A survey of papers published prior to September 2018 dealing with the topic of biological invasion in Caatinga environments was carried out using the databases Web of Science (http://apps-webofknowledge.ez19.periodicos.capes.gov.br/), Scielo (Scientific Electronic Library Online) (https://scielo.org/) and Google Scholar (https://scholar.google.com/). Searches were performed using the following keywords in Portuguese and English: biological invasion; invasive species; exotic; invasive plants; Caatinga.

Papers were selected based on the title and abstract and, in cases of doubt, the entire article. Generic papers on biological invasion were excluded. Papers, in which the authors did not make explicit whether the species were actually invasive or exotic for the Caatinga, were also excluded.

All results are based on the papers selected from the three databases cited above; however, the data presented here are not equivalent to the number of papers found, but are equivalent to the information included in the studies. For example, some papers may have involved more than one geographic area or the same article may have addressed several invasive species.

The resulting documents represent the total number of papers published on the subject. The year of publication of each article was verified so that a posteriori graph could be generated to show any increase or decrease in publications over years. The main topics addressed by authors were determined from the research objectives presented in the papers. After analysing the papers, we defined categories of themes according to the objectives presented by the authors. The themes addressed were: strategies used in the invasion process; characterisation of the invaded environment; impacts on plant diversity; population structure and/or dynamics; ethnobiological studies; literature review; list of exotic/invasive species; sociability assessment; control of invasive species; mutualistic studies; assessment of level of naturalisation; and ecological niche modelling. Information about the invaded environment and the main strategies used in the invasion process by plants can serve as important predictors of invasiveness.

We made a list of exotic and invasive plants that occur in the northeast region based on the species mentioned in the papers. The species most cited in the analysed papers received greater attention in our discussion.

The lack of published works on invasive exotic plants in the Caatinga prior to 2006 may reflect the generally late discussion of invasions in Brazil. The issue of biological invasion began to be strongly debated in an international context in the 1960s following the publication of Charles Elton’s “Ecology of Invasions by Animals and Plants” in 1958 (Richardson and Pysek 2008; Richardson 2011). The emergence of important international research programmes on biological invasions in the late 1980s contributed to the consolidation of this area of study with a significant increase in the number of publications worldwide between 1995 and 2000 (Richardson and Pysek 2008; Richardson 2011; Hui and Richardson 2017).

Nonetheless, it was not until the 1990s that the problem of invasions began to be discussed scientifically in Brazil (Petenon and Pivello 2008), with the first publications on the subject occurring in 1999 (Zenni et al. 2016). Between 2004 and 2005, two milestones conferred greater visibility to the theme in Brazil. The first was the development of the Informe Nacional Sobre Espécies Exóticas Invasoras (National Report on Invasive Exotic Species), which aimed to obtain, systematise and disseminate information on invasive species in the country (Zenni et al. 2016). The second was the I Simpósio Nacional sobre Espécies Exóticas Invasoras (First National Symposium on Invasive Exotic Species), which occurred at the end of 2005 (Petenon and Pivello 2008) and was considered one of the first steps in the development of a national programme for the control of invasive organisms (Zenni et al. 2016).

After 2005, the development of research and publications related to invasive animals and plants in the different biomes of Brazil intensified, as expected (Zenni et al. 2016). The number of publications on the Caatinga also increased during this period; however, the increase was slight when compared to other biomes (Dias et al. 2013; Frehse et al. 2016; Zenni et al. 2016). Whereas 83% of all the papers published on the topic in Brazil dealt with the Cerrado and Atlantic Forest, 5% dealt with the Caatinga (Frehse et al. 2016). This quantitative difference may be a reflection of the low number of scientific studies carried out in the Caatinga compared to other biomes over the years (Santos et al. 2011).

However, our results indicate that the number of papers published on invasion in the Caatinga is expressive and has remained constant over a period of 14 years. This scenario may indicate a consolidation of research on the theme in the biome. It is noteworthy that papers on animals are not being considered in the present study and scientific research focused on them can be seen as one of the gaps to be filled.

Some of the analysed papers sought to explain the process of invasion by certain species, such as C. procera (Leal et al. 2013; Sobrinho et al. 2013), C. madagascariensis (Cruz et al. 2016; Sousa et al. 2017), P. juliflora (Miranda et al. 2011; Gonçalves et al. 2013; Nascimento et al. 2014), P. aculeata (Gonçalves et al. 2011; Bezerra et al. 2013) and Sesbania virgata (Cav.) Pers. (Souza et al. 2010), using characteristics linked to germination and seed production. These studies are directly related to the propagation pressure hypothesis, according to which some exotic herbaceous and tree species produce a high number of seeds to occupy an invaded area (Lockwood et al. 2009). Other analysed papers sought to explain the process of invasion by the species C. procera (Fabricante et al. 2013), C. madagascariensis (Araújo et al. 2017; Silva et al. 2017b), Nicotiana glauca Graham (Fabricante et al. 2015b), A. indica and P. juliflora (Silva et al. 2018) by the production of allelopathic substances. These authors verified that substances present in the leaves of these invaders impede seedling germination and development of native species mainly in anthropised areas. Differences related to the acquisition and utilisation of resources between native and invasive exotic species were also evaluated for the Caatinga. For example, studies performed by Oliveira et al. (2014, 2017) showed that P. juliflora uses resources, such as water, light and nutrients, more efficiently than native Caatinga species.

In general, more than one hypothesis was tested for the same invasive species, suggesting that the use of multiple strategies that act together determines successful invasion of the Caatinga. Therefore, understanding the different strategies used by invasive species is extremely important since successful invasion is closely linked to the ecological and historical characteristics of the species (Van Kleunen and Richardson 2007).

Studies dealing with population and community structure, characterisation of the invaded environment, impacts on the diversity of plants and listing of exotic and invasive species were also highlighted amongst the studies. Papers that deal with community structure and dynamics are important for showing how communities behave in the presence and absence of invaders and for highlighting the role of dispersion linked to goat livestock in the process of invasion in the semi-arid region (Pegado et al. 2006). In addition, the checklists discussed here call attention to information beyond that of traditional floristic lists to provide data about the reasons for introduction. Almeida et al. (2015), for example, showed that most of the exotic and invasive plants present in the Caatinga were deliberately introduced, many for use as fodder for goats and cattle.

Although few, studies on the control of invasive exotic species in the Caatinga (Fabricante et al. 2009; Gonçalves et al. 2015) are important and need to be encouraged. Studies that deal with the main strategies used in the invasion process, the structure of the invaded community and lists of exotic and invasive species can represent a good theoretical basis for the development of new strategies, policies and studies on control. Using available information, identifying patterns and applying them in different contexts increase the chances of predicting and controlling future invasions (Crandall and Knight 2018).

Another point to be highlighted is the presence of studies that addressed the theme of ethnobiology. Plant invasions have to be treated as complex socioecological phenomena (Vaz et al. 2017). Many introduced species are useful in new geographic areas by providing resources that are indispensable for local communities (Santos et al. 2014; Vaz et al. 2017). Santos et al. (2014) showed that exotic and invasive species are used even more for fodder, medicinal and food purposes in the studied communities than native species. Studies, such as Santos et al. (2014), emphasise the positive value that these species can have within a local context, as well as the need to include local communities in regional and national strategies to deal with invasive species management.

Prosopis juliflora, popularly called algarobeira, belongs to the family Fabaceae (subfamily Mimosoideae) and occurs naturally in Mexico, Central America and northern South America (Peru, Ecuador, Colombia and Venezuela). In addition to these regions of origin, the species was introduced for fodder and wood crops in Brazil, Sudan, Sahel, South Africa and India (Andrade et al. 2010). In Brazil, the species is cultivated mainly in the northeast region and was originally introduced in Serra Talhada, Pernambuco, in 1942 with seeds coming from Piura, Peru (Gomes 1961). Some of the main characteristics of this species have been highlighted, including its aggregate distribution pattern and zoochoric dispersion syndrome (Andrade et al. 2009). Individuals occur with high frequency, dominance, density and importance value (IV) in anthropised areas relative to native species (Pegado et al. 2006). The species exhibits preferences for certain areas, such as those with wetter soils, since they are extremely efficient in capturing and utilising water from these environments (Vilar 2006). Thus, P. juliflora forms high population densities and is a better competitor in moist, anthropised and partially anthropised environments, where it affects the floristic composition, diversity and structure of autochthonous communities (Nascimento 2008).

Cryptostegia madagascariensis, belongs to the family Apocynaceae and is a native plant of the island of Madagascar, Africa, popularly known as unha-do-diabo (Devil’s claw) (Vieira et al. 2004). According to Sousa et al. (2016), this species has caused great damage to native vegetation since it has high biomass production, which prevents the passage of light to neighbouring species and thus kills by shading. The environmental preferences of the species include anthropised areas, swamps and areas of riparian forests where if forms high population densities that prevent the recruitment of local species by limiting resources (Andrade 2013).

Parkinsonia aculeata, commonly known as turco (Turk), also belongs to the family Fabaceae, but in the subfamily Caesalpinioideae (Fabricante and Feitosa 2010). It is native to arid, semi-arid and sub-humid areas between the southern United States and northern Uruguay (Hawkins et al. 2007). It has been pointed out that, amongst its main characteristics, individuals exhibit greater root growth than stem growth and thus are good invaders of floodable or temporarily floodable and humid areas, where the species forms high population densities (Bezerra et al. 2013, Fabricante and Andrade 2014). Its seeds are not negatively affected by specific environmental conditions, such as water deficiency and salinity and, in some cases, soil salinity has been found to even stimulate emergence, initial growth and chlorophyll production (Gonçalves et al. 2011). This characteristic contributes significantly to reducing density and simplifying the composition of the seed bank of an invaded area (Fabricante and Andrade 2014). It has also been observed that traditional control methods, such as cutting and burning, are ineffective, because they cause emergence of various stems (Fabricante et al. 2009).

Calotropis procera belongs to the family Apocynaceae and is popularly known as algodão-de-seda (silk cotton) or algodão de praia (beach cotton). It is native to Africa, the Arabian Peninsula and Southwest Asia and is currently considered an invasive species in Brazil (Rahman and Wilcock 1991; Chaudhary and Al-Jowaid 1999). According to Fabricante et al. (2013), the species is characterised as an invasive of anthropised environments where it forms high population densities. Its main characteristics are rapid establishment, production of large quantities of fruits and seeds and a high germination rate, plus there are indications that its tissues possess allelopathic substances. Given these characteristics, it can be affirmed that the species negatively affects the resilience of anthropised and subsequently invaded environments.

In general, the authors reported that the cited invasive species are often related to areas that are or were once anthropised (Pegado et al. 2006; Andrade et al. 2009; Andrade et al. 2010; Sousa et al. 2016; Souza et al. 2016a; Souza et al. 2016b). The Caatinga currently suffers strong anthropogenic pressures due to the exploitation of natural resources and intense human occupation (Silva et al. 2017a; Ribeiro et al. 2015). Exploitation for firewood for coal production, collective agriculture and overgrazing are some of the main chronic disturbances responsible for reduced biomass of local communities (Leal et al. 2005).

This scenario is worrying since natural or anthropogenic disturbances remove native species from communities, opening space for invasion (Mack et al. 2000). These disturbances can favour the establishment of exotic species with life cycles that are fast and/or highly productive, leading to the successful invasion of disturbed areas (Jauni et al. 2015). Another explanation for this process is that the removal of native species by disturbances increases the local availability of resources (Davis et al. 2000). The high efficiency of resource utilisation exhibited by some exotic species can certainly benefit them (Funk and Vitousek 2007) and allow them to colonise disturbed areas (Jauni et al. 2015).

Anthropisation is recognised as an important variable responsible for facilitating the process of biological invasion in Caatinga areas. Thus, policies aimed at the conservation of the Caatinga as a whole are necessary, since the recovery of managed areas occurs with greater success in places with relatively little anthropogenic disturbance (Prior et al. 2018). On the other hand, it is also necessary to draw attention to areas that have not yet been invaded, in which case, prevention is considered the most positive environmental strategy, although usually hampered by the impossibility of separating exotic invasive species from non-invasive in certain regions (Hulme 2016). In this sense, another point deserving due attention is the quantity of species listed here as invasive in the Caatinga. In the absence of an official list of invasive species for Brazil, studies such as the present are important for reducing uncertainties about local invasive species, as well as supporting adequate management decisions (Dias et al. 2013).

It is important to understand that local communities in tropical semi-arid areas appreciate invasive plants, particularly as animal fodder and for their medicinal properties (Santos et al. 2014). This point has to be taken into account since the perception of benefit or harm depends on social context (Lopes 2017; Vaz et al. 2017). Since invasive plants can provide direct economic benefits to these communities, in the same way as their removal could result in unexpected economic costs, several conflicts can arise when a species is withdrawn (Dickie et al. 2014). For a management plan to be efficient, local communities that directly use invasive species must be included (Santos et al. 2014) and alternatives to the use of these species should be offered.

Finally, the present study can serve as a source for future consultations on the topic. Using our paper, researchers interested in plant invasions in the Caatinga can check what has already been done and dedicate themselves to filling knowledge gaps. Information on the main species considered invasive and the main strategies used in the invasion process, along with the understanding that anthropised environments are more susceptible are of great relevance, can help environmental authorities and professionals make decisions, based on scientific data regarding problems related to the invasion of plants. Thus, it is important to note that investments in research aimed at public universities have been generating important products on the subject for the scientific community and society as a whole. Without investments and the strength of researchers in the region, the rise observed in the number of studies over the 14-year period would not have taken place.