Presence and Distribution of Scirtothrips dorsalis Hood (Thysanoptera: Thripidae) in Colombia.

The presence and distribution of Scirtothrips dorsalis was reported in Colombia in the Andean, Caribbean, and Orinoquia regions, from 0 to 1,200 meters of altitude (MASL) in the warm climate zone, with less than 2,000 mm rainfall per year and a temperature above 24°C, which corresponds to the tropical dry forest life zone (TDF). Larvae and adults of S. dorsalis were found on 13 plant species belonging to 12 genera in 12 families, of 181 plant species sampled from 129 genera in 47 families sampled. The botanical species with S. dorsalis presence included cotton (Gossypium hirsutum) L., mango (Mangifera indica) L., chili pepper (Capsicum frutescens) L., bell pepper (Capsicum annuum) L., orange (Citrus sinensis) L., jasmine orange (Murraya paniculata) L., rose (Rosa sp.), and the weeds Sesbania exaltata (Mill.), Phyllanthus niruri L., Ludwigia hyssopifolia Vahl, Euphorbia hypericifolia L., Echinochloa colona L., and Amaranthus spinosus (L.). S. dorsalis prefers young leaves and floral structures, but in cotton it was also associated with squares. The low number of host plants is evident, far from the extensive lists obtained by other authors in other latitudes. S. dorsalis invaded Colombia only in recent years, and this is the first study of the presence and distribution of this thrips in the area.

The family Thripidae (Thysanoptera) includes important pests of a wide number of cultivated plant species, and much direct damage is attributed to them including: distortion of leaves, defoliation, and abortion of floral structures, flowers, and fruits. In addition, they can cause indirect damage, such as the transmission of viruses that can affect yield and quality of crops (EFSA 2014). In the global context, the most important species of Thysanoptera for agricultural production are grouped in the genera Caliothrips Daniel, Retithrips Marchal, Sericothrips Haliday, Thrips Linneaeus, Frankliniella Karny, and Scirtothrips Shull (Leigh 1995, Mailhot et al. 2007). The genus Scirtothrips has 100 described species (ThripsWiki 2017), distributed in the tropical and subtropical areas of the world, with relevant species considered of economic importance for world agriculture such as Scirtothrips aurantii, Scirtothrips citri, Scirtothrips perseae, Scirtothrips inermis, Scirtothrips longipennis, Scirtothrips mangiferae, Scirtothrips oligochaetus, and Scirtothrips dorsalis (Hoddle and Mound 2003, EPPO 2005).

S. dorsalis Hood has been associated with young leaves, flowers, and terminal shoots in more than 200 species of dicotyledonous plants, in about 40 botanical families (Hoddle and Mound 2003). It preferentially feeds on the epidermis, and occasionally, on palisade tissues in young leaves and tissues of the apex of young fruits, especially when they are still hidden under the calyx. In many hosts, it can feed on upper surfaces of leaves when the levels of infestation are high. Larvae and adults are often located toward the main central vein of young leaves or near the damaged area of leaf tissues. Pupae can be found in the leaf litter on the soil, in the axils of the leaves, in curled leaves or under the calyx (Kumar et al. 2013).

The feeding damage of S. dorsalis is characterized by bronzed scars on the leaves, petals, and epidermis of fruits. A concentric ring is often formed in the scarred tissue around the apex, fruit distortion, and premature leaf senescence (Kumar et al. 2013). In addition, it has the capacity to transmit phytopathogenic viral particles of Tospovirus (= Bunyavirales: Tospoviridae: Orthotospoviruses) (Jones 2005, Adams et al. 2017) that cause considerable losses in yields, profitability, and may lead to the implementation of excessive and restrictive control measures in export markets (Riley et al. 2011, Ebratt et al. 2013).

S. dorsalis is native to tropical Asia (Dev 1964, Mound and Palmer 1981) but it currently occurs in Australia, the Solomon Islands, Israel, the Caribbean region (Kumar et al. 2013), North America (Nietschke et al. 2008) and South America in Surinam, Venezuela (Cermeli et al. 2009), and Colombia (ICA 2012). In Colombia, this species has been recorded in Huila, Tolima, and Vichada departments (ICA 2012). S. dorsalis is successful in expanding its geographical distribution range due to passive transport through human activities, new sites to colonize derived by climate change, and the lack of predators or specific plant defenses in its host that restrict its colonization (Kumar et al. 2013). This could be considered a biological invasion, as it is a nonnatural distribution process caused by human commercial activity (Gutiérrez 2006). Rodríguez (2006) and Kenis et al. (2008), recognize that the invasion of ecosystems by introduced species is one of the greatest threats to biodiversity and the structure of communities, with direct negative impacts of nonnative species on the native biota, with effects on predation, hybridization, generation of changes in ecosystem processes, loss of biodiversity, population imbalances, and competition for habitats.

In order to update the status of this economically important species in Colombia, the purpose of this study was to determine the geographical distribution and identification of plants associated and nonassociated with S. dorsalis in three ecological regions of Colombia where main agricultural crops are established.

Materials and Methods

Geographic Location and Sampling

This study was carried out between 2013 and 2016 and is part of the doctoral thesis project: Diversity and geographical distribution of Thysanoptera: Tripinae with emphasis on Frankliniella Karny in Colombia. Samples were collected covering ranges of 500 meters of altitude from 0 to 3,000 meters above sea level (MASL), in several life zones such as tropical dry forest, lower montane dry forest, lower montane moist forest, tropical moist forest, premontane moist forest, premontane very moist forest, premontane pluvial forest, lower montane very moist forest, tropical very dry forest, and premontane transitional moist forest in the Andean region; tropical dry forest, tropical very dry forest, tropical moist forest, premontane dry forest, premontane moist forest, tropical thorn forest in the Caribbean region and, the tropical dry forest and the tropical rain forest zone of the Orinoquia region (Holdridge 1967, IAvH 2014) in different agro-ecosystems of the Andean, Caribbean, and Orinoquia regions of Colombia (Fig. 1A).

Fig. 1.

Presence of S. dorsalis in Colombia: A. Total distribution of sampling points (N = 1,227). B. Points with presence of S. dorsalis in Colombia (N = 169).

For the development of the monitoring activities, we conducted collection tours to each geographical region and in different agricultural areas of 23 departments in Colombia. Six agricultural crops were selected, 10 plants per hectare were sampled and in each plant three structures. For the weeds and wild plants one to five plants were sampled. In total, 11,590 of entomological samples were collected in 4,555 monitoring points; 674 entomological samples corresponded to cotton, 33 to chili pepper, 33 to bell pepper, 221 to mango crops, 75 to orange, 225 to rose flower, 608 for weeds, and 9,721 to other plant species.

Samples were collected by beating of leaves, floral structures, and fruits in different agricultural production systems, in weeds, backyard, and wild plants using a 000 caliber brush moistened with 70% ethanol, and white plastic plates were used to collect the individuals. Finally, the specimens were deposited in 1.5 ml Eppendorf tubes with 70% ethanol. Each sample was assigned with a unique identification code.

Treatment of the Sample

From the specimens obtained, the larval stages were separated from the adults. The thrips were sorted by morphotypes and they were identified. The specimens were cleared in cold 5% KOH, followed by washing with distilled water. Subsequently, larvae and adults were mounted in Hoyer’s medium in slides, and dried in the oven for 48 h at 35–40°C, according to Mound and Marullo (1996). Few specimens of each species were mounted in Canada Balsam for the permanent collection according to Mound and Marullo (1996). The taxonomic determination of adults was carried out based on morphological characters according to available keys (Hoddle and Mound 2003, Kumar et al. 2013) and larvae were identified based on Vierbergen et al. (2010) with the help of a Leica ZOOM 2000 stereoscope and a Nikon morphological type-119YS2-T microscope.

Identification of Plants

Taxonomic keys of the Flora of Colombia were used (Gentry 1996, Bernal et al. 2015) and comparisons were made with the herbarium of the Institute of Natural Sciences of the National University of Colombia.

Data Analysis

The association of S. dorsalis with the plants according to the botanical family, genus, species, and plant structure was also determined. Based on the abundance of adults in plant structures (young leaf, floral bud, flower, and fruit), the similarity among them was determined by means of Bray–Curtis similarity index (Ramirez 2006, Pulido et al. 2015). The presence of S. dorsalis was correlated with the climatic variables temperature (°C—annual average), precipitation (mL—annual average), and altitude (MASL) by means of Spearman ranges; in addition, the probability of encounter based on the climatic variables was determined through logistic regressions. The software ArcGIS 10.1 (ArcMap version 10.1), and the programs STATISTICA v10, InfoStat v2008, Past v3, and SAS v9.2 were used.

Results

Presence of S. dorsalis in Colombia

The presence of larvae and adults of S. dorsalis was recorded in 169 (13.77%) of 1227 sampled sites (Fig. 1A). One hundred and thirty two sampled sites (78.24%) in the Andean region, especially in the warm valley of the upper Magdalena (departments of Tolima, Huila, and Cundinamarca); 27 (15.88%) in the Caribbean region in the departments of Córdoba, Sucre, Bolívar, Magdalena, and Cesar; and 10 (5.88%) in the Orinoquia region, in the natural savannah of the municipality of Puerto Carreño, northeast of the department of Vichada (Fig. 1B). All samples of S. dorsalis were collected in the tropical dry forest life zone (TDF).

Geographic and Altitudinal Distribution

S. dorsalis was reported in the Andean, Caribbean, and Orinoquia regions with an estimated altitudinal distribution from 0 to 1,200 meters (n = 295; r = −0.8003; P = 0.00001), average annual temperatures higher than 24.4°C (n = 295; r = 0.7789; P = 0.00001), and average annual precipitation less than 1683.4 ml (n = 295; r = −0.4554; P = 0.00001) (Table 1, Fig. 2). The logistic regression model adjusted for the variable temperature (P ( = 1/1 + e−(−31.798 + 1.237)) suggests that there is a probability ≥50% (r2 = 0.8462, area under the curve (AUC) = 0.760; P < 0.001) of absence of S. dorsalis in areas with temperature ≤20°C, and altitudinal ranges above ≥1,200 MASL (P ( = 1/1 + e−(−0.813 − 0.002)) (r2 = 0.9987; AUC = 0.685; P = 0.001) (Fig. 3).

Table 1.

Plants host of S. dorsalis in Colombia in the tropical dry forest (TDF) life zone

Plant Species	Family	Region	Department	Municipality	Altitude	Latitude	Longitude	System	Number of monitored plants	Number of occurrence	Number of adults	Number of larvae
Amaranthus spinosus	Amaranthaceae	Caribbean	Cesar	Bosconia	114	10,05258	−73,86875	Weed	4	1	1	6
Capsicum annuum	Solanaceae	Andean	Huila	Pital	960	2,18815	−75,81784	Crop	10	1	2	3
			Tolima	Coello	486	4,27647	−75,01584		10	1	5	5
Capsicum sp	Solanaceae	Caribbean	Bolivar	San Jacinto	265	9,82972	−75,13525	Crop	10	1	2	1
				Carmen de Bolívar	369	9,73939	−75,23635		10	1	2	1
Citrus sp	Rutaceae	Caribbean	Magdalena	Zona Bananera	40	10,76546	−74,14754	Crop	10	1	1	12
		Andean	Tolima	Espinal	318	4,20598	−74,88172		10	1	7	4
Echinochloa colona	Poaceae	Andean	Tolima	Espinal	388	4,12716	−74,84169	Weed	3	1	4	5
Euphorbia hypericifolia	Euphorbiaceae	Andean	Huila	Villavieja	387	3,19616	−75,22933	Weed	3	1	5	4
Gossypium hirsutum	Malvaceae	Andean	Tolima	Armero	312	5,01427	−74,90511	Crop	10	2	5	1
					368	5,11144	−74,89405		10	2	6	1
				Espinal	318	4,15219	−74,85444	Crop	10	4	74	23
					356	4,17713	−74,92647		10	1	3	1
					319	4,13430	−74,84708		10	2	27	9
					342	4,15691	−74,94863		10	2	4	5
				Natagaima	348	3,49755	−75,13863	Crop	10	4	33	7
					347	3,50275	−75,14116		10	4	22	2
					348	3,50591	−75,13522		10	2	17	2
					344	3,48452	−75,13911		10	2	10	8
					347	3,49030	−75,13186		10	2	9	4
					333	3,68419	−75,10452		10	3	18	9
					333	3,67433	−75,10672		10	4	54	2
					329	3,67652	−75,10160		10	4	17	4
				Coello	263	4,42388	−74,83913		10	1	8	10
			Huila	Villavieja	374	3,38175	−75,17119	Crop	10	1	3	3
					373	3,37544	−75,17169		10	3	6	2
					387	3,20475	−75,23777		10	1	1	16
					402	3,21501	−75,23422		10	2	7	1
					387	3,21638	−75,23098		10	1	9	1
					383	3,21877	−75,23001		10	2	7	1
					386	3,21363	−75,22788		10	2	29	1
					376	3,35444	−75,19886		10	3	15	3
					377	3,35588	−75,23041		10	3	12	6
					365	3,33069	−75,18863		10	1	2	9
					364	3,34506	−75,17505		10	1	3	13
					361	3,34829	−75,18559		10	1	4	3
					361	3,34933	−75,18137		10	2	6	1
					362	3,34813	−75,18202		10	1	5	5
					366	3,34512	−75,17342		10	1	4	4
				Campoalegre	519	2,69927	−75,35213	Crop	10	4	25	2
					518	2,69986	−75,35522		10	3	7	2
					509	2,70352	−75,35644		10	2	4	5
					498	2,70513	−75,35086		10	3	19	2
					507	2,71088	−75,33585		10	2	18	6
					547	2,72230	−75,31708		10	4	17	2
					526	2,68813	−75,36094		10	4	32	1
					527	2,69208	−75,36188		10	4	17	1
					525	2,69238	−75,36008		10	3	15	1
					532	2,68472	−75,36677		10	2	7	1
				Aipe	371	3,23345	−75,22752		10	1	5	5
			Cundinamarca	Ricaurte	302	4,28602	−74,74352	Crop	10	2	38	8
					301	4,28497	−74,74219		10	2	41	15
					301	4,28397	−74,74088		10	2	28	3
					305	4,29211	−74,73677		10	2	33	7
					311	4,28941	−74,73402		10	2	16	5
					345	4,29247	−74,73575		10	1	18	9
					283	4,29747	−74,74183		10	4	36	2
					291	4,30775	−74,74889		10	3	48	5
					297	4,30355	−74,74436		10	1	1	3
					290	4,28083	−74,73419		10	1	4	3
		Orinoquia	Vichada	Puerto Carreño	47	6,19768	−67,46557	Crop	10	1	5	3
					49	6,19682	−67,46499		10	1	1	1
					56	6,25074	−67,53278		10	1	10	1
					48	6,19769	−67,46581		10	1	4	3
					48	6,19735	−67,46611		10	1	5	5
					55	6,25124	−67,53229		10	1	1	5
					55	6,25148	−67,53225		10	1	2	10
					52	6,19699	−67,46453		10	1	1	1
					58	6,25049	−67,53239		10	1	4	4
					49	6,19704	−67,46591		10	1	7	2
					42	6,19589	−67,46559		10	3	15	7
					46	6,20784	−67,48207		10	2	5	1
		Caribbean	Cesar	Bosconia-Copey	114	10,05258	−73,86875	Crop	10	1	8	6
			Sucre	Sampués	111	9,12620	−75,25859	Crop	10	1	0	9
			Cordoba	Montería	17	8,77209	−75,88448	Crop	10	3	11	3
					12	8,74107	−75,80349		10	8	170	1
					11	8,78275	−75,85408		10	7	84	7
Ludwigia hyssopifolia	Onagraceae	Andean	Tolima	Espinal	388	4,12716	−74,84169	Weed	5	1	3	4
Mangifera indica	Anacardiaceae	Caribbean	Magdalena	Zona Bananera	40	10,76546	−74,14754	Crop	10	1	29	2
		Andean	Tolima	Coello	486	4,27647	−75,01584		10	1	3	5
Murraya paniculata	Rutaceae	Caribbean	Magdalena	Zona Bananera	40	10,76546	−74,14754	Backyard	10	1	1	5
Phyllanthus niruri	Euphorbiaceae	Andean	Tolima	Armero	368	4,68061	−74,90861	Weed	3	1	4	3
Rosa sp	Rosaceae	Caribbean	Bolivar	Carmen de Bolívar	310	9,73419	−75,22069	Backyard, Garden	4	1	3	13
		Andean	Tolima	Flandes	302	4,23245	−74,86134		5	1	4	21
				Guamo	303	4,01634	−74,92001		5	1	2	14
Sesbania exaltata	Fabaceae	Andean	Huila	Villavieja	383	3,21127	−75,22466	Weed	1	1	7	2
TOTAL									809	169	1267	424

Fig. 2.

Abundance of S. dorsalis in correlation with altitude (MASL), average annual temperature (°C), and average annual precipitation in Colombia.

Fig. 3.

Estimated probability of the presence of S. dorsalis in Colombia, at different temperatures (°C) (N = 1,057; r2 = 0.8462; area under the curve (AUC) = 0.760; P < 0.0001) and altitudinal gradient in meters (N = 1057; r2 = 0.9987; AUC = 0.685; P = 0.001).

Distribution According to the Life Zone

One hundred percent (100%) of the positive reports of S. dorsalis were located in the TDF in the Andean, Caribbean, and Orinoquia regions (Table 1).

Host Botanical Species and Preferences for Structures

Of the 184 botanical species sampled, which were grouped into 131 genera and 44 families, S. dorsalis was found in 13 species of 12 genera and 10 botanical families, all angiosperms. The 13 botanical species were related to agricultural systems in cotton, orange, bell pepper, chili pepper, and mango crops; the presence in plants of Rosa spp. It was also evidenced in gardens and backyards in the municipalities of Carmen de Bolívar (Department of Bolívar, Caribbean region), Flandes, and Guamo (Department of Tolima, Andean region). All plant species with presence of S. dorsalis corresponded to true host plants due to the report of larvae stages and suggest a first list of host plants for the presence in the botanical families Malvaceae, Anacardiaceae, Solanaceae, Rutaceae, Poaceae, Amaranthaceae, Euphorbiaceae, Onagraceae, Rosaceae, and Fabaceae (Table 1).

Larvae and adult stages of S. dorsalis were found in leaves, flower buds, and squares of cotton (Gossypium hirsutum); in leaves and inflorescences of mango (Mangifera indica) crop; in leaves and flowers of orange (Citrus sinensis), and in leaves and flowers of bell pepper and chili pepper (Capsicum annuum, Capsicum frutescens). S. dorsalis prefers young leaves (38.46%, n = 65 occurrence), followed by flowers (22.48%, n = 38 occurrence), fruits (21.89%, n = 37 occurrence in cotton squares), and flower buds (17.16%, n = 29 occurrence). The Bray–Curtis similarity index showed that S. dorsalis preferred young leaves and buds, compared with the other structures of the host plants (Fig. 4).

Fig. 4.

Preferences for plant structures, using the Bray–Curtis similarity index in 13 botanical species associated with S. dorsalis in three geographic regions of Colombia (N = 169; Cophenetic corr. = 0.948; read cases = 169; omitted cases = 0; P < 0.05).

In this study, S. dorsalis is nonhost in 175 botanical species in 119 genera (Table 2).

Table 2.

Species and botanical families sampled and nonhost with S. dorsalis in the Andean (A), Caribbean (C), and Orinoquia (O) regions

Plant species	Family	Region	Number of monitored plants	Plant species	Family	Region	Number of monitored plants	Plant species	Family	Region	Number of monitored plants
Acacia sp	Fabaceae	A	2	Emilia sp	Asteraceae	A,O	2	Pisum sativum	Fabaceae	A	5
Ageratum sp	Asteraceae	C	1	Euphorbia catinifolia	Euphorbiaceae	C,O	2	Pithecellobium guachapele		A	1
Allium cepa	Amaryllidaceae	A	4	Euphorbia heterophylla		C	1	Priva lappulacea	Verbenaceae	A	1
Allium fistulosum		A	1	Euphorbia hirta		A	5	Prunus domestica	Rosaceae	A	2
Allium sativum		A	3	Euphorbia postrata		A	1	Prunus persica		A	3
Amaranthus sp		A	1	Fragaria ananassa	Rosaceae	A	1	Prunus sp		A	1
Amaranthus dubius		A	5	Funastrum clausum	Apocynaceae	C	1	Psidium guajava	Myrtaceae	O	9
Anthurium sp	Araceae	A	1	Furcraea andina	Asparagaceae	A	4	Pyrus communis	Rosaceae	A	1
Arachis pintoi	Fabaceae	A, C	4	Furcraea sp		A	1	Quercus sp	Fabaceae	C	2
Asparagus officinalis	Asparagaceae	A	1	Galinsoga parviflora	Asteraceae	A	1	Raphanus raphanistrum	Brassicaceae	A	7
Baccharis prunifolia	Asteraceae	A	1	Galinsoga sp		A	1	Raphanus sp		A	5
Bauhinia forficata	Fabaceae	C	1	Gliricidia sepium	Fabaceae	A,C,O	9	Rhynchosia minima	Fabaceae	A,C	7
Bidens pilosa	Asteraceae	A	1	Hedychium coronarium	Zingiberaceae	A	1	Ricinus communis	Euphorbiaceae	A	1
Bixa orellana	Bixaceae	A	10	Helianthus annuus	Asteraceae	A	1	Rubus glaucus	Rosaceae	A	26
Boerhavia decumbens	Nyctaginaceae	A	1	Heliotropium angiospermum	Boraginaceae	A,C	3	Rubus sp		A	1
Boerhavia erecta		A	1	Heliotropium indicum		C	1	Ruellia tuberosa	Acanthaceae	A	1
Borreria capitata	Rubiaceae	C	2	Heliotropium sp		A	1	Ruta sp	Rutaceae	A	1
Borreria laevis		A	1	Hemigraphis alternata	Acanthaceae	C	1	Sambucus sp	Adoxaceae	O	1
Brassica rapa	Brassicaceae	A	1	Hevea brasiliensis	Euphorbiaceae	O	5	Sarcostemma clausum	Apocynaceae	A	1
Caesalpinia coriaria	Fabaceae	C	1	Holcus lanatus	Poaceae	A	1	Saurauia scabra	Actinidiaceae	A	1
Caesalpinia sp		C	1	Hypochaeris radicata	Asteraceae	A	1	Scoparia dulcis	Scrophylariaceae	A	1
Calliandra surinamensis		C	1	Hyptis mutabilis	Lamiaceae	C	1	Senna occidentallis	Fabaceae	A	1
Calliandra trinervia		C	1	Inga densiflora	Fabaceae	C	1	Senna sp		A,C	7
Calotropis gigantea	Apocynaceae	C	2	Inga edulis		A	2	Senna viarum		A	1
Canna indica	Cannaceae	A	1	Ipomoea pes-caprae	Convolvulaceae	O	3	Sida acuta	Malvaceae	A	1
Cannabis sp	Cannabaceae	A	1	Ipomoea sp		A	1	Sida parviflora		A	1
Cassia aeschynomene	Fabaceae	A	1	Kallstroemia maxima	Zygophyllaceae	A	2	Sida rhombifolia		A	2
Cassia fistula		C	1	Lachemilla orbiculata	Rosaceae	A	1	Silybum marianum	Asteraceae	A	1
Centrosema acutifolium		A	2	Lactuca sativa	Asteraceae	A	1	Solanum tuberosum	Solanaceae	A	8
Chamaecrista sp		C	1	Lagascea mollis		A	2	Solanum crotonifolium		A	1
Chrysanthemum sp	Asteraceae	A	2	Lantana camara	Verbenaceae	A	3	Solanum lycopersicum		A	5
Chrysobalanus icaco	Chrysobalanaceae	A, C	2	Lantana sp		A	1	Solanum melongena		A,C	6
Cissus verticillata	Vitaceae	A	1	Leptochloa scabra	Poaceae	A	11	Sorghum bicolor	Poaceae	C	1
Citrullus lanatus	Cucurbitaceae	O	1	Licania sp	Chrysobalanaceae	C	1	Sphagneticola trilobata	Asteraceae	A	2
Citrus aurantifolia	Rutaceae	A	6	Ludwigia sp	Onagraceae	A	1	Spinacia oleracea	Amaranthaceae	A	1
Citrus latifolia		A,C,O	8	Malus sylvestris	Rosaceae	A	1	Tabebuia rosea	Bignoniaceae	C	1
Citrus limon		A	2	Malus domestica		A	1	Tachigali guianensis	Fabaceae	A,O	4
Citrus reticulata		A,O	5	Malvastrum americanum	Malvaceae	A	3	Talinum fruticosum	Portulacaceae	A	2
Citrus sinensis		A,C,O	35	Malvaviscus concinnus		A	1	Talinum paniculatum		A	1
Cleome affinis	Cleomaceae	A	1	Manihot esculenta	Euphorbiaceae	A,C,O	50	Tamarindus indica	Fabaceae	C	1
Cleome spinosa		C	1	Melochia parvifolia	Malvaceae	A	6	Taraxacum officinale	Asteraceae	A	5
Cleoserrata sp		A	1	Melochia pyramidata		A	2	Taraxacum sp		A	1
Cnidoscolus urens	Euphorbiaceae	A	1	Momordica charantia	Cucurbitaceae	A	2	Tecoma stans	Bignoniaceae	A	2
Coffea arabica	Rubiaceae	A,C	31	Musa paradisiaca	Musaceae	A,C,O	65	Tephrosia sp	Fabaceae	A	1
Cordia alliodora	Boraginaceae	A	1	Musa sp		C,O	2	Thevetia peruviana	Apocynaceae	A	1
Cordia dentata		C	7	Nicotiana tabacum	Solanaceae	A,C	2	Tithonia diversifolia	Asteraceae	A,O	4
Coriandrum sativum	Apiaceae	A	3	Parthenium hysterophorus	Asteraceae	A	1	Tithonia sp		A,O	2
Crescentia cujete	Bignoniaceae	C	1	Passiflora edulis f. flavicarpa	Passifloraceae	A,C,O	39	Tridax procumbens		A,C	6
Crotalaria juncea	Fabaceae	A	1	Passiflora edulis Sims		A	2	Trifolium pratense	Fabaceae	A	3
Croton leptostachys	Euphorbiaceae	A	1	Passiflora ligularis		A	4	Trifolium repens		A	4
Cyathula achyranthoides	Amaranthaceae	A	1	Passiflora maliformis		O	2	Trifolium sp		A	5
Dahlia sp	Asteraceae	A	3	Passiflora quadrangularis		O	1	Turnera ulmifolia		A	1
Daucus carota	Apiaceae	A	1	Passiflora tarminiana		A	3	Vicia faba		A	6
Delechampia scandens	Euphorbiaceae	A	1	Pavonia sepium	Malvaceae	A	1	Zantedeschia aethiopica	Araceae	A	2
Dioclea sp	Fabaceae	A	2	Persea americana	Lauraceae	A,C,O	151	Zea mays	Poaceae	A,C,O	48
Eclipta alba	Asteraceae	A	1	Phaseolus vulgaris	Fabaceae	A,C	19
Emilia sonchifolia		A,C,O	12	Physalis sp	Solanaceae	A	1

Discussion

S. dorsalis is a nonnative polyphagous species of recent report in Colombia (Seal et al. 2010, ICA 2012) with a high capacity to occupy large areas and geographic distribution. It is also a species with a high degree of environmental and phytosanitary risk since it could generate negative effects on the native flora and on the expected yields in cultivated plants of economic interest for the agriculture of Colombia. It is evident that the introduction and establishment of species such as S. dorsalis would be able to affect existing trophic networks and could alter processes of population dynamics that allow the regulation of species (Morse and Hoddle 2006).

The presence of S. dorsalis in the warm climate zone of the Andean, Caribbean, and Orinoquia regions of Colombia, in tropical dry forest areas (TDF) (Holdridge 1967), coincides with the results obtained by other authors (Cermeli et al. 2009) and could provide elements to characterize and potentially delimit the occupation and invasion areas, by recognizing the optimal climatic variables of its ecological niche. In Colombia, tropical dry forests correspond to a plant formation that presents a continuous, nondense forest cover. They are distributed between 0 and 1,000 MASL, with one or two periods of drought per year; they consist of 8,146,000 hectares, which correspond to 11.3% of the territory of Colombia and have distinctive floristic and faunal affinities compared to other life zones (IAVH 2014). According to Hernández (1990), this formation is also known as moist temperate forest, tropical deciduous forest, or drought-deciduous tropical lowland forest in other classification systems. These forests are currently under a high degree of intervention, degradation, use, and extensive over exploitation for agricultural and livestock production.

Taking into account the objective sampling area that was included in this study and that the first report for Colombia was in the year 2012, it was determined that S. dorsalis is not widely distributed, possibly due to the negative influence of climatic conditions and the presence of different geographic barriers.

Temperature is important for the developmental rate of S. dorsalis with a critical threshold between 9.7 and 32°C with an optimal generation rate of 25°C (Tatara 1994, Shibao 1996). In this study, this species was found in an average temperature of 24.4°C in all ecological regions, which means an advantageous condition for its establishment. Additionally, S. dorsalis density has been positively correlated with temperature but negatively correlated with relative humidity and precipitation (Mannion et al. 2014). Because the Andean and Caribbean regions are characterized by a bimodal precipitation regimen with changing seasonal average temperatures, but Orinoquia regions have a monomodal rain regime between April and November with more constant average temperature throughout the year (Pacheco and León-Aristizabal 2001), we expect that geographical dispersion and establishment of S. dorsalis will be differential for these three regions.

It is evident that S. dorsalis does not adapt to cold environments with minimum daily temperatures of −4°C for at least 5 d a year; however, in subtropical and tropical environments with warm climate and prolonged droughts, 18 generations can be obtained per year with high population densities (Nietschke et al. 2008, Kumar et al. 2013). Seal et al. (2010) determined that, under the influence of subtropical environmental conditions, S. dorsalis could reach a generation every 27 d, at temperatures >20°C with food sources preferably based on young tissue of host plants such as jalapeño pepper and rose. This circumstance makes the arrival of S. dorsalis, like the arrival of any other invasive organism, reach a level of important phytosanitary risk, given the difficulty to evaluate the impacts generated on natural, intervened, and agricultural ecosystems.

Climatic conditions, which could restrict the presence and establishment of populations of S. dorsalis in the geography of Colombia, have to do with average annual temperatures below 24°C, rainfall above 2,000 mm per year and interactions related to the altitude-temperature at altitudes above 1,200 MASL. These conditions could negatively influence the biology and behavior of the insect and become a natural environmental limit for the distribution, establishment, and reproduction of the species, since they become unfavorable habitats (Nietschke et al. 2008). According to Pulido et al. (2015), to determine these habitats, it is necessary to make predictive models to recognize the distribution according to the ecological niches and potential occupation areas of S. dorsalis.

The larval stages of S. dorsalis were found in 13 plant species that constitute a true hosts (Table 1), where S. dorsalis can develop and complete its life cycle (Mound and Marullo 1996, Marullo 2004, Alves-Silva et al. 2013, Burckhardt et al. 2014). Our results are far less diverse than the reports of CABI (2018) and USDA-APHIS (2006), since S. dorsalis infests a wide variety of host plants belonging to more than 100 plant taxa among 40 families (Mound and Palmer 1981). These records does not necessarily make them true host plants, without the evidence of immature states present in them. The above is relevant, considering that Colombia is the second country with the highest plant biodiversity in the world, with heterogeneous local habitats and isolations, which historically have favored high rates of speciation and endemism in many taxa, benefited by the of climates, floristic mosaics, and biogeographical overlaps between the Andean, Orinoquia, Guyanese, Caribbean, Pacific, and Amazonia regions (Gentry 1988, Josse et al. 2012).

S. dorsalis showed preferences for young leaves and flower buds of different crops, weeds, and wild plants. The preferences for structures were oriented toward young leaves and flower buds in chili pepper, bell pepper, mango, jasmine orange, orange, and rose. In cotton crops, it was reported in the young leaves, flower buds, and fruits. This agrees with Seal et al. (2006) and this species prefers to feed on tender tissues in young leaves in some plant species, and could be limited by the adverse conditions to low temperatures and high rainfall. In general, young leaves infested by thrips have high contents in soluble protein and total nitrogen (Paine 1992). Schoonhoven et al. (2005) propose that high concentrations of secondary metabolites support the direct defense of the plant, which is effective against several species of herbivores, so insect populations must develop efficient strategies for detoxification or regulation of their intake.