalf.

Response: This work was supported by Yunnan Fundamental Research Projects [grant no. 202201AT070269]; the National Natural Science Foundation of China [grant no. 31760541]; the Reserve Talent Project of Yunnan’s Young and Middle-aged Academic and Technical Leaders [grant no. 202105AC160071]; the Young Top Talents of “High-level Talents Training Support Program in Yunnan Province” [grant no. YNWRQNBJ2020291]; and the Reserve Talents Project for the 17th Batch of Kunming’s Young and Middle-aged Academic and Technical Leaders [grant no. KMRCH2019023]. One of the corresponding authors, Dr. Xiaoming Zhang, is the funder. And we have stated what role the funder took in the study. Please see lines 476-477 in revised manuscript with track changes.

3. Please amend your list of authors on the manuscript to ensure that each author is linked to an affiliation. Authors’ affiliations should reflect the institution where the work was done (if authors moved subsequently, you can also list the new affiliation stating “current affiliation:….” as necessary).

Response: We have amended our list of authors on the manuscript to ensure that each author is linked to an affiliation. Please see lines 5-9 in revised manuscript with track changes.

4. We note that you have referenced (ie. Bewick et al. [5]) which has currently not yet been accepted for publication. Please remove this from your References and amend this to state in the body of your manuscript: (ie “Bewick et al. [Unpublished]”) as detailed online in our guide for authors

http://journals.plos.org/plosone/s/submission-guidelines#loc-reference-style

Response: We have reviewed our reference list to ensure that references we have referenced has been accepted for publication.

5. We note that Figure 1 in your submission contain [map/satellite] images which may be copyrighted. All PLOS content is published under the Creative Commons Attribution License (CC BY 4.0), which means that the manuscript, images, and Supporting Information files will be freely available online, and any third party is permitted to access, download, copy, distribute, and use these materials in any way, even commercially, with proper attribution. For these reasons, we cannot publish previously copyrighted maps or satellite images created using proprietary data, such as Google software (Google Maps, Street View, and Earth). We require you to either (1) present written permission from the copyright holder to publish these figures specifically under the CC BY 4.0 license, or (2) remove the figures from your submission:

Response: We agree with the comment. We have removed the Fig. 1. Please see lines 96 and 138 in revised manuscript with track changes.

6. Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

Response: We have reviewed our reference list to ensure that it is complete and correct.

Reviewers' comments:

Reviewer #1:

Point 1. The word abondance should be abundance

Response: We agree with the comment. We have revised “abondance” to “abundance”. Please see lines 23, 155, 157, 190, 229 and Table 2 in revised manuscript with track changes.

Point 2. L92: add “each of” before 12 tomato field plots

Response: We agree with the comment. We have added “each of” before “12 tomato field plots”, Please see line 94 in revised manuscript with track changes.

Point 3. L320-321: This sentence is not clear to me

Response: We agree with the comment. We have revised this sentence to make it clearer. Please see lines 424-425 in revised manuscript with track changes.

Point 4. The biodiversity is one key factor influencing the landscape. The authors may have a brief description of the diversity on each of the landscape type. For example, what is the main flower in the flower field landscape? How about other landscape?

Response: We agree with the comment. We have written small sub-paragraph for each of the landscapes describing their diversities. Please see lines 108-129 in revised manuscript with track changes.

Point 5. The predatory natural enemy is always co-occurrence with parasitic natural enemy. How about the role of predatory natural enemies in the field trial sites?

Response: In our study, it was found that the dominant parasitic natural enemy of Bemisia tabaci in different agricultural landscapes was Encarsia formosa, which was the specific natural enemy of whitefly. Similarly, we also investigated predatory natural enemies and found that the main predatory natural enemies were. Nesidiocoris tenuis Reuter (Hemiptera: Miridae), Chrysoperla sinica Tjeder (Neuroptera: Chrysopidae), Menochilus sexmaculata Fabricius (Coleoptera: Coccinellidae) and Harmonia axyridis Pallas (Coleoptera: Coccinellidae). However, most predatory natural enemies are omnivorous, and not only control Bemisia tabaci, due to the abundant data, we are still conducting correlation analysis on the relevant data of predatory natural enemy insects, which will be published later.

Reviewer #2:

Point 1. The hypothesis is based on finding that the parasitoids of B. tabaci could be observed in the tomato planting fields, but the number of species and black pupae of the wasps were significantly difference under different agricultural landscapes around Kunming, Yunnan Province. However, this is not published data, and it is better to put the content in the manuscript as supplementary material.

Response: We agree with the comment. We have uploaded the data as supplementary data 1 (Data S1). Please see lines 69 in revised manuscript with track changes.

Point 2. Some of the descriptions in the data analysis are confusing. Are the data normally distributed? The data should be transformed if it did not follow a normal distribution. Why use the LSD method instead of Tukey's HSD?

Response: ① We agree with the comment. We have revised the descriptions in the data analysis to make it clearer. Please see lines 193-194 in revised manuscript with track changes.

② LSD method has wider application scope than Tukey's HSD and it has high inspection efficiency. Tukey's HSD is applicable to the same number of samples in each treatment. Because the sampling number in different landscapes in the same activity period was not always the same in our study, we chose the LSD method.

Point 3. Some parts of the results lack the specific description of the statistical analysis results. For example, the comparison of relative abundance and population dynamics, the description of the results of specific statistical analysis is not found. Please supplement the results of statistical analysis.

Response: ① We have supplemented specific description of the statistical analysis results about the comparison of relative abundance. Please see lines 230-236 in revised manuscript with track changes.

② For population dynamics, we want to show the population growth and decline process of Bemisia tabaci and Encarsia formosa in different agricultural landscapes in this part, so we did not make statistical analysis. Please see lines 281-348 in revised manuscript with track changes.

Point 4. The description of the results is incomplete, please revise it.

Response: We agree with the comment. We have revised the description of the results to make it clearer and more complete. Please see lines 230-236 and 269-348 in revised manuscript with track changes.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

18 Jul 2022

Flowering agricultural landscapes enhance parasitoid biological control to Bemisia tabaci on tomato in south China

PONE-D-21-35756R1

Dear Dr. Yang,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Yonggen Lou

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: (No Response)

Reviewer #2: Accepted. The data have been uploaded as supplementary data, and the manuscrip has been well revised.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

**********

22 Jul 2022

PONE-D-21-35756R1

Flowering agricultural landscapes enhance parasitoid biological control to Bemisia tabaci on tomato in south China

Dear Dr. Yang:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Yonggen Lou

Academic Editor

PLOS ONE