Abstract: Global warming caused by elevated carbon dioxide (CO₂) is a major environmental and policy issue.  The current global average temperature has been elevated by 1°C since the industrial revolution, and it is likely to reach a temperature increase of 1.5°C between 2030 and 2052 (IPCC 2018).  Human-caused emission of CO₂ is responsible for the greenhouse effect and the atmospheric CO₂ concentration is higher now than at any other time in the past 500 000 years, and it continues to rise (Lüthi et al. 2008).  Impacts of arthropod–plant interactions on carbon dynamics and the global climate are important but often ignored.  For example, outbreaks of the mountain pine beetle, Dendroctonus ponderosae, in British Columbia during 2000–2020 will cause the release of an estimated 270 Mt carbon and convert the forest from a small carbon sink to a large carbon source (Kurz et al. 2008).   The annual carbon release due to outbreaks of this beetle is almost equivalent to the annual carbon emission from all forest fires occurring in Canada over 1959–1999 (Kurz et al. 2008).
Most studies of arthropod–plant interactions have focused on the effects of ambient CO₂ or abruptly increasing CO₂ concentrations.  In general, these studies show that elevated CO₂ has a positive direct effect on plant photosynthesis and photosynthate production (Bezemer and Jones 1998; Kim et al. 2015; Andresen et al. 2018; Thomey et al. 2019).  Most scientists expect C3 plants to benefit from this additional CO₂ and outcompete C4 species, because the efficiency of C3 photosynthesis increases with increasing CO₂ concentration to a far greater extent than it does in C4 photosynthesis (Hovenden and Newton 2018; Reich et al. 2018).  Yan et al. (2020) found that elevated CO₂ increased photosynthetic rate, nodule number, yield and total phenolic content of Medicago truncatula.  Dong et al. (2018a) reported that elevated CO₂ promoted the yield and nutritional quality of cucumber (Cucumis sativus L.).  After conducting a meta-analysis using 57 articles consisting of 1 015 observations, they found that elevated CO₂ increased the concentrations of fructose, glucose, total phenols, and total flavonoids in the edible parts of vegetables by 14.2, 13.2, 8.9, and 45.5%, respectively, but decreased the concentrations of protein and nitrate, by 9.5 and 18.0%, respectively (Dong et al. 2018b).  Robinson et al. (2012) reviewed the evidence from 170 studies and concluded that plant biomass, C:N ratio, total phenolics and flavonoids increase under elevated CO₂, while N-based secondary metabolites and plant terpenoid concentrations decrease.  Being an important limiting factor for phytophagous arthropods, changes in foliar C-based secondary metabolites (e.g., condensed tannins and phenolics) and N-based chemicals may have major effects on arthropod performance.
Numerous studies have found that elevated CO₂ indirectly influences arthropod performance via the changes in plant chemical composition (Ge et al. 2010; Xu et al. 2013; Wu 2014; Sun et al. 2018).  Wen et al. (2019) observed a significantly longer larval duration and lower fecundity of Nilaparvata lugens in elevated CO₂.  After analyzing 122 studies, Robinson et al. (2012) concluded that elevated CO₂ increases arthropod survival, abundance and relative consumption rate, but it reduces fecundity, relative growth rate and adult weight.  Many chewing pests, such as cotton bollworm (Helicoverpa armigera) and gypsy moth, exhibited lower fecundity, consumption rate and finite rate under elevated CO₂ (Foss et al. 2013; Liu et al. 2017).  The sucking pests, however, displayed varied responses to elevated CO₂.  For example, in aphids, the responses to elevated CO₂ in terms of fecundity, development and population growth varied between different species, different hosts or even different genotypes of the same host (Sudderth et al. 2005; Gao et al. 2008; Guo et al. 2013).  The studies documented above indicated that the chewing arthropods and sap feeders employ different strategies in response to elevated CO₂. 
While it is clear that arthropod–plant interactions are affected by atmospheric CO₂ concentrations, it is currently uncertain whether an abrupt increase in CO₂ causes similar responses as the gradual increase has been observed since the industrial revolution.  A recent study of Bromus inermis (a perennial grass) and its associated arbuscular mycorrhizal fungi (AMF) shows that abrupt and gradual CO₂ change regimes may not elicit the same response (Klironomos et al. 2005).  In a long-term 6-year experiment in which plants were exposed to three CO₂ regimes (ambient CO₂, gradual increase in CO₂, and abrupt increase in CO₂) for 21 successive generations, more AMF taxa were lost when CO₂ was raised abruptly than when a gradual increase of the same magnitude was implemented.  The abrupt change in CO₂ resulted in a significant change in mycorrhizal diversity in the first generation, although little change occurred in subsequent generations.  Species richness of AMF was similar in the gradual and ambient CO₂ treatments but was significantly lower in the abrupt CO₂ change treatment (Klironomos et al. 2005).  It is not known whether these effects would be similar in an intact field experiment where fungal meta-community dynamics may come into play and mediate any local species extinctions.  A comparable long-term 3-year experiment (Wu et al., unpublished data) investigating impacts of abrupt vs. gradual increases in CO₂ on life-history traits of N. lugens feeding on rice over 16 successive generations, indicated that the gradual increase in CO₂ treatment can promote the growth and physiological metabolism of N. lugens relative to the abrupt CO₂ increase treatment.  So, the effects of abrupt and gradual CO₂ change regimes on arthropods, plants and their associated organisms could differ because the changes affecting organisms are initially the greatest for the first subsequent generation in the abrupt regime, while the evolutionary responses of the interacting organisms differ between the two regimes.
Current generalizations about the effects of increasing atmospheric CO₂ on arthropod–plant interactions are mainly based on experiments using the abrupt approach.  However, a major assumption of these approaches has not been tested, i.e., whether a single-step increase in CO₂ yields similar responses in arthropod–plant systems as a gradual increase over several decades.  If a sudden increase in CO₂ does not yield a response that is similar to a gradual increase of the same magnitude, some of these generalizations could be affected.  Hovenden and Newton (2018) considered that long-term experiments show unexpected plant responses to elevated CO₂ concentrations.  Therefore, most current research may overestimate the impact of abrupt changes in CO₂ concentrations on the arthropod–plant systems.  We must be cautious when designing experiments and explaining the effects of CO₂ concentrations on the arthropod–plant system, because the magnitudes of responses to environmental changes that are significantly more abrupt may be different than those that would occur in nature.  Therefore, other model systems and intact ecosystems should be used to understand how an increase in atmospheric CO₂ influences interactions between arthropods and their host plants.