|| Home || Search || Contact || Agroforestry Program ||

In recent years, farmers have been putting trees back in, hoping that they will lower the water table and reverse the salinisation process. Yet farmers and other land managers are still largely in the dark over the potential of trees to do the job. Paul Raper of Agriculture WA recently reviewed the scientific literature on the water use of trees and shrubs in Mediterranean climates. He found that despite the hostile environments they are sometimes planted in, tree plantations can lower water tables sufficiently to reduce soil salinity. Good design and species selection are essential to achieving optimal results.

How trees can reduce the water table

The planting of trees and shrubs on land cleared for agriculture can help lower water tables in two ways: by limiting the amount of rainfall that seeps through to the water table (groundwater recharge) and by directly using groundwater for transpiration.

Scientists can measure the effects of plantations on groundwater by two means: (i) by measuring the amount of water used by individual trees and calculating the effect of this usage, and (ii) by monitoring fluctuations in water table levels for paired catchments with and without plantations. Both are limited in their reliability and require very careful experimentation.

Changes in groundwater after revegetation

A number of trials have been conducted by the Water Authority of Western Australia and other researchers who aimed to establish the effect of revegetation on groundwater levels. Most were in the catchments of the Wellington Dam and Mundaring Weir, Western Australia, where the average rainfall is between 600 and 900 mm per year. Table 1 shows the effect of different revegetation strategies on groundwater levels.

Raper found that despite below-average rainfall over the life of these trials, groundwater levels under pasture sites still rose. In addition, whilst absolute groundwater levels had not fallen under all areas where tree plantations were established, usually there was a decline relative to pasture sites.

Nevertheless, at some sites where absolute levels had declined, the effect was insufficient to reduce the discharge of saline groundwater to streams—which was the reason for planting trees in the first place. Further, under climatic conditions more in line with the long-term average, or with above-average rainfall, the effects of trees on groundwater may have been less than that observed. There are also some documented cases where tree planting failed to significantly change the rate of groundwater rise, but these failures can generally be attributed to poor site selection and the small areas planted.

More trees, more change

Figure 1 presents data on the rate of change in groundwater level against a measure of reforestation, based on trial results in the 600–900 mm rainfall zone of Western Australia. In general, the greater proportion of the catchment reforested and the greater the crown cover, the greater the effect on groundwater levels.

According to Raper, this is only what intuition would tell you. He suggests there must also be some limiting value where tree density is such that water use from the stand approaches atmospheric demand, the species reaches its physiological maximum water use, or transpiration is limited by water supply. Beyond this point, increasing planting density may not have any greater effect on the rate of groundwater reduction.
 
 

Trial site and location in the catchment	Rainfall (mm/yr)	Pan evaporation (mm/yr)	Initial depth to groundwater (m)	Initial groundwater conductivity (mS/m)	Proportion catchment cleared (%)	Proportion cleared area revegetated (%)	Change in groundwater level	Reduction in groundwater level relative to pasture (m)
Lower Slope, Discharge Zone
Stene's Valley	713	1,600	6.3	1,030	44	35	-1.47 m (12 yrs)	2.0
Maringee Farms	650	1,600	*0-5	1,870	54	22	-0.4 m (13 yrs)	3.65
Wide-Spaced Plantations
Flynn's Agroforestry	717	1,800	4.4	1,140	51	58	-2.2 m (11 yrs)	1.0
Stene's Agroforestry	713	1,600	2.7	1,140	25	57	-1.7 m (12 yrs)	3.3
Boundain	505		1.3	1,870	>95	21	-1.0 m (11 yrs)	0.75
Strip, Landscape Plantings
Flynn's Landscape	717	1,800	2.1	1,400	98	13	-1.3 m (13 yrs)	-0.1
Stene's Strip Planting	713	1,600	2.7	1,450	31	14	+0.1 m (15 yrs)	2.0
Bannister	800		0.5	640	83	14
Extensive Plantations
Flynn's Hillslope	717	1,800	*0.9-5.8	1,400	100	54	-3.0 m (13 yrs)	1.7
Stene's Arboretum	713	1,600	*3-9.5	1,030	35	70	-5.5 m (12 yrs)	7.7

 

Measuring tree water use

Raper examined experiments conducted explicitly to quantify the water use of trees and other woody perennial vegetation likely to be used in salinity control in the Mediterranean climatic zones of southern Australia. Table 2 shows reported values of annual tree water use in plantations or agroforestry situations in the agricultural regions of Western Australia.

One of the values reported in Table 2 (as well as in Table 1) is pan evaporation (Epan). This is the amount of water that would be evaporated out of a pan placed at ground level filled with a limitless supply of water. The ratio of actual tree water use (Eact) to Epan, shown in Table 2, can be used as a measure of water use potential.

Raper found comparing data from different experiments far from straight forward. He says that, unfortunately, the vast majority of tree water use estimates are made over a period of only one year, and often these are extrapolated from observations made on a single day once each month. This makes it difficult to see and account for the effects of seasonal variability, which can significantly affect tree water use. In addition, the different techniques employed to gauge tree water use do not all measure the same thing. For example, some include rainfall interception and bare soil evaporation while others measure only transpiration.

Despite these difficulties, Raper drew several conclusions. First, trees in plantations or agroforestry configurations surrounded by land managed for agriculture generally use more water than remnant vegetation. This is probably due to increased water availability. Trees planted on previously cleared land commonly use between 30–55% of annual pan evaporation, while healthy trees planted at high density can use up to about 65% of annual pan evaporation.

Raper found some evidence that under ideal conditions, maximum long-term water use could be up to 75% of annual pan evaporation. However, he also found that trees planted in landscape positions where they may be expected to have access to plentiful water have, in some instances, used only a relatively small proportion of annual pan evaporation. This suggests that supply may have been limiting during some part of the year or that there are physiological limits to tree water use.

Factors affecting tree water use

Raper also examined some of the factors that might affect tree water use. Soil salinity, for example, can reduce water use by reducing tree survival and growth rates. Similarly, waterlogging and shallow groundwater reduces tree rooting depth and hence the rate at which trees can transpire.

Different species will be more or less tolerant of soil salinity and water logging. This is illustrated in , which shows the relative growth of several eucalypt species and Casuarina obesa grown in a glasshouse under waterlogged, saline and saline waterlogged conditions.

Tree density and leaf area

Experimental evidence suggests that water use increases with leaf area, or crown cover, until canopy closure. Planting density tends to affect water use per tree rather than per unit area. The key to achieving maximum water use in a plantation is leaf area—in the early years, a higher leaf area may be achieved by denser planting, but once canopy closure has occurred the number of trees appears to be immaterial.

Given the importance of leaf area, Raper draws an important and logical conclusion: the tree that is healthiest and which can grow the largest canopy at a location is the one that will transpire the most water. Selection of species in relation to site conditions is therefore critical in maximising the effect of trees on the local water balance and hence on groundwater levels.

Modelling tree water use

Based on data produced by various trials reviewed in his report, Raper developed a simple model to determine the annual tree water use which incorporates pan evaporation, average depth to groundwater and soil salinity. The model doesn’t allow for differences in water use that arise from soil texture and depth, orientation and planting density.

Raper suggests that the model could be used as a guide to land managers interested in estimating the effect of revegetation on the local water balance in situations where detailed site information is not available. It could also form a working hypothesis to be tested by future research.

References

Carbon, B.A., Roberts, F.J., Farrington, P. and Beresford, J.D., 1982. Deep drainage and water use of forests and pastures grown on deep sands in a mediterranean environment. J. Hydrol., 55: 53-64.

Eastham, J., Scott, P.R. and Steckis, R.A., 1993. Evaluation of Eucalyptus camaldulensis (river gum) and Chamaecytisus proliferus (tagasaste) for salinity control by agroforestry. Land Degrad. Rehab., 4: 113–122.

George, R.J., 1990. Reclaiming sandplain seeps by intercepting perched groundwater with Eucalypts. Land Degrad. Rehab., 2: 13-25.

Greenwood, E.A.N., Beresford, J.D. and Bartle, J.R., 1981. Evaporation from vegetation in landscapes developing secondary salinity using the ventilated-chamber technique. III. Evaporation from a Pinus radiata tree and the surrounding pasture in an agroforestry plantation. J. Hydrol., 50: 155-166.

Greenwood, E.A.N., Beresford, J.D., Bartle, J.R. and Barron, R.J.W., 1982. Evaporation from vegetation in landscapes developing secondary salinity using the ventilated-chamber technique. IV. Evaporation from a regenerating forest of Eucalyptus wandoo on land formerly cleared for agriculture. J. Hydrol., 58: 357-366.

Greenwood, E.A.N., Klein, L., Beresford, J.D. and Watson, G.D., 1985a. Differences in annual evaporation between grazed pasture and Eucalyptus species in plantations on a saline farm catchment. J. Hydrol.,78: 261-278.

Greenwood, E.A.N., Klein, L., Beresford, J.D., Watson, G.D. and Wright, K.D., 1985b. Evaporation from the understorey in the Jarrah (Eucalyptus marginata Don ex Sm.) forest, southwestern Australia. J. Hydrol., 80: 337-349.

Hookey, G.R., Loh, I.C. and Bartle, J.R., 1987. Water use of Eucalypts above saline groundwater. Water Auth. W.A., Hydrology Branch WH 32. Perth, WA, 89 pp.

Marshall, J.K., Morgan, A.L., Akilan, K., Farrell, R.C.C. and Bell, D.T., 1997. Water uptake by two river red gum (Eucalyptus camaldulensis) clones in a discharge site plantation in the Western Australian wheatbelt. J. Hydrol., 200: 136–148.

Salama, R.B., Bartle, G.A. and Farrington, P., 1994. Water Use of plantation Eucalyptus camaldulensis estimated by groundwater hydrograph separation techniques and heat pulse method. J. Hydrol.,156(1-4): 163-180.