Effects of productive management of an agroforestry system on the associated diversity of parasitoids (Hymenoptera: Braconidae)

Tropical agroecosystems have emerged from the continuous modification of natural environments, as a sustainability alternative for food production and biodiversity conservation. This work explores how the diversity of parasitoids is modified in environments where plant diversity is limited e.g. crops and when these are adjacent to secondary vegetation, i.e. a scenario fragmented continuously in a limited space. It was found that there is no direct effect of plant diversity in the group of parasitoids studied; but the number of specialist species is high, which indicates that in diversified agroecosystems these probably function as remnants of natural habitat or as a refuge for parasitoids that disperse to different types of management within the agroecosystem. Therefore, it is necessary to consider in future studies the controls exerted by the plant diversity effects bottom-up and consumer top-down. Adding to this the context of the interactions that occur in agroecosystems.


24
Agroecosystems arise in the tropics derived from the loss of vegetation cover, the conversion 25 and fragmentation of natural areas in ecosystems. These activities modify the species that 26 inhabit and depend on vegetation,with effects that depend on the range of the species and its 27 habitat requirements (Scolozzi & Geneletti, 2012;Campos-Navarrete et al 2015a). On the 28 other hand, the speed in the loss of species has increased the interest in the study of 29 biological diversity in agroecosystems, mainly insects, because they constitute the most 30 important fraction of the diversity of a territory since they provide multiple ecological  The effect of land use, in conjunction with the associated plant diversity at higher trophic 43 levels, has been explained primarily because greater diversity, by generating a more complex 44 environment, consequently offers a greater number of shelters and prey (Russell, 1989;45 Obermaier, et al 2008; Moreira et al 2016),which in turn generates increases in predation 46 rates, causing a reduction in the abundance of parasitoid prey ("up-down" effects). Evidently, 47 this type of effects are highly relevant to be considered in the design of productive systems 48 such as forestcrops, due to their potential in pest control (Russell, 1989;Abdala-Roberts et 49 al 2015;2016a). This last factor can reduce or increase the abundance of prey (herbivores) 50 depending on characteristics such as the specialization of their diet (generalists vs.

59
The present work explores the effect of four productive areas in an agroecosystem of multiple 60 production in relation to the diversity of parasitoids (Braconidae) associated. Particularly 61 changes in richness, abundance, similarity in areas and in parasitism strategies represented 62 by the proportion of koinobiont (specialist) and idiobiont (generalist) species. Trying to infer 63 how the diversity of parasitoids is modified in environments where plant diversity is limited 64 e.g. crops and when these are adjacent to secondary vegetation, i.e. a scenario with 65 continuously fragmented in a limited space.

67
Study area. The present work was carried out in the east of Yucatán in the municipality of 68 Tizimin, Yucatán, Mexico. This area is characterized by the conversion of land use from 69 native jungle to grasslands for the production of pasture for cattle feed. Livestock represents 70 30% of the economic activity of this area (INEGI 2015). The agroecosystem of study is 71 located in the area of agricultural and livestock production of the TecNM Campus Tizimin, 72 located at the end of the airport Cupul s/n C.P. 97700 with the coordinates 21°09′29" N 73 88°10′21"W.

74
The agroecosystem has three areas of crops: a plantation of Cocos nucifera "PC" 75 (monoculture),a plantation of Citrus lemon "PL" (monoculture), in the livestock production 76 area is located the grassland area with star grass Cynodon sp. named grassland "PT" and the 77 fourth type is the matrix of secondary vegetation with more than 30 years of recovery "VS".

78
This contrast in a limited space provides a frequent scenario today, originated by the 79 processes of fragmentation, originating contrasting sites.  Parasitology of the TecNM Campus Tizimín, before the identification the braconids were 91 separated from the rest of the insects, later they were counted and labeled with the 92 corresponding data, for their identification the specimens were assembled according to the 93 technique traditionally used for these organisms. Identification was carried out using the 94 taxonomic key for Braconidae (Wharton et al 1998).

95
The assembly was carried out using entomological pins correctly placing each insect, which 96 were also labeled for a systematic control of these. The reason for mounting is that in this 97 way it is easier to observe insects when they are dry, since when they are wet it is difficult to 98 observe certain characteristics.

99
The material was determined up to the taxonomic level of genus subsequently the concept of

112
The similarity between the braconid communities was estimated using the Sorensen index 113 (Krebs, 1989) which is based on the presence/absence of species and the number of species 114 whether common or rare (Spellerberg, 1991). It relates the number of species in common to 115 the arithmetic mean of the species at the sites (Magurran, 1988).

116
To estimate the representativeness of the species richness of the samples in each type of 117 vegetation, the EstimateS version 9 program was used, which presents specialized estimators 118 for different types and sizes of data. We used ice wealth estimators based on the number of 119 rare species (those observed in less than 10 sampling units) and jacknife first order which is 120 based on the number of unique species (Colwell and Coddington, 1994).

121
With the data of the morphospecies recorded by area, using relative abundance, as the 122 response variable, the change between the four areas was evaluated. This is through a model   (Table 1).

139
Richness and abundance of species by typearea. Wealth in each area of the agroecosystem 140 was distributed as follows from highest to lowest. PL with the largest number of subfamilies, 141 followed by PC, VS and PT (Table 1). As for gender from highest to lowest PL, PC, VS 142 and PT (Table 1). The morphospecies were distributed from highest to lowest in the areass 143 in the following order first PL, second PC and with the same number PT and VS (Table 1).

144
Finally, for the number of individuals the largest was concentrated in PL, followed by PC,

145
PT and finally VS (Table 1). For wealth in general it was observed that the PL and 146 PC,considereds monocultures concentrated the largest number of individuals and 147 taxonomic wealth, while VS occupied the third site and with the least wealth and abundance 148 the PT with the simplest vegetation structure composed solely of grass.

149
The general pattern for the number of rare and unique morphospecies, as well as species 150 diversity according to the Shannon Index by area is described below. For rare species the 151 order of greatest number was in PT, PL, PC and VS ( Table 2). As for unique species from 152 highest to lowest these were found in PC, PT, PL and finally VS (Table 2). In general, it is 153 observed that the VS considered a polyculture and with greater complexity was the one with 154 the lowest number of rare and unique species present. This contrasts with what was found 155 for the Shannon-Wienner diversity index, where the highest diversity was observed in PL, 156 followed by VS, finally PC and PT (Table 2).

157
Specific wealth estimation: The wealth estimators used were ICE and Jacknife of the first 158 order. According to the ICE, 60% of the species were captured on average for the 159 agroecosystem, while for the jacknife of the first order 70% were captured according to the 160 estimate for the agroecosystem. Table 3 shows in particular the percentage of wealth 161 estimators by area.

162
Similarity ofcommunities: The Sorensen index indicated that the sites that presented the 163 greatest similarity were PC with VS, followed by PT with VS, in third place, PL with VS 164 and those that presented the least similarity were PC with PT (Table 4). The above highlights 165 the importance of conserving remnants of secondary vegetation in agroecosystems, since it 166 was observed that VS is similar to crop areas.

167
Effect of the areas on the abundance. As for the quantitative response in the abundance 168 relative to the area of cultivation and the functional role of the braconist in this system, it was 169 observed that the Area has a statistically significant effect on abundance (F 1, 3 = 2.8125 p = 170 0.000). Higher values of relative abundance were observed in PC, PT and VS with respect 171 to PL (Figure 1). As for the parasitism strategy, a significant effect was found on abundance 172 (F 1.2 =8.5947 p =0.000), being greater in koinobionts (specialists) compared to idiobionts 173 (generalists) (Figure 2). However, no interaction effect was found between the Area and 174 parasitism strategies (F 1.1 = 0.6097 p = 0.65)inthis agroecosystem.
Proportion of the estrategia of parasitism. The distribution was as follows, in general 176 the Koinobionte (specialist) represented 63% on average, while 36% on average is classified 177 as Idiobiont (generalists) and only 1% that corresponds to the registror a species of the genus 178 Epsilogaster (Mendesellinae) of which its biology is unknown (Table 5). The percentage 179 from highest to lowest in the areas of koinobiont species was higher percentage was 180 presentedor in PT, followed by PC, VS and finally PL ( Table 5). As for Idiobionts with 181 the highest and equal percentages PL and VS, followed by PC and finally PT (Table 5).

193
In the study of this diversified agroecosystem, it was found that the areas contain high levels  196 This supports the benefit of agroecosystems as conservation strategies, as some insects (e.g. there were no significant differences in the abundance of bracor nests.

227
In this agroecosystem it was found that the largest number of individuals were classified as 228 specialists (koinobionts) and that the presence of these is independent of the area of 229 cultivation. One mechanism that could explain the patterns found in this agroecosystem may