In the present study, the physiological data and some of the presence points of ACT were sourced from the scientific literature. This encourages data reuse for future research, allowing future research work to progress more efficiently and effectively 68–69. Unlike the mechanistic models, the CLIMEX model uses the occurrence records of a species and its physiological stress factors to establish the potential distribution of species. The predictions are consistent with the known global T. erytreae distribution. The model predicted suitable areas in regions that were earlier predicted using different machine learning algorithms, such as MaxEnt 50. Again, the prediction covered suitable areas, including the validation area, in which the data were not used to estimate the geographical distribution of the pest. Overall, of the 301 occurrence points, only one occurrence record was found in areas deemed unsuitable by the model fit, thus most of the occurrence records inside the validation area matched the CLIMEX-estimated suitable areas for T. erytreae invasion and spread.
The geographical distribution of invasive species like T. erytreae has been evaluated through SDMs, but most studies were conducted locally. For example, Kyalo et al. 50, combined MaxEnt with remotely-sensed vegetation variables to predict the spatial distribution of T. erytreae in Kenya. A study by Aidoo et al. 70, indicated the potential distribution of T. erytreae using bioclimatic variables and elevation datasets in Kenya using MaxEnt. A recent study predicted the expansion of T. erytreae in the Iberian Peninsula using a pest risk analysis approach 51. A more recent study modeled the potentially suitable areas of T. erytreae in two European countries, Portugal and Spain using MaxEnt 15. Finally, Espinosa-Zaragoza et al. 44, predicted the suitable areas of T. erytreae in Mexico based on MaxEnt and only bioclimatic variables. However, none of these studies evaluated the impact of climate change on the global geographical distribution of T. erytreae, and under current and future scenarios.
Prevention of invasive alien pests is the surest way of reducing or avoiding their impact on sustainable agriculture production and food security thus prevention of biological invasion is less expensive and more manageable than post-entry management 71. Therefore, it is essential to understand and assess the impact of climate change on the potential distribution and range shifts of invasive species like T. erytreae for a better biosecurity plan. The study's information may be used to develop a proactive, adaptive, and integrated invasive species management approach to curtail and prevent further spread.
In the present study, we used the CLIMEX model to predict habitat suitability for T. erytreae, and the suitable areas ranged from unsuitable (EI < 0) to high suitability (EI < 0 to EI > 30) is concentrated in the north-south temperate zone around the world. The model predicts an expansion of suitable areas outside reported countries in Europe (i.e., Spain and Portugal). These areas were predicted to have moderate (0 < EI < 30) to high suitability (EI > 30) for T. erytreae. The new moderately suitable regions identified by the model include Cyprus, Croatia, Greece, France, Albania, and these countries produce citrus in Europe; thus the introduction of T. erytreae into these zones could threaten their citrus industry. To date, T. erytreae is still restricted to Span and Portugal 72. However, the pest can invade and establish in other European countries where citrus is cultivated 73. However, suggests that there is a need for early detection and eradication measures to avoid the spread of this invasive alien pest across Europe. In Oceania, T. erytreae has not been reported, but our study predicts highly suitable areas, particularly in the east coast of Australia. If T. erytreae invade Australia, its establishment will be facilitated by the suitable temperate climate, especially its mild winter temperatures 74, 75. Therefore, continuous strict compliance of the continent biosecurity measures is required at the entry points (i.e., railway stations, harbors, airports, and lorry parks) to intercept host materials. Apart from citrus, alternative host plants belonging to the family Rutaceae, such as the curry tree, Murraya koenigii (Bergera koenigii), and Clausena anisata (Willd.) Hook. f. ex Benth., 70 should be regularly checked for all stages of T. erytreae at the Australian entry points. According to Australian Plant Biosecurity report 76, there is a need to pay attention to additional types of invasion pathways, including natural spread, human travelers and their luggage, and infested machinery.
Detection and monitoring of T. erytreae using sticky traps can be of great importance to where the pest exists or is at risk of invasion. In Kenya, Aidoo et al. 45, demonstrated that sticky card traps, especially the yellow ones, effectively detected field populations of T. erytreae even at low densities. Other sampling methods, such as visual observation of adults immature and collection of symptomatic leaves, could facilitate early detection of the pest, particularly after it has invaded new locations. In most areas, anthropogenic activities have been strongly associated with the spread of T. erytreae through the alternate hosts' movement. Therefore, regular visits to areas where these alternate hosts are present may help early detection and monitoring.
We demonstrated that the overall potential habitat suitability for T. erytreae will decline in the future under SRES A1B and A2 scenarios compared to the current climate. The A1B and A2 SRES scenarios and the global climate model (GCM) CSIRO-Mk3.0 datasets predict a temperature increase of 2.11°C and a precipitation loss of 14% by 2100 77. Previous studies have shown that temperatures above 32°C coupled with 30% relative humidity degrees are detrimental to T. erytreae development64–65. As a result, in regions near the equator where the annual mean temperature is around 31°C, a rise in 2.11°C may increase T. erytreae mortality due to its climatic requirements77–78. In contrast, areas with high elevations, such as Kenya, Tanzania in East Africa, will remain suitable because the changing climate may favor the survival and distribution of the pest in those areas. The areas that will reduce suitable areas are primarily concentrated in countries, such as India and Mexico. Because it has been demonstrated that T. erytreae prefers cool and moist climatic conditions79, the extreme temperature may cause high mortalities14. On the other hand, the pest has adapted to and settled in a wide range of ecological conditions, including equatorial, arid, and warm temperate climates with varying temperatures and rainfall13. In the future, our model predicts more areas of moderate suitability (0 < EI < 30) than high (EI > 30). However, parts of the five major citrus-producing countries in the world (i.e., China, Brazil, Mexico, India, and the USA) will continue to remain highly suitable for the pest until 2070. None of these countries has reported T. erytreae72. Nevertheless, our study will guide policymakers and plant protection and regulatory services to implement quarantine and preventive measures to avoid future invasion by the pest.
Trioza erytreae and the agent it transmits (CLaf), are heat-sensitive 64, and the development of symptomatic leaves varies based on temperature 16. In addition, detecting the disease in leaves is problematic because ‘Ca. L. species’ are of low concentration and unevenly distributed within their host tissues 80, 81. The difficulty in detection may facilitate the spread of CLaf in new areas when T. erytreae is introduced through pathways, such as citrus fruit82, nursery stock (live plants) 83, budwood 84, fresh leaves 82, and infected Psyllid 85. Ajene et al. 86 predicted habitat suitability for CLaf in several regions, including Western, Eastern, and sub-Saharan Africa in Africa; South and Central America in the Americas; the United Kingdom and the Iberian Peninsula in Europe, Australia, and Asia. Similarly, these areas were predicted to be suitable for T. erytreae, the disease's primary vector. Because these regions are suitable for both T. erytreae and CLaf, there is a need to develop proactive measures against the pest and the disease to safeguard the citrus industry 76.
It is important to note that there are limiting abiotic and biotic factors, such as natural enemies; entomopathogens, predators, parasitoids; human factors, urban accessibility; and host plant availability; geographical factors; vegetation, land cover, and elevation, that may limit the distribution of the pest, but were not included in the current model. Other factors that were not considered, include competition with native species, propagule pressure, gene editing, sterilized inset technology, farm-level management strategies, policies, and quarantine measures. All these factors should be considered when interpreting the current study results. Richard et al. 50 found that including remotely-sensed vegetation in a MaxEnt model improved the prediction of suitable areas of T. erytreae. Therefore, we recommend that future studies include these factors in the CLIMEX model when predicting the suitable areas of T. erytreae.
As predicted in the current study, the suitable areas in Ghana, Swaziland, Zimbabwe, the Democratic Republic of the Congo, and the Madagascar Mediterranean coast of North Africa, where citrus is cultivated, require plant protection and regulatory plans. Moreover, these areas need regular inspection or testing for pests, prohibition of portions of the host, a pre-entry or post-entry quarantine system, stated conditions on consignment preparation, specified treatment of the consignment, restrictions on end-use, distribution, and entry periods of the commodity are among the measures that can be implemented 76.