Landscape context does not constrain biological control of Phenacoccus manihoti in intensified cassava systems of southern Vietnam
Introduction
The cassava mealybug, Phenacoccus manihoti Matile-Ferrero (Homoptera: Pseudococcidae) is a prominent pest of cassava (Manihot esculenta Crantz; Euphorbiaceae) and one of the world’s most notorious invasive species. Endemic to the Paraguay River basin, P. manihoti was inadvertently introduced into Africa during the early 1970s and subsequently spread through the continent’s extensive cassava belt (Herren and Neuenschwander, 1991, Bellotti et al., 2012). Capable of inflicting yield losses up to 58–84% (Nwanze, 1982, Schulthess et al., 1991), P. manihoti devastated local cassava production and impacted food security for underprivileged rural populations across sub-Saharan Africa. In late 2008, this same pest was detected in Thailand, where it was reportedly causing root yield reductions up to 50% and economic losses over US $30 million nationally (Muniappan et al., 2009, TTTA, 2011). By 2014, P. manihoti had spread extensively into neighboring countries and several Indonesian islands (Sartiami et al., 2015, Graziosi et al., 2016). Climate-based niche modeling further revealed that other key production areas in eastern Indonesia and the Philippines are also climatically suitable for P. manihoti (Yonow et al., 2017). As Southeast Asia houses a multi-billion dollar cassava industry and accounts for nearly 95% of the world’s cassava exports (Cramb et al., 2017), the (socio-)economic impacts of this pest were projected to be exceptionally large.
With the 1981 introduction of Anagyrus lopezi (De Santis) (Hymenoptera: Encyrtidae) in West Africa, a globally-acclaimed biological control program against P. manihoti was started (Neuenschwander, 2001). This solitary, host-specific parasitoid had earlier been collected in Paraguay and southern Brazil from small P. manihoti colonies on cassava (Lohr et al., 1990). Following its release in Nigeria, A. lopezi promptly established and suppressed P. manihoti population levels from more than 100 to fewer than 10–20 individuals per tip (Hammond et al., 1987). In less than three years following its release, A. lopezi had effectively dispersed over 200,000 km2 and colonized the vast majority of cassava fields within this range (Herren et al., 1987). Though multiple endemic primary parasitoids and hyper-parasitoids were recorded in mealybug-invaded areas in Africa (Neuenschwander et al., 1987, Neuenschwander and Hammond, 1988), these largely did not impede the success of A. lopezi as biological control agent (Neuenschwander, 2001). Overall, the parasitic wasp successfully established in 26 different African countries, prevented wide-spread famine and generated long-term economic benefits of US$ 9.4-20.2 billion (Zeddies et al., 2001).
In late 2009, A. lopezi was introduced from West Africa into Thailand and subsequently into Indonesia, through a joint endeavor between the Food and Agriculture Organization (FAO), CGIAR centers and Thai governmental institutions (Winotai et al., 2010, Wyckhuys et al., 2015). Other methods promoted for P. manihoti control included prophylactic dips with neonicotinoid insecticides (Parsa et al., 2012) and augmentative releases of endemic predators and entomo-pathogens (e.g., Saengyot and Burikam, 2012, Sattayawong et al., 2016). However, it has largely been deemed that A. lopezi effectively suppressed local mealybug populations. In a first regional assessment during 2014–2015, A. lopezi was routinely found in P. manihoti-affected fields at parasitism levels of 10–57% (Wyckhuys et al., 2017a). Yet, from smallholder fields in Cambodia, a diverse and speciose complex of hyperparasitoids or mummy parasitoids was equally recorded (Wyckhuys et al., 2017b).
Natural enemy abundance and performance, as much as pest pressure, are shaped by a wide range of variables at a field, farm, and agro-landscape level. In Asian cassava fields, patch-level characteristics, such as soil parameters and a plant’s phytopathogen infection status, readily modulate A. lopezi × P. manihoti interactions (Wyckhuys et al., 2017a, Wyckhuys et al., 2017b). Landscape-dependent impacts on cassava mealybug biological control have rarely been inferred in past studies, and so far have not been studied in-depth. Though the effects of landscape structure on natural enemy abundance, diversity, and activity have been relatively well investigated (e.g., Bianchi et al., 2006, Chaplin-Kramer et al., 2011, Veres et al., 2013, Schellhorn et al., 2015b), much less is known about its impacts on pest pressure or (natural) biological control. Yet, for specialist parasitoids such as A. lopezi, habitat loss and landscape simplification could be particularly disruptive (Cagnolo et al., 2009). Also, while landscape-level interactions have been assessed to a fair extent for annual cropping systems under temperate conditions, they have only received scant attention in (semi-)perennial cropping systems in the tropics (but see Tylianakis et al., 2007, Pak et al., 2015). Lastly, fourth-trophic level organisms, such as hyperparasitoids, have only received peripheral attention in landscape ecology studies (Rand et al., 2012, Plećaš et al., 2014), even though they are highly responsive to landscape complexity, are connected to arthropod population dynamics across habitats, and may release herbivores, such as P. manihoti, from biological control (Sullivan and Volkl, 1999).
In this study, we examine landscape-level effects on P. manihoti biological control in intensified cassava cropping systems of southern Vietnam (i.e., Tay Ninh province). We characterize overall P. manihoti population levels across two successive growing seasons, assess A. lopezi establishment and impact, and describe the resident hyperparasitoid community. Furthermore, we contrast mealybug-parasitoid-hyperparasitoid dynamics in fields embedded within simplified, large-scale landscapes vs. complex, small-scale settings. This study adopts a landscape ecology approach, yet is no landscape analysis sensu stricto as it is not guided by structural indicators of landscape composition or spatial configuration (Mühlner et al., 2010, Birkhofer et al., 2018). Our work constitutes the first, comprehensive assessment of A. lopezi parasitism rates and parasitoid × host dynamics from a key cassava-growing area in SE Asia, examines P. manihoti biological control through a (novel) landscape ecology lens, and provides valuable insights to guide further (invasive) pest mitigation programs.
Section snippets
Study sites
Our study was conducted in several rural communes of Tay Ninh province, southern Vietnam (Fig. 1); an area characterized by highly-intensified cassava production, with staggered (overlapping) planting and near-continuous, year-long cultivation. Local cassava fields are routinely established with locally-sourced stem cuttings (i.e., stakes), receive ample fertilizer and herbicide inputs during the first 3–4 months, and are manually harvested at 9–12 months after planting. Overall, two different
Mealybug & parasitoid community composition
In field surveys from 2014 to 2015, a total of four different mealybug species were recorded from cassava fields: P. manihoti, Pseudococcus jackbeardsleyi Gimpel and Miller, Paracoccus marginatus Williams and Granara de Willink, and Ferrisia virgata (Cockerell). Across sites and years, P. manihoti constituted 91.4% of the mealybug complex, with other species representing 5.7%, 2.8% and 0.2%, respectively. Field-level incidence of P. manihoti ranged from 0% to 82% across both years, with an
Discussion
As the stage for one of the world’s greatest biological control successes (Neuenschwander, 2001), cassava cropping systems possess unique and noteworthy features. Being long-season crops exposed to comparatively few disturbances, cassava systems provide habitats of prolonged durational stability and vegetational complexity for myriad natural enemies. The secretion of extra-floral nectar by the cassava plant also constitutes a favorable trait for multiple beneficial organisms, including
Acknowledgments
This manuscript presents original datasets that result from collaborative research, with trials jointly conceptualized, defined and executed by Vietnamese counterparts, CIAT personnel and international cooperators. We would like to thank Drs. Nguyen Van Liem, Trinh Xuan Hoat and Le Xuan Vi at Vietnam’s Plant Protection Research Institute (PPRI-VAAS), for their support to field staff and for facilitating this 2-year insect survey. We are grateful to collaborators from Hung Loc Research station
References (69)
- et al.
Relationships between multiple biodiversity components and ecosystem services along a landscape complexity gradient
Biol. Conserv.
(2018) - et al.
Tools and techniques for investigating impacts of habitat complexity on biological control
Biol. Control
(2014) - et al.
Relationships between biodiversity and biological control in agroecosystems: current status and future challenges
Biol. Control
(2014) Interactions between parasitoids and higher order natural enemies: intraguild predation and hyperparasitoids
Curr. Opin. Insect Sci.
(2016)- et al.
Introduction and dispersal of Epidinocarsis lopezi (Hym., Encyrtidae), an exotic parasitoid of the cassava mealybug, Phenacoccus manihoti (Hom., Pseudococcidae), in Africa
Agric. Ecosyst. Environ.
(1987) Biological control of the cassava mealybug in Africa: a review
Biol. Control
(2001)- et al.
Parasitoid wasps benefit from shade tree size and landscape complexity in Mexican coffee agro-ecosystems
Agric. Ecosyst. Environ.
(2015) - et al.
Current knowledge and future research perspectives on cassava (Manihot esculenta Crantz) chemical defenses: an agro-ecological view
Phytochemistry
(2016) - et al.
Landscape composition and configuration influence cereal aphid-parasitoid-hyperparasitoid interactions and biological control differentially across years
Agric. Ecosyst. Environ.
(2014) - et al.
Natural enemies on the landscape – integrating life-history theory and landscapes
Biol. Control
(2014)