Volumes and Efficiencies of Water-Use Within Selected Indigenous and Introduced Tree Species in South Africa: Current Results and Potential Applications

VOLUMES AND EFFICIENCIES OF WATER-USE WITHIN SELECTED INDIGENOUS AND INTRODUCED TREE SPECIES IN SOUTH AFRICA: CURRENT RESULTS AND POTENTIAL APPLICATIONS

M.B. Gush¹, P.J. Dye², C.J. Geldenhuys³ and H.H. Bulcock 4

¹CSIR, Stellenbosch

²University of the Witwatersrand, Johannesburg

³FORESTWOOD cc, Pretoria

4University of KwaZulu-Natal, Pietermaritzburg

Corresponding author: [email protected]

Abstract

South African indigenous forests provide goods and services which are recognised as valuable natural capital, and are well documented. However, the limited extent of these forests has forced South Africa to accelerate the expansion of its own plantation forest industry over the last century, using fast-growing introduced tree species to meet the timber needs of the country. The resultant impacts on streamflow and water resources have been the subject of considerable research, and first led to regulation of this industry in 1972 due to water-use concerns. Conversely, there is widespread belief that indigenous tree species use little water and deserve to be planted more widely. However, research and data on the water-use of indigenous trees and forests has historically been limited. This paper discusses current progress in a Water Research Commission solicited project on the measurement and modelling of water-use and growth in selected South African indigenous tree species. Hourly sap flow rates (water use) over a 12-month period were recorded in a diverse selection of indigenous tree species, while stem circumferences were recorded at the start and end of the monitoring period, to derive biomass increments. Rates of growth and water-use were used to calculate water-use efficiency, defined as mass of utilisable wood produced per unit of water transpired, and were compared to existing data for introduced plantation species. Water-use efficiency in the indigenous species studied was lower than for introduced plantation species, however overall water-use was also generally lower in the indigenous species. It was concluded that the relatively lower water-use efficiency of the indigenous species studied was primarily a consequence of slow growth rates as opposed to high water-use rates. Implications and potential applications of these findings in alternative forms of indigenous forestry and sustainable resource use are discussed.

1. Introduction

South Africa is very reliant on its plantations of introduced tree species to meet its pulp and timber needs, and the benefits of this industry in terms of production, income generation and job provision are undisputed (Chamberlain et al. , 2005). The downside is that these benefits come at some environmental cost, not least the impact of the industry on water resources (Dye & Versfeld, 2007). Many catchment areas are consequently now closed to further afforestation, but economic growth and development continue unabated. Imports and improved productivity are potential solutions to continue meeting the demand for timber and forest products, but further consideration of the feasibility of expanding indigenous tree resources is also warranted. With over 1000 species of indigenous trees in the country, South Africa is extremely rich in natural arboreal diversity (von Breitenbach, 1990). The numerous benefits of indigenous trees and forests, in terms of the goods and services that they offer, are widely recognised (Lawes et al. , 2004; Shackleton et al. , 2007). There are also widespread perceptions that indigenous tree species use less water than introduced tree plantations, however up to now these have been unsubstantiated. While data from previous studies are available on the water-use efficiency (WUE) of common introduced plantation species in South Africa (Olbrich et al. , 1996; Dye et al ., 2001), information on the water-use of indigenous trees and forests is scarce 1

With a growing awareness of the socio-economic and environmental challenges being posed by finite water supplies in a developing country such as South Africa, there is renewed interest in the possibility of low water-use forms of forestry. This information is required in order to facilitate sustainable land-use planning from a hydrological perspective (Dye et al. , 2008). New and innovative techniques to quantify the water-use (transpiration and evapotranspiration) of a range of tree species and forest types are available (Jarmain et al ., 2008), and these may be used to broaden our understanding of forest hydrological processes, and their associated effects on water resources in this country. Evidence of low and efficient water-use in indigenous tree species would make them an attractive alternative forestry solution, particularly in catchments that are water-stressed. The overall efficiency of water-use for biomass production, and the net benefit of the water used are also important criteria that need to be understood to permit the evaluation of different land use scenarios. Expanded indigenous forestry systems in South Africa could, under certain conditions, offer attractive alternative land-use scenarios to plantations of introduced timber species, but the feasibility of their expansion needs to be thoroughly evaluated from socio-economic and environmental perspectives. A research project, solicited, managed and funded by the Water Research Commission (WRC) with co-funding by the Working for Water Programme of the Department of Water Affairs and Forestry, was commissioned to study the water-use, growth rates and economic value of the biomass of indigenous trees and forests in South Africa. Research questions posed by the study include:

1. Do indigenous tree species use less water than introduced plantation tree species? 2. Do indigenous tree species use water more efficiently than introduced plantation tree species? 3. Is there scope for the expansion of indigenous tree systems in South Africa? This is a 6-year project (2009 – 2015), currently nearing the end of its second year (Water Research Commission, 2010). This research project follows on from an earlier pioneering WRC study, which explored water-use and growth rates in various indigenous tree systems in South Africa (Dye et al ., 2008). This paper reports on some of the recent findings of the project in terms of comparative growth and water-use studies between indigenous and introduced tree species, and discusses potential applications of the research.

2. Materials and methods

The objective of the measurement methodology was to determine the water-use efficiency of selected indigenous and introduced tree species by means of accurate measurements of water-use and growth. To do that, measurements were conducted for one year to incorporate seasonal variations in, and responses to, climate. In essence, the sampling strategy employed was to conduct continuous sap flow (transpiration / water-use) monitoring on an hourly basis, together with annual stem biomass increments at the start and end of the monitoring period. Hourly measurements of a full suite of climatic variables (solar radiation, temperature, relative humidity, wind speed and rainfall), together with soil water content measurements in the A-horizon complemented these.

2.1 Site and species

Two monitoring sites were selected on the Mondi De Magtenburg forestry estate in the Karkloof region of the KwaZulu-Natal (KZN) midlands in South Africa (S29° 21' 25.2''; E30° 11' 49.3'', alt. 1 148 m.a.m.s.l). The site has a mean annual precipitation of approximately 1100 mm (Lynch & Schulze, 2006), and is an established commercial forestry estate, consisting primarily of pine ( Pinus patula ) and wattle ( Acacia mearnsii ) stands in former grassland. The stands selected for this study consisted of:

1. An indigenous Podocarpus henkelii (Stapf ex Dallim.) & Jacks (Yellowwood) plantation. 2. An introduced Pinus patula (Schiede ex Schlechtendal) & Chamisso (Mexican weeping pine) plantation.

The small (<1ha) P. henkelii stand is situated in a riparian area close to a small stream on the estate, and is one of very few formally established indigenous tree plantations in South Africa. Due to changes in ownership of this particular farm details on the stand are limited, and the trees are of an unknown age. However, by virtue of their size (average tree height of 8 m) and stem diameters (DBH: 15-30 cm) the trees 2

Gush, M.B., Dye, P.J., Geldenhuys, C.J. and Bulcock, H.H., 2011. Volumes and efficiencies of water-use within selected indigenous and introduced tree species in South Africa: Current results and potential applications. In : Proceedings of the 5th Natural Forests and Woodlands Symposium , Richards Bay, 11-14 April. are estimated to be roughly 40 years old. The stand is relatively weed free, having achieved canopy closure, but a few invasive A. mearnsii and Solanum mauritianum (Bugweed) plants are present in the stand. The trees have not actively been pruned and planting espacement is somewhat irregular but an average distance between trees (3 m X 3 m) translates into a planting density of approximately 1111 trees per hectare. Sap flow (water-use) and stem increment (growth) rates were measured in three trees at this site (Table 1). The study trees were selected after first conducting a survey of the range of stem sizes present in the plantation. All trees were subsequently assigned to one of 3 stem circumference size classes, with each size class being represented by one sample tree on which sap flow measurements were conducted. The annual transpiration totals for the 3 study trees were weighted according to the number of trees in the size class that they represented, and a single transpiration rate for the plantation was determined, accounting for variations in stem circumference and water-use within the Yellowwood plantation. The second site is situated within 50 m of the Yellowwood stand and consists of a Pinus patula stand with trees of comparable size to the Yellowwood trees (8-10 m high, DBH: 18-25 cm). Tree spacing within this non-riparian stand is also somewhat irregular but an assumed planting distance of 3.5 m X 3.5 m (a common espacement for production of saw timber) translated into a planting density of approximately 816 trees per hectare. Sap flow (water-use) and stem increment (growth) rates were measured in two trees at this site (Table 1). Bark thickness of the sample trees were determined by excising bark sections from the stems. Measurements of sapwood depth, required to determine the insertion depths of thermocouple probes for water-use measurements, were obtained using a 5 mm inside-diameter increment corer (Haglöf, Sweden). Cores were subsequently analysed for sapwood depth using measurements of the visual distinction between lighter coloured sapwood and darker coloured heartwood. Wood density for the two tree species was determined using mass and volume measurements (Archimedes Principle) on stem-wood samples chiselled from the trees. Monitoring at both sites commenced on 13 August 2009 and continued for an entire year. Simultaneous measurements of certain meteorological variables (rainfall, solar radiation, air temperature and relative humidity) and soil water content in the top 10 cm of the soil profile, took place hourly at the site for the corresponding period.

Table 1. Sample tree details

Diameter at Tree Sapwood Wood Wound Bark width Species breast height height depth density width (mm) (mm) -3 (cm) (m) (cm) (g.cm ) P.henkelii 1 14 6.34 5.5 3 7 0.468 P.henkelii 2 23 7.33 9.5 3 7 0.468 P.henkelii 3 18 7.00 7.5 3 7 0.468 P. patula 1 20 8.77 8.5 4 10 0.380 P. patula 2 24 10.79 10.0 4 10 0.380

2.2 Sap flow measurements

The Heat Pulse Velocity (HPV) technique is an internationally accepted method for the measurement of sap flow (water-use) in woody plants and has been extensively applied in South Africa (Dye & Olbrich, 1993; Dye, 1996; Dye, Soko & Poulter, 1996; Dye et al. , 1996; Gush, 2008; Gush & Dye, 2009). The heat ratio method (HRM) of the HPV technique (Burgess et al. , 2001) was selected for sap flow measurements in this study because of its ability to accurately measure low rates of sap flow, expected to be the case in indigenous tree species. The HRM requires a line-heater to be inserted in the xylem at the vertical midpoint (commonly 5 mm) between two temperature sensors (thermocouples). Heat pulses are used as a tracer, carried by the flow of sap up the stem. This allows the velocity of individual heat pulses to be determined by recording the ratio of the increase in temperature measured by the thermocouples (TCs), following the release of a pulse of heat by the line heater. For these measurements TC pairs and heater probes were positioned 80 cm up the main stem of each tree, below the first branches. TCs were inserted to four different depths within the sapwood to determine radial variations in sap flow. Insertion depths of the TC’s were calculated after first determining the total sapwood depth for each species, and then spacing the probes evenly throughout. All drilling was performed with a battery-operated drill using a drill guide strapped to the tree, to ensure that the holes were as close to parallel as possible. CR1000 data loggers connected to AM16/32 multiplexers (Campbell Scientific, Logan, UT) were programmed to initiate the heat pulses and 3

Heat pulse velocities derived using the HRM were corrected for sapwood wounding caused by the drilling procedure, using wound correction coefficients described by Swanson & Whitfield (1981). The corrected heat pulse velocities were then converted to sap flux densities according to the method described by Marshall (1958). Finally, the sap flux densities were converted to whole-tree total sap flow by calculating the sum of the products of sap flux density and cross-sectional area for individual tree stem annuli (determined by below-bark individual probe insertion depths and sapwood depth). Hourly sap flow values were recorded from all the trees. Periods of missing data were patched and the complete record was aggregated into daily, monthly and annual totals. Individual-tree sap-flow volumes (L.annum ¹) were scaled up to a hectare using the planting density to also derive sap flow (transpiration) totals in mm-equivalent volumes for the year.

2.3 Stem growth measurements and water-use efficiency

Stem biomass increment surveys were conducted on all the sample trees in conjunction with sap flow measurements. Initial biomass determination was carried out shortly after the individual trees had been instrumented with the HPV systems, and the final surveys were performed one year thereafter to incorporate 1-year seasonal variation in both water-use and growth. Biomass increments were calculated for all sample trees as the difference between the initial and final surveys. Stem circumferences at increasing heights up the tree were measured. These measurements were converted to volumes by assuming that the stem consisted of a series of truncated cones with a complete cone at the top. The volumes of individual cones V (m³) were calculated using (Eq. 1):

V =  ( πr2h) / 3. (Eq. 1.) where r is radius of the base of the cone (m), and h is height of the cone (m). The volumes of the truncated cones were calculated using (Eq. 2):

2 2 V =  ( πh (r 1 + r 1 r2 + r 2 )) / 3 (Eq. 2.) where r1 is radius of the base of the truncated cone (m), r2 is radius of the top of the truncated cone (m), and h is height of the truncated cone (m).

The individual stem section volumes were totalled for each tree, which allowed for the calculation of stem volume increase in the year. Stem biomass increments were converted from volume to mass using wood densities determined for each species (Table 1). In conjunction with the sap flow (water-use) results this allowed the calculation of WUE, defined as mass of woody biomass produced (g) per unit of water transpired (L), and results were compared against existing data for indigenous and introduced tree species available from previous studies (Olbrich et al ., 1996; Dye et al ., 2001; Gush & Dye, 2009).

3. Results

3.1 Weather

In general, the weather conditions at the site exhibited the seasonal pattern typical of this area (Table 2). The site experienced an extremely wet spring season initially. The start of the rainy season in this area is usually in October; however the rains commenced early, with significant amounts falling in August (51 mm) and September (75 mm) already. The rainfall season was also prolonged, with rainfall of 53 mm as late as April, and well distributed over the season. A high percentage of rain-days (irrespective of amounts) were recorded during the year (there were 197 rain days, or 54% of the days, with rainfall). No drought periods were experienced; however the total precipitation for the year was only 1025 mm, being slightly less than the long- term mean. The high percentage of rain days influenced the other weather variables, so while daily maximum temperatures regularly peaked above 30°C (with the highest recorded temperature being 41.1°C on 17 December 2009), monthly means of maximum temperature were mild initially, peaking in February and March. The frequent overcast conditions also limited the extent to which temperatures cooled at night, with the lowest recorded temperature being 0.8°C on 12 July 2010. Daily average solar radiation values were in February 2010, but dropped ¹־ day.²־ also very consistent initially, increasing gradually to 22.3 MJ.m

Wind speeds were .¹־ day.²־ dramatically in June and July 2010 to daily averages of approximately 7.7 MJ.m .in August which is normal ¹־ peaking at a daily average of 1.4m.s (¹־ generally low (±0.5m.s

Table 2. Monthly values of meteorological variables recorded at the site

Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Variable / Month ‘09 ‘09 ‘09 ’09 ’09 ‘10 ‘10 ‘10 ‘10 ‘10 ‘10 ‘10

Rainfall Totals (mm) 51.0 75.3 179.6 123.6 167.2 190.6 93.1 98.3 53.0 6.1 11.6 11.5

Ave. Daily Max T (°C) 19.6 21.6 21.2 22.7 26.3 27.2 31.0 29.6 27.7 27.1 22.6 21.6

Ave. Daily Min T (°C) 8.9 10.5 11.5 11.8 13.6 14.8 16.2 14.5 12.9 11.5 7.1 7.6

Ave. Daily Solar. Rad. 16.3 16.7 16.2 17.4 17.9 17.4 22.3 18.2 14.4 12.8 7.7 7.8 (¹־ day.²־ MJ.m)

Ave. Daily Wind 1.4 1.3 1.0 1.2 1.1 0.7 0.6 0.4 0.4 0.5 0.8 0.5 (¹־ Speed (m.s

3.2 Sap flow

3.2.1 Podocarpus henkelii

Sap flow (water-use) volumes recorded in the Yellowwood trees illustrate consistent transpiration rates year- round (Figure 1). This is attributable to the evergreen nature of this species and the lack of seasonal water stress due to the riparian location. A slight decline in sap flow rates is evident during the dry season, particularly around September when leaf exchange takes place. Thereafter, sap flow rates increase towards mid-summer, as leaf areas, temperatures and available water from rainfall increase. Towards the end of March sap flow rates gradually decline once more and the cycle is repeated. Individual days of wet, overcast weather in the summer (when available energy from solar radiation is limited) are characterised by very low sap flow rates. Daily volumes of water-use in these trees peaked at between 10 and 20 L.day ¹ during the summer, declining marginally to between 5 and 15 L.day ¹ in winter. Annual transpiration totals for the 3 study trees were weighted according to the number of trees in the size class that they represented, and a single transpiration rate for the plantation was determined, accounting for variations in stem circumference and water-use, within the Yellowwood plantation (Table 3). Individual tree water-use volumes (L.annum ¹) were also scaled up to weighted plantation equivalent depths of water-use (mm) using planting density in the stand.

Table 3. Weighting of observed transpiration volumes in 3 Podocarpus henkelii trees relative to stem circumference variation within the plantation, to determine a representative total water-use

Stem circ. No. of trees HPV HPV tree 1-yr Water - 1-yr Water- size classes in sub- tree stem use Weighting use (mm) (cm) sample no. circ. (cm) (L.tree ¹)

≤45 cm 10 1 44.17 1755 194.98 29.41%

45-65 cm 10 3 60.00 3554 394.85 29.41%

≥65 cm 14 2 73.48 5033 559.17 41.18%

Weighted Total for Plantation 3634 403.7 100.0%

120 0

110 10

100 20

90 30

80 40

70 50

60 60

50 70 Daily Rainfall (mm) 40 Transpiration -Podocarpus henkelii Tree1 (L/day) 80 Daily Rainfall (mm) Rainfall Daily

30 90 Daily Sap Flow (Transpiration) (l.day SapFlow (Transpiration) Daily ¹)

20 100

10 110

0 120 09 09 10 10 10 10 10 10 10 10 10 09 09 09 09 09 09 09 09 09 10 10 10 10 10 10 10 ------Jul Jul Jul Oct Oct Apr Apr Jan Jan Jun Jun - - - Feb Feb Mar Mar Aug Aug Sep Sep Nov Nov Dec Dec Dec Aug ------May May ------01 15 29 08 22 08 22 14 28 03 17 11 25 11 25 13 27 10 24 05 19 03 17 31 12 06 20 120 0

110 10

100 20

90 30

80 40

70 50

60 60 Daily Rainfall (mm)

50 Transpiration - Podocarpus henkelii Tree 2 (L/day) 70

40 80 Daily Rainfall (mm) Rainfall Daily

30 90 Daily Sap Flow (Transpiration) (l.day SapFlow(Transpiration) Daily ¹)

20 100

10 110

0 120 09 09 10 10 10 10 10 10 10 10 10 09 09 09 09 09 09 09 09 09 10 10 10 10 10 10 10 ------Jul Jul Jul Oct Oct Apr Apr Jan Jan Jun Jun - - - Feb Feb Mar Mar Aug Aug Sep Sep Nov Nov Dec Dec Dec Aug - - - - May May ------01 15 29 08 22 08 22 14 28 03 17 11 25 11 25 13 27 10 24 05 19 03 17 31 12 06 20 120 0

110 10

100 20

90 30

80 40

70 50

60 60 Daily Rainfall (mm)

50 Transpiration - Podocarpus henkelii Tree 3 (L/day) 70

40 80 Daily Rainfall Rainfall (mm) Daily

30 90 Daily Sap Flow (Transpiration) (l.day SapFlow(Transpiration) Daily ¹)

20 100

10 110

0 120 09 09 10 10 10 10 10 10 10 10 10 10 10 10 10 09 09 09 09 09 09 09 09 09 10 10 10 ------Jul Jul Jul Oct Oct Apr Apr - - - Jan Jan Jun Jun Feb Feb Mar Mar Aug Aug Sep Sep Nov Nov Dec Dec Dec Aug ------May May ------01 15 29 08 22 08 22 14 28 03 17 11 25 11 25 13 27 10 24 05 19 03 17 31 12 06 20 Figure 1. Daily sap flow (water-use) volumes (L.day¹) recorded in 3 indigenous Podocarpus henkelii trees

3.2.2 Pinus patula

Sap flow (water-use) volumes recorded in the P. patula trees illustrate a more distinct seasonal transpiration pattern (Figure 2). The trees are evergreen and clearly transpire throughout the year, however peak flow rates were recorded in February and March. These are the warmest months of the year, towards the end of the rainy season when water availability is good. In the middle of the growing season peak sap flow volumes of 50-100 L.day ¹ were recorded in these trees, which is five times greater than the volumes recorded in the yellowwood trees. These dropped to approximately 30 L.day ¹ in the late winter months.

120 0

110 10

100 20

90 30

80 Daily Rainfall (mm) 40

70 Transpiration -Pinus patula 50 Tree 1 (L/day) 60 60

50 70

40 80 Daily Rainfall (mm) Rainfall Daily

30 90 Daily Sap Flow (Transpiration) (l.day SapFlow(Transpiration) Daily ¹)

20 100

10 110

0 120 09 09 10 10 10 10 10 10 10 10 10 10 10 10 10 09 09 09 09 09 09 09 09 09 10 10 10 ------Jul Jul Jul Oct Oct Apr Apr Jan Jan Jun Jun - - - Feb Feb Mar Mar Aug Aug Sep Sep Nov Nov Dec Dec Dec Aug ------May May ------01 15 29 08 22 08 22 14 28 03 17 11 25 11 25 13 27 10 24 05 19 03 17 31 12 06 20

120 0

110 10 Daily Rainfall (mm) 100 20 Transpiration - Pinus patula 90 Tree 2 (L/day) 30

80 40

70 50

60 60

50 70

40 80 Daily Rainfall (mm) Rainfall Daily

30 90 Daily Sap Flow (Transpiration) (l.day SapFlow(Transpiration) Daily ¹)

20 100

10 110

0 120 09 09 10 10 10 10 10 10 10 10 10 09 09 09 09 09 09 09 09 09 10 10 10 10 10 10 10 ------Jul Jul Jul Oct Oct Apr Apr - - - Jan Jan Jun Jun Feb Feb Mar Mar Aug Aug Sep Sep Nov Nov Dec Dec Dec Aug - - - - May May ------01 15 29 08 22 08 22 14 28 03 17 11 25 11 25 13 27 10 24 05 19 03 17 31 12 06 20

Figure 2. Daily sap flow (water-use) volumes (L.day¹) recorded in 2 Pinus patula trees

Annual transpiration totals for the 2 study trees were averaged to determine a single transpiration rate for the Pinus patula plantation (Table 4).

Table 4. Weighting of observed transpiration volumes in 2 Pinus patula trees, to determine a representative total water-use

1-yr Water - 1-yr Water- HPV tree no. DBH (cm) use Weighting use (mm) (L.tree ¹)

Pinus patula 1 20.0 9849 804 50%

Pinus patula 2 25.6 16067 1312 50%

Weighted Total for plantation 12958 1058 100.0%

3.3 Stem growth increments and water-use efficiency

Using observed stem growth increments from the respective sample trees, water-use efficiency (defined as amount of stem biomass produced (g) per unit of water transpired (L)) was calculated for the respective study trees (Table 5).

Table 5. Summary of WUE data for indigenous Podocarpus henkelii trees and introduced Pinus patula trees, as calculated from a mass-based ratio of biomass increment (stem wood) over water-use 1-yr Wood WUE Stem Volume Stem Mass Tree Water- Density (g stem wood / L Increment (m³) Increment (g) use (L) (g cm ³) water transpired) P. henkelii 1 1755 0.00215 0.468 1006.20 0.5733 P. henkelii 2 5033 0.01088 0.468 5091.84 1.0117

P. henkelii 3 3554 0.00524 0.468 2452.32 0.6900

P. patula 1 9849 0.05157 0.38 19596.60 1.9897 P. patula 2 16067 0.09035 0.38 34333 2.1369

4. Discussion

The water-use efficiency results from this study show the introduced P. patula trees to be 2-4 times more water-use efficient than the indigenous P. henkelii trees. This trend is consistent with results from earlier studies (Olbrich et al., 1996; Dye et al., 2001; Gush & Dye, 2009), and supports the hypothesis that introduced species such as pines and eucalypts use water more efficiently than indigenous species (Figure 3). Evidence is emerging that the growth rate of plants is linked to WUE. Efficiency of resource use within forests (including water-use efficiency) has been shown to increase as forests increase their productivity and rate of resource use (Binkley at al. , 2004). This hypothesis has been tested and supported by various studies including Gyenge et al. (2008), who showed that the water-use efficiency (WUE) of a mixed species native forest (less productive system) was below that of an introduced Pseudotsuga menziesii (Mirb.) Franco (Douglas-fir) plantation (more productive system), both growing in the same area in Patagonia, Argentina. Using case studies from Eucalyptus plantations, Binkley et al ., (2004) and Stape et al. , (2004) have both demonstrated that more productive sites tend to have higher efficiencies of resource use than less productive sites, and silvicultural treatments may increase both resource supply and efficiencies of resource use. This is consistent with results from other studies, which have also shown that WUE is often well correlated with growth rate (Almeida et al. , 2007, Forrester et al. , 2010). So more productive systems (introduced plantations) appear to use water more efficiently than less productive systems (indigenous trees 8

6 Indigenous Tree Species Exotic Plantation Species 5 Mean

) Standard Deviation -1

1 Water-use Efficiency water L Efficiency (g wood. Water-use

0 P. patula P. patula P. patula P. patula P. patula P. patula P. P. patula P. B.zeyheri E.grandis E.grandis E.grandis E.grandis E.grandis E.grandis E.grandis E.grandis E.grandis E.grandis E.grandis E.grandis E.grandis E.grandis E.grandis E.grandis E.grandis E.grandis E.grandis P.falcatus P.falcatus C. africana C. T. orientalis T. P. obliquum P. O. europaea O. Species Eucalyptusclone Eucalyptusclone Eucalyptusclone Eucalyptusclone Eucalyptusclone Eucalyptusclone Eucalyptusclone Eucalyptusclone Eucalyptusclone P. patula P. patula 1 - study this P. patula 2 - study this P. henkelii P.henkelii 1 - this study P.henkelii 2 - this study P.henkelii 3 - this study Figure 3. A comparison of water-use efficiency data (using 1-year transpiration and stem mass increment measurements) from Eucalyptus clones, Eucalyptus grandis and Pinus patula trees (Olbrich et al., 1996; Dye et al., 2001; this study), and selected South African indigenous tree species (Gush & Dye, 2009; this study)

In terms of volumes of water-use on the other hand, it is noteworthy that compared to existing data, the indigenous species appear to use consistently less water (on average) than introduced species (Figure 4). Similar findings have been reported by Kagawa et al. (2009) and Little et al . (2009), who found that the water-use of native tree species, in Hawaii and Chile respectively, was considerably lower than that of introduced timber species. This finding is corroborated by similar results obtained by Licata et al. (2008) amongst gymnosperms in Patagonia, Argentina. They found that an introduced pine species ( Pinus ponderosa Doug. ex. Laws) used significantly more water than a native cypress specie ( Austrocedrus chilensis (D. Don) Pic. Serm. et Bizzarri).

The relatively lower water-use of the indigenous trees used in this study compared to introduced plantation species has some important implications. One potential application of this benefit could be the planting of indigenous tree species in riparian zones within commercially afforested areas. These zones are difficult to manage from grassland conservation and weed control perspectives as they are often narrow riparian corridors, in which it is dangerous to perform bi-annual burning regimes due to fire-risk within the plantations, and which are thus often heavily infested with alien invasive plants. The observed low water-use rates of indigenous tree species relative to introduced timber species make them a viable alternative land-use in these areas from a hydrological perspective. Other sectors in which the low water-use of indigenous tree species could play an important role include reforestation programmes, urban greening (e.g. gardens and street trees) and woodlots in support of rural livelihoods.

25000 Indigenous Tree Species Exotic Plantation Species Mean 20000 Standard Deviation ¹) ¹ ¹ year 15000

10000 Transpiration (kg tree Transpiration

5000

0 P. patula P. patula P. patula P. patula P. patula P. patula P. patula P. thisstudy B.zeyheri E.grandis E.grandis E.grandis E.grandis E.grandis E.grandis E.grandis E.grandis E.grandis E.grandis E.grandis E.grandis E.grandis E.grandis E.grandis E.grandis E.grandis E.grandis E.grandis P.falcatus P.falcatus C. africana C. - T. orientalis T. P. obliquum P. O. europaea O. 1 1 Eucalyptusclone Eucalyptusclone Eucalyptusclone Eucalyptusclone Eucalyptusclone Eucalyptusclone Eucalyptusclone Eucalyptusclone Eucalyptusclone Species P.patula P.patula2 - thisstudy P. henkelii1 henkelii1 P. - thisstudy henkelii2 P. - thisstudy henkelii3 P. - thisstudy Figure 4 A comparison between 1-yr total sap flow (transpiration), for Eucalyptus clones, Eucalyptus grandis and Pinus patula trees (Olbrich et al., 1996; Dye et al., 2001; this study), and selected South African indigenous tree species (Gush & Dye, 2009; this study – last three bars)

A number of issues are worthy of consideration regarding the potential expansion of indigenous tree systems. Firstly, the conversion of natural vegetation types to artificial vegetation types, e.g. grassland to indigenous tree plantations or even "forest" should be cautioned (Theo Stehle, Pers. Comm. Feb 2010). Plantation owners have to accept responsibility for the natural areas in-between plantations and to manage these as natural corridors, including natural forests, grasslands, fynbos and wetlands. These remnant natural ecosystems are regarded as very important to conserve natural habitats and species (fauna & flora). In terms of the CARA and other laws, riparian zones and wetlands are now managed as natural habitats and it is also a requirement for FSC certification. Converting these “open” areas into artificial ecosystems would defeat these objectives. It is wrong to think that all riparian zones need to be treed (forested). In nature this is only the case where habitat is suitable for forest. Arguments may be made about plantations creating unnatural habitats through absence of fire and micro-climate effects, however very well managed riparian zones and wetlands in between introduced plantations do exist, which demonstrates that these can in fact be successfully managed as natural ecosystems. However, if the riparian zone is continuously weed-infested and has limited bio-diversity value then an indigenous tree cover could be considered as an alternative option.

It is worth remembering that indigenous forests are not fire retardants, they just persist in fire ‘refugia’ areas, and it is still possible for a fire to run through an indigenous forest. This is important when considering potential planting sites – i.e. outside the fire zone and inside the fire ‘refugia’ areas. Planted and protected stands of introduced trees, intensive agriculture, road networks, urban development and settlements all create new fire ‘refugia’ areas, where the planting of indigenous tree production systems could be considered. However, distinction needs to be made between artificial and natural fire ‘refugia’. Artificial fire ‘refugia’ are ephemeral, such as within highly flammable pine stands, with greater inherent risks associated with those areas.

Nevertheless, there is certainly scope for the expansion of indigenous forest resources in the country (taking into account habitat requirements and fire risks) in terms of restoration of previously degraded forest areas. The rehabilitation of forests where they once occurred naturally and then disappeared by some or other cause, natural or man-made, is highly desirable. In those instances it would be beneficial to limit the water impacts and so the water-use of the species or combination of species to be re-established would be an important consideration. Furthermore, in the case of erosion control, appropriate measures to counteract erosion and rehabilitate eroded lands is a priority, even if suitable tree species have to be planted. In this case it would be important to understand the relative water use of different component tree species, for use in those areas to be planted up.

5. Conclusions

Results from the project so far indicate that indigenous tree species exhibit both lower water-use efficiency and lower overall water-use rates compared to introduced tree species. In other words, the relatively lower WUE of the indigenous species studied is more a consequence of slow growth rates as opposed to high water-use rates. As has been the experience with introduced plantation genera, genetic breeding for fast growth rates and the application of silvicultural practices employed in commercial plantations (e.g. pruning and thinning) could increase growth rates and improve the water-use efficiencies of indigenous species. However, the paradox of this is that increased growth rates will in turn lead to increased water-use rates, thereby negating the benefits of low water-use.

Given the wide range of climatic and site conditions around South Africa, the large number of indigenous species that are found in this country, and our dwindling water resources, it is important to identify new and sustainable indigenous forest and woodland production systems. Considering that on the one hand further afforestation with commercial forest species is now severely restricted due to concerns about reductions in catchment water yields, while on the other hand significant potential exists to expand indigenous tree systems, the possibility of low water-use forms of forestry is an attractive proposition. There have already been some pioneering attempts to increase the area under indigenous trees in the form of experimental plantations of indigenous trees (Hans Merensky, Komatiland forests), as well as indigenous tree planting / livelihood programmes such as the Wildlands Conservation Trust “Indigenous Trees for Life” and Sappi Sandisa Imvelo initiatives. While growth rates are still slow, these initiatives together with other financial incentives for planting indigenous trees such as carbon sequestration credits and payment for ecosystem services (PES) schemes look set to expand the area under indigenous trees in South Africa. As this happens it will be useful to know more about the water-use requirements, growth rates and economic potential of South Africa’s indigenous tree resources.

Acknowledgements

The research reported on here formed part of a project solicited and initiated by the Water Research Commission of South Africa (WRC), and was co-funded by the Working for Water Programme of the South African Dept. of Water Affairs. Their support is gratefully acknowledged. Mondi (particularly Doug Burden) is thanked for allowing the monitoring of trees on their estate. Technical assistance in the field by CSIR colleagues Mr. Alistair Clulow and Mr. Vivek Naiken is appreciated.

References

ALMEIDA, A.C., SOARES, J.V., LANDSBERG, J.J. & REZENDE, G.D. (2007). Growth and water balance of Eucalyptus grandis hybrid plantations in Brazil during a rotation for pulp production. Forest Ecology and Management , 251: 10–21.

BINKLEY, D., STAPE, J.L., & RYAN, M.G. (2004). Thinking about efficiency of resource use in forests, Forest Ecology and Management , 193: 5-16.

BURGESS, S.O., ADAMS, M.A., TURNER, N.C., BEVERLY, C.R, ONG, C.K., KHAN, A.A.H. & BLEBY, T.M. (2001). An improved heat pulse method to measure low and reverse rates of sap flow in woody plants. Tree Physiology 21: 589-598. 11

CHAMBERLAIN, D., ESSOP, H., HOUGAARD, C., MALHERBE, S. & WALKER, R. (2005). The GENESIS Report, Part I: The contribution, costs and development opportunities of the Forestry, Timber, Pulp and Paper industries in South Africa. Report to the Department of Trade and Industry and the Pulp and Paper Manufacturers Association, South Africa . Genesis Analytics.

DYE, P.J. (1996). Response of Eucalyptus grandis trees to soil water deficits. Tree Physiology , 16: 233–238.

DYE, P.J. & OLBRICH, B.W. (1993). Estimating transpiration from 6-year-old Eucalyptus grandis trees: development of a canopy conductance model and comparison with independent sap flux measurements. Plant Cell Environment , 16: 45–53.

DYE, P.J. & VERSFELD, D. (2007). Managing the hydrological impacts of South African plantation forests: An overview. Forest Ecology and Management , 251: 121-158.

DYE, P.J., GUSH, M.B., EVERSON, C.S., JARMAIN, C., CLULOW, A., MENGISTU, M., GELDENHUYS, C.J., WISE, R., SCHOLES, R.J., ARCHIBALD, S. & SAVAGE, M.J. (2008). Water-use in relation to biomass of indigenous tree species in woodland, forest and/or plantation conditions. WRC Report TT 361/08, Water Research Commission, Pretoria, South Africa.

DYE, P.J., POULTER, A.G., HUDSON, K.E. & SOKO, S. (1996). A comparison of the water-use of common riparian forests and grasslands. Report FOR-DEA 962 , Division of Forest Science and Technology, CSIR, Pretoria, South Africa.

DYE, P.J., SOKO, S & POULTER, A.G. (1996). Evaluation of the heat pulse velocity method for measuring sap flow in Pinus patula . Journal of Experimental Botany , 47, 975–981.

DYE, P.J., VILAKAZI, P., GUSH, M.B., NDLELA, R. & ROYAPPEN, M. (2001). Investigation of the feasibility of using trunk growth increments to estimate water use of Eucalyptus grandis and Pinus patula plantations. WRC Report 809/1/01 , Water Research Commission, Pretoria, South Africa.

FORRESTER, D.I., COLLOPY, J.J. & MORRIS, J.D. (2010). Transpiration along an age series of Eucalyptus globulus plantations in southeastern Australia, Forest Ecology and Management, 259: 1754–1760.

GUSH, M.B. (2008). Measurement of water-use by Jatropha curcas L. using the heat pulse velocity technique. Water SA, Vol. 34 (5): 579-583.

GUSH, M.B. & DYE, P.J. (2009). Water-use efficiency within a selection of indigenous and exotic tree species in South Africa as determined using sap flow and biomass measurements. Acta Horticulturae, 846: 323-330.

GYENGE, J., FERNÁNDEZ, M.E., SARASOLA, M. & SCHLICHTER, T. (2008). Testing a hypothesis of the relationship between productivity and water use efficiency in Patagonian forests with native and exotic species. Forest Ecology and Management , 255: 3281-3287.

JARMAIN, C., EVERSON, C.S., SAVAGE, M.J., MENGISTU, M.G., CLULOW, A.D. & GUSH, M.B. (2008). Refining tools for evaporation monitoring in support of water resources management. WRC Report No. 1567/1/08 , Water Research Commission, Pretoria.

KAGAWA, A. SACK, L. DUARTE, K. & JAMES, S. (2009). Hawaiian native forest conserves water relative to timber plantation: Species and stand traits influence water use. Ecological Applications , 19 (6): 1429-1443.

LAWES, M.J., OBIRI, J.A.F. & EELEY, H.A.C. (2004). The uses and value of indigenous forest resources in South Africa. In: LAWES, M.J., EELEY, H.A.C, SHACKLETON, C.M. & GEACH, B.G.S. (eds). Indigenous forests and woodlands in South Africa: Policy, people and practice. University of KwaZulu-Natal Press, Scottsville, South Africa. P. 227-273.

LICATA, J.A., GYENGE, J.E., FERNÁNDEZ, M.E., SCHLICHTER, T.M. & BOND, B.J. (2008). Increased water use by ponderosa pine plantations in northwestern Patagonia, Argentina compared with native forest vegetation. Forest Ecology and Management, 255: 753–764.

LITTLE, C., LARA, A., MCPHEE, J. & URRUTIA, R. (2009). Revealing the impact of forest exotic plantations on water yield in large scale watersheds in South-Central Chile. Journal of Hydrology , 374: 162– 170.

LYNCH, S. D. & SCHULZE, R. E. (2006). Rainfall Databases. In: Schulze, R. E. (Ed). 2006. South African Atlas of Climatology and Agrohydrology. Water Research Commission Report, 1489/1/06. , Section 2.2. WRC, Pretoria, South Africa.

MARSHALL, D.C. (1958). Measurement of sap flow in conifers by heat transport. Plant Physiology , 33: 385- 396.

MUCINA, L. & RUTHERFORD, M.C. (2006). The vegetation of South Africa, Lesotho and Swaziland. Strelitzia 19. South African National Biodiversity Institute, Pretoria.

OLBRICH, B., OLBRICH, K., DYE, P. & SOKO, S. (1996). A year-long comparison of water use efficiency of stressed and non-stressed E. grandis and P. patula : findings and management recommendations. Un-published Report FOR-DEA 958, CSIR, Pretoria.

SHACKLETON, C.M., SHACKLETON S.E, BUITEN, E. AND BIRD, N. (2007). The importance of dry woodlands and forests in rural livelihoods and poverty alleviation in South Africa. Forest Policy and Economics , 9 (5): 558-577.

STAPE, J.L., BINKLEY, D., & RYAN, M.G. (2004). Eucalyptus production and the supply, use and efficiency of use of water, light and nitrogen across a geographic gradient in Brazil. Forest Ecology and Management, 193: 17-31.

SWANSON, R.H. & WHITFIELD, D.W.A. (1981). A numerical analysis of heat pulse velocity theory and practice. Journal of Experimental Botany, 32: 221-239.

VERRYN, S. D. (2000). Eucalypt hybrid breeding in South Africa. In: DUNGEY, H. S., DIETERS, M. J., & NIKLES, D. G. (eds). Hybrid Breeding and Genetics of Forest Trees . pp. 191-199.QFRI/CRC-SPF, Noosa, Australia.

VON BREITENBACH, F. (1990). National list of indigenous trees. Dendrological Foundation , Pretoria.

WATER RESEARCH COMMISSION. (2010). Abridged Knowledge Review 2009/10 - Growing Knowledge for South Africa’s Water Future. WRC, Pretoria. pp. 59-60.