1. Introduction
Most
Eucalyptus plantings in Florida before 1970 were windbreaks, ornamentals, and shade trees in central and southern Florida [
1]. Reevaluation of a number of species starting in the 1970s [
2,
3] included tree improvement efforts with
E. amplifolia and
C. torelliana (formerly
E. torelliana), which had been considered of limited potential and still are minimally deployed. For both, the starting germplasm was derived from seed or trees resulting from earlier screening efforts, which resulted in the Florida Division of Forestry retaining and growing small quantities of
E. amplifolia and
C. torelliana for ornamental purposes.
Matching
E. amplifolia and
C. torelliana to the lower southeastern USA’s diverse weather and soils is challenging. Climatic regions based on average low temperatures or numbers of freezes provide some broad guidelines, but freeze aberrations [
4] and extended cold periods impact their freeze susceptibility at young ages. Unpredictable, extended dry spells make Florida’s summer rainfall climate difficult for successful planting and early growth. Within climatic regions, soils available for planting
E. amplifolia and
C. torelliana can range from sandy and infertile to heavy clay to limestone to organic.
Broad climatic regions guide the deployment of
E. amplifolia and
C. torelliana in Florida (
Figure 1). Typically tolerant of the colder winters of northern Florida and similar regions,
E. amplifolia requires good fertility with pH
> 5.6, unless infertile, poorly drained soils are amended. From southern into central Florida,
C. torelliana tolerates typical winter conditions and grows well across sites, especially when irrigated on deep sands.
This paper expands previous reviews [
6,
7] by compiling tree improvement and utilization research with
E. amplifolia and
C. torelliana through 2021. It highlights genetic resources, describes current and future uses, identifies ongoing research activities and needs, and acknowledges significant collaborators in this research.
2. Genetic Resources
The tree improvement strategy followed for
E. grandis in Florida was used in developing SSOs of these two species [
1,
6,
7,
8,
9]. This inexpensive, effective strategy utilized short generation time of about four years and rapid growth to concurrently test provenances, progenies, and new, primarily single-tree accessions in one place, followed by early selection and use of pedigrees to minimize inbreeding and achieve rapid genetic gains.
Two generations of
E. amplifolia orchards (
Table 1) produce seed for freeze-frequent northern Florida and similar areas. The 1st generation genetic base population included many new accessions, particularly individual tree accessions from frost frequent portions of the natural distribution of
E. amplifolia subsp.
amplifolia. Most of those accessions were retained in AO92. Two 2nd generation SSOs (AO03 and AO10) included seedlings from AO92, but were mostly composed of additional new accessions. CSO AO13 was established beginning in 2013. AO03, AO10, and AO13 can supply commercial seed for central Florida northward into southern Georgia, Alabama, Mississippi, Louisiana, and Texas based on BVs calculated from progeny tests established since 1998.
The 2008
C. torelliana genetic base population (
Table 1) included 960 trees from 29 trees in windbreaks in Florida. This tropical species has demonstrated tolerance to freezing temperatures, and all 69 trees in TO08 combine freeze tolerance with good growth and tree form. Twenty-five new Australian accessions were included in the 2nd generation base population that became TO12. These SSOs can supply seed for southern and central Florida.
Severe freezes made selection of fast growing, freeze resilient cloning candidates possible (
Table 2). Collectively, 115
E. amplifolia cloning candidates were identified, with 35 entered in tests. Since
E. amplifolia rooting percent is highly variable but typically half that is of
E. grandis [
10], many more candidates are needed for commercialization. Four
C. torelliana cloning candidates have been identified and captured by tissue culture but have not been field tested.
A novel method has been used to propagate E. amplifolia commercial cultivar A1 because it does not respond to rooting hormones and its tip cuttings fail to root. Several weeks after cutting a clone down in March–April at a height of ~0.75 m, when stump shoots are ~1 m long, 3-node cuttings are cut from the firm sections of the shoots. The cuttings are then completely submerged in a solution of seaweed extract (dry concentrate from Arcadian Seaplants at 5000 ppm in water) for 1–2 h, followed by double wounding of the basal segments to promote more robust root development, then insertion into a rooting mix of 45% perlite and 55% pine bark amended with controlled release fertilizer, cutting the leaves on the upper two nodes in half to reduce water loss, and, after rooting under mist, being moved to 50% shade and fertilized weekly with water soluble 20–20–20 at half rate.
Cultivar A1 has very good form with a wider crown than most
E. amplifolia, making it well suited for ornamental use (
Figure 2). Its relatively long retention of juvenile foliage contributes to a very rapid growth rate. It has excellent cold-hardiness, withstanding temperatures to −9 °C when established, and also is blue gum chalcid (
Leptocybe invasa) resistant. Seedlings grown from A1 seed have also been chalcid resistant and have exceptional form.
BVs have been developed for five
E. amplifolia and five
C. torelliana traits that are important for best deployment of genotypes (
Table 3). The numbers of progeny tests contributing to these BVs were 15 in Florida and the southeastern USA for
E. amplifolia and seven in Florida for
C. torelliana. All test data were first processed in SAS version 9.4 (Copyright© 2002–2012 by SAS Institute Inc., Cary, NC, USA). For each continuous trait, an analysis was run by test to determine if there were any heritability for the trait. For those tests with heritability, the data were standardized by the square root of the progeny variance, making the standardized progeny variance equal 1. Data were then assembled and analyzed in ASREML ver. 4.1, 2014 [
11]. For binary traits, the data were run in ASREML using a logit link function. The results from the logit analysis were then back transformed into percentages. For
E. amplifolia, the gains for continuous traits were calculated against the overall mean, and the binary trait gains were calculated as though the incidence were 50%. For
C. torelliana, the gains for all traits were calculated against the overall least square mean performance of the TO08 orchard trees. The general formula for gain calculations in percentage was 100 ∗ ((prediction + population mean) − checklot mean)/checklot mean.
Collectively, these BVs guide the selection of genotypes for individual applications. In Tree Basal Area and Plot Basal Area, which incorporate survival, BVs indicate individual tree size and per ha productivity, respectively. Regarding Stem Quality, on a 1 = good to 5 = poor scale, and Freeze Resilience, on a similar scale, BVs rate tree form and freeze tolerance and/or ability to regrow vigorously afterwards, respectively. As for Chalcid Resistance and Pest Resistance, both on 0 = no incidence or 1 = infected scale, BVs reflect relative resistance to critical pests. C. torelliana‘s Flowering BV assesses early flowering.
3. Potential Uses
Market opportunities in Florida for
E. amplifolia and
C. torelliana wood are now limited but could expand. Near existing pulp mills, specialty pulps may utilize eucalypts, but the hardwood pulpwood market forecast in the 1970s [
1] has become a more local mulchwood market [
5]. As cypress mulchwood availability decreases, more eucalypt wood may replace it.
Many wood markets for eucalypts, such as MDF, are undeveloped in Florida. In preliminary testing of their suitability for MDF, genetic variation between and particularly within
E. amplifolia and
C. torelliana affected their MDF potential [
12]. These two studies of factors (wood characteristics, refining system, and resin system) influencing their suitability found that log specific gravity (SG), fines, MDF SG, and fiber length were most influential. A study using 4% phenol-formaldehyde resin detected some variation between species, considerable variation within species, minor variation within a tree, and some influence of basic wood characteristics. The second study involving the “best” genotype of each species and three resins concluded that certain genotypes may be suitable and that resin type and rate and percentage of fines influenced MDF properties. Refining and MDF-making aspects have such major impacts on MDF properties that specific processing requirements may be needed to optimize MDF production from
E. amplifolia and
C. torelliana genotypes.
E. amplifolia may also be suitable for wood-cement boards, plywood, and oriented strand board [
13,
14,
15].
Eucalyptus energywood uses have been demonstrated [
16].
Eucalyptus has been considered as a feedstock for energy generation at pulp mills in Florida, and it has potential for use in biorefineries associated with pulp mills [
17].
Eucalyptus has been proposed as the primary feedstock for stand-alone biomass power plants in Florida [
18].
Eucalyptus amplifolia and
C. torelliana have shown promise as biochar feedstocks and for sequestering carbon. The biochar of both species compares favorably with the commercial Polchar made from oak trees in Europe (
Table 4). Biochar production in southern Florida may foster planting of the more dense
C. torelliana (
Table 5).
E. amplifolia is suitable for biofuel and bioenergy production. Its major wood chemicals in are lignin, glucan, and xylan [
20]. For ethanol and methanol production, certain wood properties are typically favored: higher wood density, lower moisture content, and higher extractives content [
21]. There were differences between and within
E. amplifolia and
C. torelliana for some of these properties based on limited genotypes and ages (
Table 5), which suggests that clonal deployment would be advantageous in producing energy products. Similar variation in refined fiber characteristics also emphasized the importance of genetic variation in making other products.
Cost-effective capture of the many silvichemicals in Eucalyptus species is critical to their commercial use. Steam pretreatment of wood chips of
E. amplifolia and
C. torelliana yielded a multitude of components in condensate extracts but no compound was in sufficient quantity to make separation and recovery commercially viable [
20]. Capturing them as incidental byproducts of other wood processing may be an option.
SRWC systems maximize productivity for the above uses [
11,
22,
23,
24,
25,
26,
27,
28,
29,
30,
31]. Due to propagation ease, rapid growth, tolerance to high stand density, response to intensive culture, and coppicing,
E. amplifolia can produce up to 67 green mt ha
−1yr
−1 in multiple SRWC rotations as short as three years. It is very responsive to soil amendments, vegetation control, and irrigation.
Eucalyptus amplifolia and
C. torelliana also have other uses.
E. amplifolia Cultivar A1 has been planted in multiple southern Georgia (USDA Zone 8b) locations, primarily for ornamental purposes in yards and at a local golf course (
Figure 1).
E. amplifolia is also being considered for windbreaks around citrus orchards to reduce the spread of disease and insects and to provide cold protection in the winter. Since
E. amplifolia loses its lower limbs when mature, double row windbreaks using another low growing evergreen species to block low level wind provide the best protection. Another potential use is to interplant
E. amplifolia in citrus orchards, especially those involving satsuma mandarins, to provide up to 30% dappled shade that reduces sunburn injury on fruit, to perhaps confuse insects that feed on citrus trees, and also to provide some frost protection in more cold frequent areas like southern Georgia.
One of the
E. amplifolia uses in Australia is for honey production [
32] because it flowers prolifically in late winter and early spring when other flowers are not available.
C. torelliana produces abundant flowers four times a year. With the significant reduction in the number of citrus trees in Florida due to citrus greening, beekeepers are seeking other plants to support their honey bees. Other uses include tall visual barriers and sound barriers along highways.
While both
Eucalyptus amplifolia and
C. torelliana can be used in windbreaks [
11,
33,
34,
35,
36,
37,
38],
C. torelliana has been more widely planted around citrus groves and vegetable fields in central and southern Florida (
Figure 3). Since 2010, PRCN has grown potted
C. torelliana seedlings from TO08 for agricultural, residential, and commercial windbreaks and privacy screening. TO08 seed are dried, sieved, purified, weighed, cataloged, packed in coin envelopes, and then stored in airtight containers at 4.4–5.5 °C with silica gel desiccant. Seed is taken from cold storage and germinated in 5 × 10 cm Ellepots in 45 cell trays in a greenhouse. Germination/selection rate for stored seeds averages about 72%. About 15-cm-tall seedlings are transplanted into 15 cm nursery pots to grow to 45–60 cm height outdoors.
Two unpublished studies in Florida illustrate the performance of
C. torelliana in windbreaks. In June 2012, two-row windbreaks consisting of
E. grandis cultivars and
C. torelliana progenies in adjacent staggered rows 2.4 m apart were established around two Rapid Infiltration Basins (RIB 2-3 and RIB 3-2) near Winter Garden. The trees were subsequently irrigated with reclaimed water and periodically measured. Carbon sequestration was estimated by applying equations for
E. grandis in Brazil [
39] and
C. torelliana in Florida [
40]. The cultivars were bigger in RIB 2-3 while the progenies were similar across the RIBs (
Table 6). Consequently, at 16 months, the cultivars had higher sequestration in RIB 2-3, and sequestration by the progenies was basically the same in both RIBs, but not as high as the cultivars in RIB 2-3. Comparison of
C. torelliana carbon sequestration with that of
E. grandis in this windbreak study suggests that the typically higher wood density of
C. torelliana (
Table 5) may offset its somewhat slower growth compared to
E. grandis.
The second windbreak study involved two windbreaks established in late March 2014 around a citrus grove near Clermont, Florida, following Roundup herbicide in mid-March. In the northern windbreak, seedlings of C. torelliana progeny 1–14 were planted in a single row at 1.5 m spacing, and in the double row southern windbreak at 2.4 m spacing, E. grandis cultivar G3 was established in the interior (north) row and C. torelliana progeny 1–14 in the staggered (1.2 m offset) exterior (south) row. The trees were subsequently irrigated for four years and measured through May 2020. At age 74 months, C. torelliana in a single row configuration was taller (12.6 vs. 10.8 m) but had similar DBH (17.9 vs. 17.5 cm) compared to its size in the 2-row configuration. In the double row southern windbreak, 1–14 seedlings in the exterior row were smaller than G3 in the interior row (10.8 vs. 24.1 m in height, 17.5 vs. 24.2 cm in DBH), a relationship that makes this two species combination ideal for quickly reaching maximum windbreak height while maintaining full canopy closure from the ground up.
Eucalyptus amplifolia and
C. torelliana may also be used for dendroremediation and weed suppression. For dendroremediation,
E. amplifolia can be very effective [
41,
42,
43,
44,
45,
46,
47,
48,
49,
50].
Eucalyptus grandis and
E. amplifolia progenies and
Populus deltoides clones were evaluated for energywood production and dendroremediation of accumulated soil nitrogen (N) and phosphorus (P) in 27-month stand rotations at the Tallahassee, Florida, municipal waste spray field [
50].
Eucalyptus grandis grows fastest and thus has the greatest potential nutrient capture, but
E. amplifolia is more cold-hardy and appropriate in freeze-frequent areas.
Eucalyptus is evergreen, an advantage for year-round nutrient and contaminant capture.
Eucalyptus also regenerates effectively from stump sprouts (coppicing) following harvest, facilitating reproduction in short rotations optimizing both nutrient removal and energywood production. The highest biomass producing
P. deltoides clone (110412) removed 215% of N and 615% of P effluent water inputs, indicating effectiveness in reducing soil nutrient concentrations and mitigating potential groundwater pollution.
Eucalyptus amplifolia and
E. grandis survived and grew very poorly as the result of below-normal winter freezing temperatures during the study period and were not evaluated for nutrient removal. However, this discriminating event provided an opportunity to assess potential
E. amplifolia progeny recovery. Tree height growth indicated greater recovery for progeny 5050 and 5108 than others tested. Eucalypts may be “bridge crops” to convert lands infested with invasive species to agricultural uses [
51]. Planting trees at high density results in canopy closure in a few months, shading out understory plants.
4. Research Needs
Genetic and silvicultural improvements with E. amplifolia and C. torelliana have dramatically improved their productivity, but still more progress may be made through research in freeze resilience, growth, coppicing, pest resistance, and propagation. While advanced generation breeding could produce more advanced SSOs and CSOs to improve these traits, significant improvements may accrue from clonal selection and testing.
In collaboration with Camcore, North Carolina State University, over 1.8 million seed were collected from the AO03, AO10, and AO13 seed orchards in November 2019. A set of 38 family seedlots will be used to establish progeny tests by 15–20 Camcore members in 11 countries (Argentina, Brazil, Chile, Colombia, Ghana, Sierra Leone, Indonesia, Kenya, Mexico, South Africa, Uruguay). In March 2020, pollen was collected from 13 E. amplifolia clones for a collaborative eucalyptus hybridization project that will attempt to produce more than 20 different hybrid combinations in total: E. globulus x E. amplifolia and E. grandis x E. amplifolia, as well as a three-way hybrid, (E. grandis x E.nitens) x E. amplifolia.
For proven clones, economical and rapid propagation is necessary. Current vegetative propagules are more expensive than seedlings. Florida’s seasonal planting schedule further necessitates the need for periodic rapid buildups of propagules.
Silvicultural enhancements are needed because of the infertility, low pH, and low organic matter of many available planting sites. Appropriate environmentally friendly amendments such as organic fertilizers and biochar need study and documentation. Application of available wastewaters to plantations needs to be commercialized. Growth and yield models reflecting genetic and silvicultural improvements are needed.
Expanded markets for E. amplifolia and C. torelliana in Florida may depend on energy project development and technology improvement. The current market for mulchwood is met by existing plantations, but the mulchwood market could expand as cypress availability decreases. Pellet plants, biomass-fueled utility plants, and especially biochar production facilities could significantly increase demand. Improvements in biomass conversion at pulpmills and stand-alone biorefineries would also increase demand.