Predicting the spread of an invasive tree pest: The pine wood nematode in Southern Europe

Abstract

Invasive tree pests are increasingly threatening forest ecosystems. Understanding and controlling their spread presents a major challenge for managers, researchers, and policy makers. In 1999, the pine wood nematode (PWN) Bursaphelenchus xylophilus, causal agent of pine wilt disease and native to North America, was first detected in Europe, more precisely in south‐western Portugal. Since then, it has spread to more than 30% of the country, producing large‐scale damages in Portuguese forests. We applied a new spatio‐temporal network‐based model to predict, at yearly intervals, the natural spread of the PWN on the Iberian Peninsula, as dispersed by the longhorn beetle Monochamus galloprovincialis, its only known vector in Europe. We compared the model predictions with independent observations of the natural spread of pine wilt disease in Portugal between 2005 and 2015. We simulated the effect of potential containment measures on the disease spread. The model predicted the status (infected or non‐infected) of forest areas with 93% accuracy, had 93% sensitivity and specificity, and AUC = 0.96 (averages for 8 years with validation data). Simulated clear‐cut belts stopped the spread only if they were wider than 30 km, although thinner belts delayed invasion. Furthermore, clear‐cuts were much more effective in slowing down the invasion when reinforced, in the adjacent areas, by the reduction in the vector beetle population through mass trapping and by the early detection and removal of infected trees. Model projections identified areas through which the nematode is most likely to invade, by natural means, the phytosanitary buffer zone established along the Portuguese border, and Spain. Synthesis and applications. In the absence of effective containment measures, the pine wood nematode may naturally spread into Spain in about 5 years. In less than 10 years, it may reach the major forest and climatic corridors that provide a gateway for subsequent expansion towards the rest of the Iberian Peninsula and, in the longer term, towards other European countries. Our model can considerably support management efforts by forecasting when areas are at highest risk of vector‐mediated invasion as well as the effects of specific control measures on the disease spread.

Importance of spatio–temporal connectivity to maintain species experiencing range shifts

“…Although some recent studies highlighted the importance of accounting for temporal aspects in connectivity modeling (Alagador et al 2014, Mui et al 2017, only a few of them accounted for dispersal processes over generations (Saura et al 2014, de la Fuente et al 2018, the variation in importance of habitat patches in maintaining connectivity due to seasonal habitat change (Mui et al 2017, Bishop-Taylor et al 2018, or the implications of ephemeral patches and transient connectivity change in metapopulation dynamics (Reigada et al 2015, Perry andLee 2019). Yet, the importance of temporal dynamics of habitat patches has largely been ignored (Blonder et al 2012).…”

Effectiveness of clear‐cuttings in non‐fragmented pine forests in relation to EU regulations for the eradication of the pine wood nematode

“…Besides, the PWN can spread at long distance with the transportation of infected wood products. In Portugal, the PWN has locally spread at 5.3 km/year on average (De la Fuente, Saura, & Beck, 2018) but it can propagate further through human-assisted dispersal (on average at 150 km; Robinet, Opstal, Baker, & Roques, 2011).…”

Pest categorisation of non‐EU Monochamus spp.

“…In Japan, Osada et al (2018) analysed historical records of PWD infection and vector abundance across 403 municipalities in northern Honshu during the period 1980-2011, and found that the yearly local spread of PWD (due to vector movements) was 2.2-3.6 km. Analysing the spread of PWD in Portugal between 2005, de la Fuente et al (2018 found a yearly spread into adjacent areas (thus due to vector movements) of 5.3 km in the average, with a maximum of 8.3 km.…”