THE INFLUENCE OF DEFORESTATION ON SOIL WATER CONSERVATION IN A PINE FOREST IN TENERIFE (CANARY ISLANDS, SPAIN)

ISCO 2004 - 13th International Soil Conservation Organisation Conference – Brisbane, July 2004
Conserving Soil and Water for Society: Sharing Solutions


THE INFLUENCE OF DEFORESTATION ON SOIL WATER CONSERVATION IN A PINE
FOREST IN TENERIFE (CANARY ISLANDS, SPAIN)

M. Tejedor, C. Jiménez, C. Monteverde and J. Parra
University of La Laguna, Department of Soil Science and Geology, Faculty of Biology, Avda. Astrofísico
Francisco Sánchez s/n, 38204 La Laguna, Tenerife, Canary Islands, Spain, e-mail: martesa@ull.es

Abstract
The present work studies the effect of deforestation on soil water conservation in a pine forest in Tenerife (Canary
Islands, Spain). We compare two adjoining plots, with originally similar environmental characteristics but now
differentiated by the presence or absence of canarian pines (Pinus canariensis) following the removal from one of
the plots of trees for use as timber. In both areas soil moisture was monitored monthly every 10 cm to a depth of 1
metre over a period of three years. Also analysed were the spatial variability of the soil water in the pine forest and
the influence of the type of vegetation thereon. The results obtained show differences in the soil moisture content
distribution throughout the year. The importance of water uptake from condensation is also apparent. The increase
in seasonal contrasts and the prolongation of the dry period caused by deforestation influence the soil moisture
regime, which changes from udic under the pine trees to xeric in the deforested zone.

Additional Keywords: Land use, soil moisture regime, Soil Taxonomy, condensation

Introduction
The direct and indirect influence of vegetation on the soil water cycle has received extensive consideration in the
literature (Zinke 1962; Huber and Oyarzun 1990; Domingo et al. 1994; Putuhena and Cordery 1996; Calvo and
Gómez 2002). Some works have focused on the type of species, the age, height and degree of cover of the
vegetation, while others have considered the stemflow effect of the trunks and the consequences this has for water
concentration at the tree base and for the spatial variability. Others, meanwhile, have focused on the importance of
the water falling on the undergrowth from tree branches and leaves or on the impact of rainwater interception and
condensation phenomena. Fewer papers have addressed specifically the influence exerted by changes in plant cover
on soil moisture regimes from the Soil Taxonomy classification point of view (Soil Survey Staff 1999), even
though the type of cover is often mentioned in discussing the regimes of a specific region.

In the Canary Islands, where the use of groundwater is essential, studies contributing to better knowledge of
aquifers and aquifer recharge is of great interest. In the present paper we analyse the effect of deforestation of a
Canarian pine forest in Tenerife on soil water reserves and on the soil moisture regime. We examine also the spatial
variability of the soil moisture in the pine forest and how this is influenced by the distance to the trees.

Materials and Methods
Study area
The island of Tenerife, Canary Islands, Spain (highest point, 3718 m.a.s.l.) has a wide variety of microclimates,
depending on altitude, orientation, orography, the effects of the trade winds and influence of the sea. Moisture from
the trade winds condenses extensively on the northern side at heights between 900-1400 m.a.s.l. (our study zone is
situated at the latter figure). The total amount of water produced by condensation can be several times greater than
the level of rainfall (Kämmer 1974; Santana 1990; Marzol 1981, 1993).

Two adjacent zones at 1300 m.a.s.l. were selected for the study. The first had canarian pine with virtually no
underbrush, a tree density of 0.045 individuals/m2, a basal area of 80 m2/ha and over 75% cover. The second,
smaller in size, was replanted with eucalyptus trees (Eucaliptus globulus) which were felled for wood shortly
before the experiment commenced. During the study period, this second zone became covered with herbaceous
plants and species of tree-heath (Erica arborea), none of which exceeded 1 metre in height. The average annual air
temperature was 14.2ºC and annual rainfall of approximately 700 mm with a seasonal distribution of 47% in
winter, 24% in spring, 3% in summer and 26 % in autumn. The soils were Dystric Hapludands/Haploxerands (Soil
Survey Staff 1999), silty loam in texture and rich in organic matter, which is fully deeply incorporated (more than
6% at a depth of 1 metre). They were formed by a mixture of pyroclasts and basaltic scoria. In some parts,
particularly those covered with the herbaceous plants, the scoria is found at less than one metre from the surface.
The modification of the vegetation has led to differences in the surface horizon: in the soil under pine cover the O
Paper

No.760
page
1

---

ISCO 2004 - 13th International Soil Conservation Organisation Conference – Brisbane, July 2004
Conserving Soil and Water for Society: Sharing Solutions


horizon is more or less compact, with accumulation of organic matter, pine needles. This horizon is not present in
the soil covered by herbaceous plants.

Methods
In both study zones, sampling for soil moisture was conducted monthly over a period of three years (October 2000-
December 2003) every 10 cm to a depth of 1 metre. However, in the deforested plot the maximum sampling depth
tended to be around 80cm due to the presence of a layer of scoria. Three random replications were made on each
date and for each plot. Sampling never took place within 48 h of rainfall. The samples were taken using an
Eijelkamp probe and moisture content was determined by the gravimetric method. The monthly moisture content in
both zones was compared using the Wilcoxon non-parametric test, given that the data distribution did not meet the
requirements for parametric analysis.

To study the spatial variability of the soil moisture and the influence exerted by the pine forest, a 20x20 plot was
selected and the position of each tree was mapped. A 10x10 sub-plot was also selected in which 40 points were
chosen at random and the distance was recorded between each point and the three nearest trees (distance 1,
distance 2 and distance 3). In March 2004 each point was sampled every 10 cm between 10-60 cm depth. During
the previous three years, the greatest differences between the pine and herbaceous soils had been observed in the
same month and in the same depth segment. Unilateral Spearman correlations and Principal Components Analysis
were used for the statistical analysis. The statistics packages used were SPSS 11.0.1 (SPSS Inc. 2001) and Canoco
for Windows 4.0 (ter Braak and Smilauer 1998).

Results and Discussion
The results of the Principal Components Analysis (Figure 1) show a clear distribution pattern for the soil moisture
in the pine-forest plot studied. Most of the variation (55.4%) is linked to the distance to the nearest tree, which is
inversely related to the soil moisture content. This result is corroborated to 95% significance in the unilateral
Spearman correlation, with a negative sign between soil moisture content and distance to the nearest tree (Table 1).
As is logical, analysis of the distances to the second and third trees (22.0% and 8.7% of variance respectively)
corroborates that the influence on soil moisture diminishes as the distance increases. In this heterogeneous
distribution pattern of the soil water, trickle from the trunks appear more influential than water translocation or
direct precipitation not intercepted by the undergrowth. The high dispersion with respect to the factors indicates
that although the trend is clearly defined, the extent of the influence of the trees is limited. In the case of the pine
forest, the area of influence of each tree overlaps with that of near neighbours and the moisture gradient effect is
thus cushioned by the presence of other gradients.

Table 1. Descriptive parameters of spatial variation and Spearman correlations between cumulative soil
moisture and distance 1, distance 2 and distance 3 (no: non-sig.,*: 95% sig.)

Mean (mm)
St. dev
St. error
Cumulative soil moisture content
118.5
12.9
3.6

distance 1
distance 2
distance 3
Spearman correlations
-.364 *
-.122 no
-.314 *

Having seen the representativeness of the sampling over the three years, we will now consider the analysis. Figure
2 shows the evolution of cumulative moisture content at a depth of 80 cm in the plot under pine forest and the zone
covered by herbaceous plants for all the sample dates. The annual tendency in the soil water content seen over the
three years is maintained: higher content in the deforested plot during rainy periods and during the two or three
subsequent months, lower content during summer and autumn. These differences are statistically significant during
January, February, March, April, May and August.

Paper

No.760
page
2

---

ISCO 2004 - 13th International Soil Conservation Organisation Conference – Brisbane, July 2004
Conserving Soil and Water for Society: Sharing Solutions


40
31
distance 2
1
24
16
21
4
27
32
11
5
30
cumulative moist.
39 29
distance 3
36
26
23
28
6
15
22
35
17
8
3
33
2
13
14
19
25
distance 1
20
9
18
10
37
12
34
7
Axes: 1 2 3 4
Eigenvalues: .554 .220 .139 .087

Figure 1. Principal Components Analysis. Influence of distances to nearest trees
on cumulative soil moisture content

Analysis of the seasonal distribution of cumulative soil water throughout the year produced the following results.
Pine forest: 29% in winter, 28% in spring, 19% in summer and 24% in autumn. Deforested zone: 31% in winter,
28% in spring, 15% in winter, and 26% in autumn. The regular soil moisture distribution in the pine forest, which
differs greatly to that corresponding to rainfall, indicates the presence of additional non-rainfall sources of water,
due -we believe- of the effect of condensation. The importance of this effect is not so much the amount of water but
its year-round distribution (Parra 2001), particularly during months of little rain. It is worth noting that in August
the amount of water is significantly higher in the pine forest soil, where condensation is greater. The condensation
effect is still seen in the deforested soil, which is more heavily influenced by rainfall. This effect may be due to the
fact that the selected site is surrounded by pine forest and during the study period became populated with
herbaceous plants, which also propitiate condensation of fog. In addition to the important influence of the
condensation, the role played by the different types of cover in reducing moisture loss through evaporation and,
secondly, the water consumed by the vegetation are also factors to be taken into account.
525
475
)

(
%
425
r
e
u
ist
o 375
m
il
325
i
ve so
lat
u 275
m
u
C
225
2 0 0 1
2 0 0 2
2 0 0 3
175
n
j
m
m
j
s
n
j
m
m
j
s
n
j
m
m
j
s
o
d
Date (m onth)
Pi newood
Def or ested

Figure 2. Evolution of cumulative moisture content. Effective soil depth: 80 cm

The differences in water balance influence the soil moisture regimes (Soil Survey Staff 1999). These regimes are
defined in the Soil Taxonomy on the basis of the presence or absence of water at pressures below 1500 kPa, over a
given number of days per year and at a given depth segment. Deforestation has propitiated seasonal contrasts and
Paper

No.760
page
3

---

ISCO 2004 - 13th International Soil Conservation Organisation Conference – Brisbane, July 2004
Conserving Soil and Water for Society: Sharing Solutions


an increase in the number of days on which the soil is totally dry (over 90 days). Given this circumstance and the
fact that the soil temperature regime is isomesic in the case of the pine forest (Tejedor et al. 2003) and mesic for the
deforested zone (Rodríguez Paz 2004), we could tentatively define an udic regime for the pine soil and a xeric
regime for the deforested soil, to be confirmed in further studies.

References
Calvo de Anta, R., Gómez Rey, M.X. (2002). Distribución espacial del ciclo del agua en suelos forestales con Pinus radiata de Galicia (NO
de España). Edafología 9(1), 61-84.
Domingo, F., Puigfabregas, J., Moro, M.J., Bellot, J. (1994). Role of vegetation cover in the biogeochemical balances of a small a forested
catchment in southeastern Spainh. Journal of Hydrology 159, 275-289.
Huber, A.W., Oyarzun, C.
990). Variacion
E. (1
es anuales en precipitación, escurrimiento e intercepción en un
de Pinu
bosque adulto
s
radiata. Turrialba 40, 503-508.
Kämmer, F. (1974). Klima und vegetation auf Tenerife besonders in Hiblick auf Nebelniederschlag. Scripta Geobotanica. Vol. 7 Edit Erich
Goltze KG. Göttingen.
Marzol, M.V. (1981). El clima de montaña de la isla de Tenerife. Variaciones en el gradiente térmico. VII Coloquio de Geografía, Pamplona.
Tomo 1. p 163-168.
Marzol, M.V. (1993). El Clima: Rasgos generales. Geografía de Canarias. Consejería de Obras Públicas, Vivienda y Aguas, Gobierno de
Canarias. p101-116.
Parra López, J. (2001). Estimación del régimen hídrico del suelo de dos zonas de la Isla de Tenerife. Trabajo de fin de
e
carr ra. Escuela
Técnica Superior de Ing. Agrónomos y Montes. Universidad de Córdoba.
Putuhena, W.M., Cordery, I. (1996). Estimation of interception capacity of the forest floor. Journal of Hydrology 180, 283-299.
Rodríguez Paz (2004). Comunicación personal.
Santana, L. (1990). La importancia hidrológica de las nieblas en las cumbres del Parque Nacional de Garajonay, en P.L. Pérez de Paz (e d.)
“Parque Nacional de Garajonay, Patrimonio Mundial”. Exmo. Cabildo Insular de La Gomera, ICONA. p66-71.
Soil Survey Staff. (1999). Soil Taxonomy; a basic system of soil classification for making and interpreting soil surveys. 2ª Edition. United
States Department of Agriculture. Natural Resources Conservation Service, Agricultural Handbook, 436. U.S. Gov. Print. Office,
Washington, DC.
Tejedor, M., Jiménez, C., Rodriguez Paz, M., Hernández Moreno, J.M. (2003). Soil Temperature Regimes on the island of Tenerife.
Altitudinal sequence. 2003 Annual Meeting Changing Sciences for a Changing World: Building a Broader Vision. Denver Colorado.
Ter Braak, C.F.J., Smilauer, P. (1998). Canoco Reference Manual and User’s Guide to Cano
Canonical
co for Windows: Software for
Community Ordination, Version 4. Microcomputer Power, Ithaca, NY.
Zinke, PJ. (1962). The pattern of influence of individual forest trees on soil properties. Ecology 43, 130-133.



Paper

No.760
page
4

---