Geographic distribution and the invasive scope of aquatic plants in México

Aquatic plants are expanding their geographic distributions necessitating updated inventories that include evaluations of potential impacts on biodiversity, environmental services, and water management. This study describes the geographic distribution of native and non-native aquatic plants from 11 hydrological regions and basins of México, including 56 reservoirs, 14 lakes, two irrigation channels, two drainages, and one river. Forty-two aquatic plant species were recorded, of which eleven were non-native and invasive, and 31 were native. Of the native species, all but four are considered to have the potential to become invasive in novel habitats. The estimated percent surface coverage by different species of plants recorded in this survey ranged from 5% to 100% of the water bodies. Eichhornia crassipes (water hyacinth) was the most common, non-native species observed in the Mexican inland waters. This is the first report on the geographic distribution of invasive aquatic plants that affect biodiversity, water availability and uses in México.


Introduction
In México, 800 non-natives species have been reported, including 665 (83.12%) plants (IMTA 2008).Species such as water hyacinth, Eichhornia crassipes, (Mart.)Solms; water lettuce, Pistia stratiotes L, giant reed, Arundo donax L; and hydrilla, Hydrilla verticillata (L.f.) Royle, were introduced to México more than 100 years ago, via anthropogenic activities (IMTA 2008).These non-native, invasive plants cause serious economic, ecological, and health problems, including the loss of water and reduction of the storage capacity in reservoirs due to evapotranspiration and increases of sediment filling in the water column (Robertson and Coll 2019).Aquatic plants may also limit fishing and recreational activities, and clog irrigation channels and water supply systems in multipurpose or irrigation infrastructure (Gopal 1987;Rowantree 1991).Ecological problems include the accumulation of large quantities of aquatic plants resulting in water stagnation, the reduction of dissolved oxygen, and ultimately, the death of other aquatic species (Barrett and Forno 1982).They may also provide habitat for disease vectors that cause some illnesses such as dengue, filariasis, helminthiasis, encephalitis, malaria, and yellow fever (Hernández and Pérez 1995).
Two of the most important factors in invasion success are propagule pressure and dispersal pathways (Lockwood et al. 2005;Colautti et al. 2006;Simberloff 2009;Wilson et al. 2009;Cassey et al. 2018).Both factors increase the probability of successful establishment of non-native species, alongside with their strategies of producing large amounts of seeds capable of surviving in sediments for a long time until the environmental conditions allow their successful germination (Leck 2003;Mouton et al. 2019;Szabó et al. 2019).Due to the lack of control or maintenance programs for these plants, there is a yearly progressive accumulation of "seed banks", which allows the regular re-establishment of native or non-native species in the water bodies.In the case of E. crassipes, Albano et al. ( 2011) observed a seed density in soil up to 2,534 seed/m 2 and demonstrated the existence of soil seed banks in reservoirs and rivers in South Africa.Furthermore, the large supply of nutrients from agricultural, urban, and industrial wastewater ensures non-native species growth and dispersion (Kuntz et al. 2014).The high reproductive rate of these species (Colautti et al. 2006), together with the high concentration of nutrients and the absence of natural enemies, result in explosive growth and the coverage of almost the entire surface of many water bodies in México.
In addition, federal and local government agencies and social institutions do not have the permanent, effective, and efficient control programs required to prevent the spread of these species.Information concerning the identification and distribution of non-native or native invasive aquatic plants, and the degree of coverage of water bodies at the national level is rarely available and traceable.International databases on invasive plants (IPNI) and national inventories focus primarily on taxonomy and distribution (Fulmer and Robinson 2008).Recently, aspects of biodiversity, health, and water management are being included (Wang et al. 2016).The objective of this study was to provide a record of native and non-native plant species in water bodies located in 11 hydrological regions of Mexico as well as to highlight how some of these species may threaten water availability by covering, in some cases, more than 50% of the water surface of the reservoirs.To our knowledge, this is the first report that provides an update on the geographic distribution and invasive range of aquatic plants in México, especially those species considered serious threats to water resources and aquatic biodiversity.

Study area
Surveys were conducted in 75 water bodies of 11 hydrological regions of the Mexican Water Authority (CONAGUA) (Figure 1).These hydrological regions comprise water bodies related to irrigation activities, hydropower generation or water supply.The Northern hydrological regions comprise 49% of the national territory with climates from dry to very dry.Temperatures in this region range from 18 °C to more than 26 °C and rainfall ranges from 100 to 600 mm/year.The central part of the country, approximately 23% of the territory, is mild humid and sub-humid, with temperatures ranging from 10 °C to 22 °C and rainfall ranging from 700 to 1000 mm/year.On the coasts and the Southeast of México, the climate is warm humid (28% of the territory), with a temperature higher than 26 °C and rainfall ranging from 1,000 to 4,000 mm/year (INEGI 1991).

Estimation of aquatic plant coverage
Water body surveys were conducted from July 2015 to August 2019.For each survey, we examined the perimeter of lakes and reservoirs to visually estimate plant coverage.Although reeds are not strictly aquatic, they were included in this study as they threaten water availability.Coverage maps were constructed from field observations and coverage estimation, complemented by boat surveys.Additionally, we used the Google Earth Pro tool to analyze satellite images from Google Earth.We calculated the geometric surface area of the water body, estimated the degree of plant coverage, and compared the result with field observations.
To estimate the approximate area covered by plants, the number of floating and submerged plants was counted in randomly sampled plots of 1 m 2 .In the case of the reeds, the number of stems was counted (Ramos et al. 2004).Depending on the degree of coverage observed, between 4 and 6 plots were surveyed per water body.

Species identification and classification
We identified all species via photographs taken of all plant species in each water body, and all specimens that could not be identified in situ were collected and cultivated to a stage that facilitated identification in the laboratory.Algae were only identified at the genus level since no reproductive stages were observed.Classification into the native or nonnative species criteria was based on: 1) USDA Natural Resources Conservation Service (2020); 2) Enciclovida ( 2020), Invasive Species Compendium (2018); 3) Global Invasive Species Database (2020), and 4) The University of Florida Center for Aquatic and Invasive Plants (2020).All scientific names were verified with the International Plant Names Index (2020).

Results
In total, 75 water bodies were surveyed across the 11 hydrological regions of México and included 56 reservoirs, 14 lakes, two irrigation channels, two irrigation drainages and 1 river (Supplementary material Tables S1, S2).Large areas covered by aquatic plants were observed in 51 of the water bodies.Only 22 reservoirs and one irrigation channel were free of plant species (Table S2).
Forty-two species were recorded, of these, 11 were observed in 4 or more water bodies (Figure 1).Thirty-one species were classified as native (Enciclovida 2020; Table 1).According to the observations during surveys, and the comments of the local people, four of the native species had not been reported invading the surface of the reservoirs: Cyperus eragrostis, Cyperus strigosus, Ludwigia peploides and Verbena bonariensis.Eleven species were classified as non-native and invasive (Lowe et al. 2000; Global Invasive Species Database 2020; Table 2) and were found in seven of the 11 hydrological regions.Percent of plant coverage estimated ranged from 5% to 100% of the surface of water bodies (Table S1).

Discussion
Fifty-two of the 75 water bodies surveyed in this study contained aquatic plants (Table S1), while no aquatic plants were recorded from the remaining 23 sites (Table S2).Of the 52 sites with aquatic plants, 9 contained only native species, 18 contained non-native and, the remaining 25 contained both native and non-native.Water surface coverage by aquatic plants was greater than 30% all 52 water bodies, and 22 exceeded 65% coverage.
Ten of the 42 species of aquatic plants recorded are listed among 100 of the world's worst invasive alien species (Lowe et al. 2000).Twenty-seven of the 31 native species were observed to be colonizing new habitats and show coverages from 20 to 100% of the water surface of lakes or reservoirs (Table S1).Barreto et al. (2000) observed that the most "aggressive" aquatic weeds are native to the Neotropics, as are T. latifolia and T. dominguensis, which were observed spreading to new habitats, reducing the water surface of dams or lakes.Unfortunately, few control measures have been carried out against these invasive native species in México, even though they clog and reduce water flow in ditches and irrigation channels of Sinaloa and Sonora states (Table S1).Eutrophication via the influx of nitrates and phosphates from urban, agricultural, and industrial wastewater is the primary driver of these native species becoming invasive.These species have been reported as a serious threat to water availability and biodiversity (Gutierrez et al. 2007;IMTA 2008;Wang et al. 2016), and at least one, Najas marina, negatively impacts fishing activity in Mitla, a brackish coastal lagoon in Guerrero state (Table S1).Despite being a known problem in reservoir systems (Agami et al. 1980), this species has yet to be considered as such in México.
In the largest and best-known lakes in México, including Chapala, Patzcuaro, Cuitzeo and Zirahuen, non-native or native species such as A. donax, P. australis, Typha spp.and S. acutus were recorded (Table S1).Typical control of these species is carried out by cutting and/or burning the stems, leaving the rhizomes (i.e.reproductive structure) intact.In 2008, about 1,700 ha of Patzcuaro Lake were invaded by these species, and because it was shrinking the flooded area, the local government implemented control measures that included mechanical removal (Huerto Delgadillo et al. 2012).
Species belonging to the genus Typha, native to México, are a serious aquatic invasive in many countries (IMTA 2008;Lorenzi 1991;Bansal et al. 2019), and the large amount of biomass and organic matter produced by these species negatively impact ecosystem functions and diversity (Barko and Smart 1981).Angeloni et al. (2006) observed decreases in plant diversity, higher concentrations of nutrients, including nitrates and phosphates, and high sediment accumulation in areas invaded by Typha spp.The large accumulation of organic matter from the decomposition of the plants creates large deposits in riparian environments, which people then use for cultivation or new settlements.As this happens, fishermen shift their occupation from fishing to farming, leading to an increase in pollution from fertilizers and pesticides.The degradation of these habitats is further exacerbated by the increased numbers of people in villages near dams or lakes.Most of these villages lack centralized sewage systems and instead rely on septic tanks, which leak nutrients and pollutants into the surface and groundwater (Torres 1993;Zurita et al. 2012).As such, the proliferation and dispersion of invasive plants dramatically reduces the flooded area of the lakes, further degrading these habitats (Mendoza et al. 2002;Von Bertrab 2003).Typha spp.are not the only invasive species impacting these systems, M. aquaticum and C. demersum were also observed invading the water surface of lakes and dams and are known to be serious aquatic invasive species (Teles and Silva 1974; Global Invasive Species Database 2020).Unfortunately, there is no in México on the environmental impact of these plants.
The most frequently observed species was E. crassipes, found in almost 43% of the sites (Figure 1) and is the only one the 11 non-native species with some control program in Mexico (Gutierrez et al. 2007).Eichhornia crassipes is listed among the 100 worst invasive alien species (Lowe et al. 2000).The main obstacles in controlling E. crassipes are the rapid growth rate and the ability to re-infest via the seed bank or by flood-borne plants (Coetzee et al. 2009).Herbicides and mechanical removal have been used to control this weed, but these are expensive and require repeated treatments indefinitely.As an effort to control E. crassipes, insect and plant pathogens are being used (under specific regulations) as biocontrol agents in some water bodies in México (Martínez Jiménez and Gomez Balandra 2007;Martínez Jiménez et al. 2018).
Arundo donax, a non-native species, is another aquatic invasive plant known for its severe impacts on riparian zones of rivers and dams (Frandsen and Jackson 1994;Goolsby 2008).It was initially introduced for erosion control at the edges of the Rio Grande, but has since spread and become the primary non-native invasive plant negatively impacting water availability in semi-arid regions along the US and Mexican border (Figure 1).According to Chenghai et al. (2011), in northern hydrological regions, where water availability is limited and little rainfall, species such A. donax covers 5,981 ha along the Rio Grande, 3,714 ha on the USA side and 2,267 ha on the Mexican side (Table S1).Large stands of A. donax can significantly increase water loss from underground aquifers due to a high evapotranspiration rate (Jain et al. 2015), and several reports indicate that Arundo donax invasions are correlated with increased natural fire regimes along the riparian corridors (D'Antonio and Vitousek 1992;Scott 1994;Bell 1997;Bond and Keeley 2005).To establish a program for controlling this weed, release insects as biological agents of A. donax was made through a bi-national collaboration USA-México on both sides of the Rio Grande.Insects are now widely established and beginning to show impacts on plants at the initial sites of release, without damage in non-target plants (Goolsby et al. 2016(Goolsby et al. , 2020;;Martínez Jiménez et al. 2017).Other species in which biocontrol agents have been evaluated include Azolla filiculoides (Hill 1998), E. densa, M. aquaticum, P. stratiotes (Cilliers 1991), P. repens (Barreto et al. 2000).
In reservoirs where no coverage by aquatic plants was observed (Table S2), it is important to note that some of these sites have limited anthropogenic influence.Most are deep canyons, located far from urban and/or agricultural areas and do not receive nutrients from agricultural, urban and/or industrial wastewater.Similarly, these water bodies are not recreational areas or visited by tourists, making them unlikely to have invasive species transported to them via human activity.In Requena and Endhó dams of the state of Hidalgo (Table S2), no aquatic plants were observed because the local Ministry of the Environment maintains an active mechanical control program to prevent.Although aquatic weeds were not found in the Sonora State dams (Table S2), we observed in the irrigation channels of the Irrigation District 038, ten species of which three are non-native (reported as invasive by the Global Invasive Species Database (2020)) and six natives covering 95% of irrigation canals (TS1).
The Sinaloa dams presented the same pattern, where all bypass dams, and irrigation channels contained large populations of Eichhornia crassipes, Lemna minor, Pistia stratiotes, Paspalum repens, and Typha latifolia (Table S1).In reservoirs of the state of Chihuahua, Tamarix ramosisima and Arundo donax were observed invading irrigation channels (Table S1).It is important to note that in these regions, producers practice extensive agriculture, and the use of restricted fertilizers is releasing nutrients that cause explosive growth and reproduction of aquatic plants.
The species reported in this survey, whether native or non-native, share characteristics associated with invasiveness, including rapid growth rates, flexible reproductive strategies (asexual via stolons, rhizomes, or fragments, and sexual via seeds), high seed production, high dispersal capabilities, phenotypic plasticity, and tolerance to a wide variety of environmental conditions (Hofstra et al, 2020;Rejmanek and Richardson 1996).These traits contribute to rapid colonizing of new habitats and continued spread in the absence of active control measures.As they spread, free-floating plants form dense mats and create large amounts of sediments facilitating the establishment of other native or non-native species (such as reeds).In turn, water availability in invaded reservoirs and lakes is reduced through the combination of evapotranspiration and sedimentation.Additionally, increased eutrophication from pollutants from cities and agricultural areas further facilitates establishment and negatively impacts water navigation, recreational activities, and fishing.Dukes and Mooney (1999) indicated altering nutrient cycles and other factors, such as global warming or hydrological changes, have the potential to accelerate biological invasions.The high degree of aquatic plant coverage in areas surveyed during this study is the result of anthropogenic inputs, primarily eutrophication, and represents a serious threat to aquatic resources in México.
Most water bodies surveyed in this study were impacted by native and/or non-native invasive plants, and few resources are currently allocated to control these plants.As such, the water supply problems continue to compound with each passing year.Despite the serious impact of these plants on biological diversity and human activities, few control programs have been implemented in the country, and almost none have long-term support.Exacerbating the control issue is the fact that many of the recorded plants are sold as ornamental plants for ponds or fountains as well as for the breeding of ornamental fish.
In various meetings held with personnel of the National Water Commission and the municipalities, we noted lack of information on the problems posed by non-native or native invasive plants, as well as the absence of federal or state programs to address this situation.Control programs to control these species and eliminate the causes of their spread are critical for preserving the non-renewable water resources.
This study is the most complete national survey on non-native and native aquatic plants in México and provides useful information for stakeholders regarding this serious threat to the national aquatic resources.This baseline data will be used to develop "A National Register of Introduced and Invasive Species" database, which will inform the control of invasive species and help in developing plans to mitigate their impacts.

Figure 1 .
Figure 1.Eleven hydrological regions surveyed in this study from the 37 designated by The Mexican Water Authority (Comisión Nacional del Agua, CONAGUA) showing native (N), non-native (NN) and Invasive (I) plants found in 4 or more of the water bodies in each of these regions.

Table 1 .
Native aquatic plants recorded from the 11 hydrological regions surveyed in México.

Table 2 .
Non-native aquatic plants recorded in 11 hydrological regions surveyed in México.