DTD 5

Forest Ecology and Management xxx (2005) xxx–xxx

1 www.elsevier.com/locate/foreco

2 Use of quantitative methods to determine leaf biomass on

3 15 woody shrub species in northeastern Mexico

OF
4 Rahim Foroughbakhch *, Gregorio Reyes, Marco A. Alvarado-Vázquez,
5 Jorge Hernández-Piñero, Alejandra Rocha-Estrada
6 Departamento de Botánica, Facultad de Ciencias Biológicas, Universidad Autónoma de Nuevo León,
7 Box 2-F, Cd. Universitaria, San Nicolás de los Garza, NL 66451, Mexico

RO
8 Received 14 January 2005; received in revised form 21 May 2005; accepted 23 May 2005
9
10
11

12 Abstract

13
14
DP
The coast of the Gulf of Mexico is characterized by dry regions with a high variation in climatic conditions. This area is rich
in drought-tolerant or subhumid species. In order to determine the most appropriate method in evaluating the leaf biomass of 15
15 shrubs species in a native matorral (thornscrub) of northeastern Mexico, a study was conducted to compare the following non-
16 destructive methods: (1) Adelaide, (2) double sampling or dimensional analysis and (3) double sampling of branches. The non-
17 destructive methods allow indirect relationships between leaf biomass and some ecological characteristics of the plants by using
18 regression models. The methods were used to estimate leaf weight in each species. Ecological and morphological characteristics
TE

19 of these species were determined using a structural analysis. No single standard method estimated leaf biomass for all species on
20 site, because of the diversity of forms in these species. However, Adelaide and the dimensional were the most precise, practical
21 and simplest methods so they could be considered the method of choice for measuring the forage leaf biomass of many shrub
22 species like Acacia rigidula Benth. (r2 = 0.98), Bernardia myricaefolia Wats. (r2 = 0.94), Caesalpinia mexicana A. Gray
23 (r2 = 0.92), Leucophyllum frutescens (Berl) I.M. Johnst. (r2 = 0.95) y Zanthoxylum fagara (L.) Sarg. (r2 = 0.93), Celtis pallida
Torr. (r2 = 0.99), Cordia boissieri A. DC. (r2 = 0.83), Parkinsonia aculeata L. (r2 = 0.83), among other species.
EC

24
25 # 2005 Published by Elsevier B.V.

26 Keywords: Shrubs; Leaf biomass; Structural analysis; Tamaulipan thornscrub; Matorral; Mexico
28
27
RR

29 1. Introduction composition, height, density and plant associations 33

(Battey, 2000; Eviner, 2003). 34
30 The great variability in climatic and edaphic The various species occurring in the northeastern 35
31 conditions of arid and semiarid zones causes region of Mexico can be categorized in several groups 36
32 extremely diverse shrublands in terms of the species based on their ecological adaptations and forestry 37
CO

uses. The Tamaulipan shrubland occurs in the state of 38

* Corresponding author. Tel.: +52 81 8114 3465/1238 6249.

Nuevo Leon, extending from the coastal plain of the 39
E-mail addresses: [email protected], [email protected] Gulf of Mexico to the southern rim of Texas, USA 40
(R. Foroughbakhch). (Foroughbakhch, 1992; Foroughbakhch et al., 2001). 41

0378-1127/$ – see front matter # 2005 Published by Elsevier B.V.

UN

doi:10.1016/j.foreco.2005.05.046

FORECO 9357 1–8

 DTD 5

2 R. Foroughbakhch et al. / Forest Ecology and Management xxx (2005) xxx–xxx

42 Woody plants of shrubberies are an economically The climate is semiarid with two rainy seasons 86
43 important forage source for rural population (von (summer and autumn) and a dry spell between 87
44 Maydel, 1996). Grazing and fragmentation in large November and April. Mean annual precipitation is 88
45 scale has been practiced for 350 years in these areas 749 mm (Foroughbakhch, 1992). Mean annual tem- 89
46 (Fahring, 2003). This type of grazing, in the long run, perature is 22.3 8C with temperature over 40 8C during 90
47 results in loss of both quality and quantity of biomass the summer, and frost from December to March. The 91
48 and forage plant species, followed by reduction in the water budget is unbalanced. The ratio of precipitation 92
49 plant layer which covers and protects the soil (Reid to free evaporation is 0.48 and precipitation to 93
50 et al., 1990). potential evaporation is 0.62. 94
51 This situation could only be improved through a Most soils of region are of rocky type of Upper 95

OF
52 management plan under a sound livestock and Cretaceous calcite and dolomite. The dominant soils 96
53 agroforestry scheme, in which foliar biomass produc- are deep, dark grey, lime-clay vertisols which are the 97
54 tion of the shrubland is related to animal productivity. result of aluvial and coluvial processes. They are 98
55 So far, the animals have received more attention than characterized by high clay and calcium carbonate 99
56 plants, despite the fact that plants indeed form the content (pH 7.5–8.5) and low organic matter content. 100

RO
57 basis for livestock exploitation (Blair, 1990). Further- Analysis of major nutrients reveals phosphorus and 101
58 more, little effort has been done in measuring the nitrogen deficiencies. Such soil can be 3 m deep or 102
59 foliar biomass production of woody species (Ludwig more and are preferred for agricultural seepage. 103
60 et al., 1975; Clutter et al., 1983). Underground water is hard, but non-saline. 104
61 To fill this gap, studies are required to estimate
62 forage potential of shrubs on site. Therefore, use of 2.2. Selection of shrub species
DP 105
63 non-destructive methods might be a good alternative
64 for evaluating the production of the leaf biomass of The choice of species was made only after careful 106
65 wood producing plants (Grigal and Ohmann, 1977). consideration of their ecological importance, uses and 107
66 In the light of the aforementioned, a study was preference by the rural population. The tree species 108
67 conducted with the objective of evaluating the non- selected to determine foliar biomass and structural 109
68 destructive methods to quantify leaf biomass of 15 analysis, were: Acacia berlandieri Benth., Acacia 110
TE

69 woody plant species of Tamaulipan thornscrub. farnesiana (L.) Wild., Acacia rigidula Benth., Amyris 111
70 Another objective of this research was to determine texana (Buckl.) Wilson, Bernardia myricaefolia 112
71 whether or not the precision level of these methods Wats., Caesalpinia mexicana Gray, Celtis pallida 113
72 varied with morphological characteristics of the Torr., Cordia boissieri DC., Helietta parvifolia (Gray) 114
EC

73 species under study. Benth., Leucophyllum frutescens Johnst, Parkinsonia 115

aculeata L., Ebenopsis ebano (Benth.) Coulter, 116
Prosopis leavigata Torr. Viguiera stenoloba Blake, 117
74 2. Materials and methods and Zanthoxylum fagara (L.) Sarg (Table 1). 118
All the species used in the investigation are native 119
RR

75 2.1. Description of the study area to arid and semiarid zones. 120

76 The experimental area is situated on the plain at

77 430 m altitude in the piedmont of the Sierra Madre 3. Structural analysis 121
78 Oriental in Mexico (248270 north latitude and 998 320
79 west longitude). Originally all the area was covered by The three methods for leaf biomass estimation
CO

122
80 the typical tamaulipan thornscrub dominated by employ mathematical equations that require some 123
81 woody plants, which was mainly cleared to support relevant quantitative data related to the characteristics 124
82 cattle production and crops. Some areas with of the structure of the vegetation, such as the tree total 125
83 permanent matorral were conserved as reserves and height, tree canopy projection, number of branches 126
84 those that have been used for forestry and silvopastoral and their diameters. The data on the structural analysis 127
UN

85 vegetation dynamic and ecological studies. for this work was obtained by standard dendrometric 128

FORECO 9357 1–8

 DTD 5

R. Foroughbakhch et al. / Forest Ecology and Management xxx (2005) xxx–xxx 3

Table 1
Outstanding characteristics of 15 woody plant species selected for study
Family Species Life form and plant height Usesa
Asteraceae Viguiera stenoloba Blake Sub-small shrub (<1 m) For.
Boraginaceae Cordia boissieri A. DC. Medium tree (2–3 m) Orn., For.
Caesalpiniaceae Caesalpinia mexicana A. Gray Medium shrub (2–4 m) For., Fir., Honey
Caesalpiniaceae Parkinsonia aculeata L. Tree (3–5 m) For., Fir.
Euphorbiaceae Bernardia myricaefolia Wats. Small shrub (1–2 m) For.
Mimosaceae Acacia berlandieri Benth. Spiny shrub (3–5 m) For., Fir.
Mimosaceae Acacia farnesiana (L.) Willd. Tree (3–6 m) For., Wood, Fir.
Mimosaceae Acacia rigidula Benth. Medium shrub (2–4 m) For.

OF
Mimosaceae Ebenopsis ebano (Berl.) Barneby Tree (3–6 m) Wood, Fir., For.
Mimosaceae Prosopis laevigata (Humb. & Bonpl. Ex Willd) M.C. Johnst. Tree (3–6 m) Wood, Fir., For.
Rutaceae Amyris texana (Buckl.) Wilson Medium shrub (2–4 m) For., Med.
Rutaceae Helietta parvifolia (Gray ex Hemsl.) Benth. Tree (3–8 m) Wood, For., Med.
Rutaceae Zanthoxylum fagara (L.) Sarg. Spiny shrub (3–5 m) For., Fir., Wood
Scrophulariaceae Leucophyllum frutescens (Berl) I.M. Johnst. Small shrub (1–2 m) For., Med., Orn.

RO
Ulmaceae Celtis pallida Torr. Medium shrub (2–4 m) For., Fruit
a
For.: forage; Fir.: firewood; Orn.: ornamental; Med.: medicine.

129 methods (Mueller-Dombois and Ellenberg, 1974; was determined using standard equations of several 153
130 Dieguez et al., 2003). geometric figures based on dimensional measure- 154
ments (height and diameter of canopy). After
DP 155
obtaining the regression equation, the amount of 156
131 4. Leaf biomass estimation methods forage of other unharvested plants is estimated 157
(Pieper, 1973). 158
132 4.1. Adelaide method
4.3. Double sampling of branch method 159
133 The method consists in selecting a branch from
TE

134 each species which is taken from outside the study With this method, proposed by Reid et al. (1988) a 160
135 area. This branch is called the reference unit (Andrew regression equation of the basal diameter of the 161
136 et al., 1979, 1981; Cabral and West, 1986) and it branches on leaf biomass of the same branches is 162
137 should be representative of the form and foliar density estimated. This equation is then used to estimate the 163
EC

138 of the branches for each species under study. Then, leaf biomass of other shrubs based on diameter 164
139 using this reference unit, the number of the branch measurement of their branches. 165
140 units for each sampled shrub was estimated. The shrub Two groups of 15 individuals each, from each of the 166
141 was harvested at the end of the measurement period to 15 species, were selected in order to sample all the 167
142 determine its leaf biomass. Afterwards, the regression range of different sizes. The aforementioned estima- 168
RR

143 equation which best fits the relationship between leaf tion methods were applied to both groups of each 169
144 dry matter and the number of units was chosen to species and afterwards, all individual plants were 170
145 predict the leaf biomass as forage on site for other harvested. 171
146 individual shrubs of the same species. Leaves of each individual plant were cut (sepa- 172
rately) and then placed in an oven at a temperature of 173
147 4.2. Dimensional method 55–60 8C for a period of 48 h to obtain the dry weight.
CO

174
The individuals of one group were used to obtain 175
148 This method consists in establishing a lineal the regression line which best fitted the relationship 176
149 regression of leaf biomass and some measured between biomass and the independent variable to 177
150 morphological characteristics of shrub or tree species. predict the dry weight (leaf biomass) of the other 178
151 In this study, the independent variable used for group. Three regression models were used and the 179
UN

152 regression was the canopy volume of the plant. This selection of the best model was based on the 180

FORECO 9357 1–8

 DTD 5

4 R. Foroughbakhch et al. / Forest Ecology and Management xxx (2005) xxx–xxx

181 coefficient of determination (r2): the similarity analysis (Table 2) showed a clear 209
dominance of C. boissieri (7.9% of importance value), 210
Origin : Y ¼ bX H. parvifolia (11.5%), E. ebano (12.9%), P. leavigata 211
(9.8%), V. stenoloba (11.9%) and Z. fagara (8.9%) in 212
183
182 Linear : Y ¼ a þ bX
tamaulipan thornscrub of northeastern Mexico. 213
185
184
Quadratic : Y ¼ a þ bX þ cX 2
5.1. Regression equations for estimating 214
188
186
187
where X = independent variable and Y = dependent foliar biomass 215
189
190 variable (Zar, 1996).
The other group was processed in such a manner The foliar biomass analysis reveals that the method 216
191

OF
192 that the registered weight of its individuals were with the highest coefficient of determination (r2), was 217
compared with the estimated weights generated by the the Adelaide (Table 3) for the following species: A. 218
193
194 models and, in this way, the most precise method of rigidula (r2 = 0.98), B. myricaefolia (r2 = 0.94), C. 219

195 estimation was chosen based on the highest value of mexicana (r2 = 0.92), L. frutescens (r2 = 0.95) y Z. 220

196 the coefficient of determination and the lowest value fagara (r2 = 0.93). The Adelaide method showed the 221

RO
of the relative difference (%D) between estimated and closest relationship between leaf biomass and the 222
197
observed weight. Relative difference (percentage) was number of branch units, as compared to the 223
198
199 calculated based on ðSEX̄ =X̄Þ100. independent variables used in the two other methods. 224
Statistical analyses were computed using the SPSS These results are in agreement with those of Andrew 225
200
(v. 10.0) statistical package. The regression and et al. (1981) and Sanchez and Febles (1999), where the 226
201
202 correlation models were applied to the increment
parameters and leaf biomass (Zar, 1996).
DP
highest value of the coefficient of determination was
obtained also for the Adelaide method compared to
other estimation methods.
227
228
229
The results of the correlation coefficient analysis 230
203
5. Results and discussion and the relative differences between the predicted and 231
the measured leaf biomass (Table 4), showed that the 232
204
In order to determine the vegetative diversity and Adelaide method produced the best fit for most of the 233
TE

205
its impact on the ecosystem, it was necessary to species under study. The dimensional method had the 234
206
determine the community indices, which are used to highest precision in the case of C. pallida (r = 0.97; 235
207
establish the importance of each species within the D = 29.5), C. boissieri (r = 0.91; D = 24.6), P. 236
208
community in a quantitative manner. The results on aculeata (r = 0.91; D = 21.3), despite the fact that 237
EC

Table 2
Physical structure of some tree species in the high stratum of matorral in northeastern Mexico
Species Relative frequency (%) Relative abundance (%) Relative density (%) Importance value (%)
Acacia berlandieri 5.1 8.4 3.6 5.7
RR

Acacia farnesiana 3.5 4.7 7.2 5.1

Acacia rigidula 4.9 7.5 3.1 5.2
Amyris texana 5.6 4.8 0.8 3.7
Bernardia myricaefolia 8.9 10.1 6.5 8.5
Caesalpinia mexicana 1.5 0.9 0.1 0.8
Celtis pallida 6.9 5.0 4.4 5.4
Cordia boissieri 4.8 2.8 16.0 7.9
CO

Helietta parvifolia 14.2 8.7 11.5 11.5

Leucophyllum frutescens 6.9 8.4 1.6 5.6
Parkinsonia aculeate 1.8 2.7 9.1 4.2
Ebenopsis ebano 10.6 10.1 18.1 12.9
Prosopis leavigata 8.6 7.2 13.7 9.8
Viguiera stenoloba 9.7 21.2 4.9 11.9
UN

Zanthoxylum fagara 7.9 6.8 12.1 8.9

FORECO 9357 1–8

DTD 5

R. Foroughbakhch et al. / Forest Ecology and Management xxx (2005) xxx–xxx 5

Table 3
Regression equations using three methods of estimations of foliar biomass in 15 woody plant species in northeastern Mexico
Species Method Function r2
Acacia berlandieri A Ŷ ¼ 14:68 þ 3:31U þ 0:88ðUÞ2 0.66
Di Ŷ ¼ 213:11V 0.81
B Ŷ ¼ 5:99  1:4D þ 0:13ðDÞ2 0.94

Acacia farnesiana A Ŷ ¼ 21:26 þ 12:42U 0.68

Di Ŷ ¼ 156:21V 0.45
B Ŷ ¼ 6:81  1:24D þ 0:26ðDÞ2 0.87

Ŷ ¼ 3:50 þ 4:06U þ 1:62ðUÞ2

OF
Acacia rigidula A 0.98
Di Ŷ ¼ 18:32  191:54V þ 2578:5ðVÞ2 0.87
B Ŷ ¼ 8:87  2:19D þ 0:18ðDÞ2 0.85

Amyris texana A Ŷ ¼ 6:28U 0.71

Di Ŷ ¼ 15:52 þ 218:02V 0.78

RO
Bernardia myricaefolia A Ŷ ¼ 8:68U 0.94
Di Ŷ ¼ 27:26ŶV 0.45

Caesalpinia mexicana A Ŷ ¼ 8:36U 0.92

Di Ŷ ¼ 123:0V 0.77
B Ŷ ¼ 9:72 þ 1:58D 0.45

Celtis pallida A
Di
DP
Ŷ ¼ 28:16 þ 13:21U þ 0:22ðUÞ2
Ŷ ¼ 185:39V
0.53
0.99
B Ŷ ¼ 10:18  3:55D þ 0:37ðDÞ2 0.87

Cordia boissieri A Ŷ ¼ 12:98 þ 25:90U þ 0:42ðUÞ2 0.43

Di Ŷ ¼ 236:54V 0.83
TE
Helietta parvifolia A Ŷ ¼ 44:28 þ 36:96U 0.89
Di Ŷ ¼ 39:57V 0.28
B Ŷ ¼ 26  66 þ 4:9D  0:08ðDÞ2 0.49

Leucophyllum frutescens A Ŷ ¼ 10:14U 0.95

Di Ŷ ¼ 3:02 þ 128:94V  17:95ðVÞ2 0.78
EC

Parkinsonia aculeate A Ŷ ¼ 87:86 þ 254:90U 0.43

Di Ŷ ¼ 93:49V 0.83

Ebenopsis ebano A Ŷ ¼ 24:86 þ 59:21U 0.91

Di Ŷ ¼ 160:44V 0.67
RR

Prosopis leavigata A Ŷ ¼ 2:81 þ 23:19U þ 1:24ðUÞ2 0.88

Di Ŷ ¼ 362:47V 0.53

Viguiera stenoloba A Ŷ ¼ 14:60U 0.82

Di Ŷ ¼ 169:88V 0.75

Zanthoxylum fagara A Ŷ ¼ 55:36U 0.93

CO

B Ŷ ¼ 3:04  0:39D þ 0:10ðDÞ2 0.66

A: Adelaide method; Di: dimensional method and B: branch method. V: canopy volume of individuals of each species; U: number of units
contained in individuals of each species; D: relative difference between estimated and observed leaf biomass.
UN

FORECO 9357 1–8

DTD 5

6 R. Foroughbakhch et al. / Forest Ecology and Management xxx (2005) xxx–xxx

Table 4 Table 4 (Continued )

Correlation coefficient (r) and relative difference (D) between Species Methods
predicted and measured leaf weights using three different estimation Adelaide Dimensional Branch
methods in 15 woody plant species
Species Methods Zanthoxylum fagara
r 0.93 0.83 0.85
Adelaide Dimensional Branch
D 38.10a 45.10 41.30
Acacia berlandieri a
The lowest value of relative difference (%) indicating the most
r 0.81 0.90 0.96 precise method.
D 38.90 69.10 28.70a
Acacia farnesiana these species were morphologically the most hetero- 238

OF
r 0.82 0.67 0.93
D 48.90 41.60 27.96a
geneous species. We have confirmed that the dimen- 239
sional method provided good estimations for only a 240
Acacia rigidula
few species, mainly because the determination of a 241
r 0.99 0.81 0.91
D 43.70 49.70 31.20a specific geometric form of the individuals of each 242
species is very difficult. The branching method gave 243

RO
Amyris texana
r 0.89 0.63 0.75
good fits for A. berlandieri (r = 0.96; D = 28.7), and A. 244
D 25.30a 39.60 51.80 farnesiana (r = 0.93; D = 27.9). Therefore, our results 245
are consistent with those of Andrew et al. (1981), 246
Bernardia myricaefolia
r 0.95 0.93 0.35 Clark and Messina (1998) and Whelan (2001) who 247
D 39.20a 45.70 57.90 confirmed that the Adelaide and the dimensional 248

Caesalpinia mexicana
methods are simpler, more practical and precise in the
DP 249
r 0.90 0.91 0.73 estimating the amount of forage despite the highly 250
D 17.70a 28.70 40.40 varied external characteristics of woody plants. 251
Celtis pallida In general terms, plant architecture and morpho- 252
r 0.98 0.97 0.88 logical characteristics of woody plants play an 253
D 50.80 29.50a 65.10
Table 5
TE

Cordia boissieri
Correlations between some plant characteristics and the foliar
r 0.65 0.91 0.38
biomass from structural analysis in 15 woody shrub species
D 45.10 24.65a 72.84
Species Foliar biomass (dependent variable)
Helietta parvifolia
r 0.99 0.59 0.57 TPHa CPa BDa NB a
39.10a
EC

D 58.90 42.70 Acacia berlandieri 0.26 0.69 *

0.59 0.61*
Leucophyllum frutescens Acacia farnesiana 0.34 0.78 0.57 0.82*
r 0.94 0.66 0.63 Acacia rigidula 0.83 0.90** 0.57 0.23
D 24.20a 63.60 49.76 Amyris texana 0.29 0.69* 0.23 0.35
Bernardia myricaefolia 0.63* 0.90** 0.43 0.65*
Parkinsonia aculeate Caesalpinia mexicana 0.24 0.43 0.30 0.45
RR

r 0.65 0.91 0.52 Celtis pallida 0.38 0.81** 0.28 0.72*

D 44.25 21.34a 68.91 Cordia boissieri 0.20 0.86** 0.61 0.73*
Ebenopsis ebano Helietta parvifolia 0.59 0.56 0.02 0.58
r 0.95 0.81 0.62 Leucophyllum frutescens 0.69* 0.75* 0.16 0.43
D 23.54a 40.22 72.01 Parkinsonia aculeate 0.78* 0.89** 0.46 0.75
Ebenopsis ebano 0.92** 0.97** 0.68 0.81*
Prosopis leavigata Prosopis leavigata 0.73 0.85* 0.64 0.80*
CO

r 0.94 0.72 0.56 Viguiera stenoloba 0.54 0.77* 0.28 0.39

D 18.90a 37.80 58.72 Zanthoxylum fagara 0.90** 0.85** 0.70* 0.07
Viguiera stenoloba TPH: total plant height; CP: canopy projection; BD: branch dia-
r 0.95 0.83 0.37 meter; NB: number of branches.
a
D 22.10a 49.30 65.80 Independent variables.
*
P < 0.05.
UN

**
P < 0.01.

FORECO 9357 1–8

 DTD 5

R. Foroughbakhch et al. / Forest Ecology and Management xxx (2005) xxx–xxx 7

Table 6 These results agree with those of Melgoza and Fierro 281
List of 15 woody shrub species and their suitable methods to
(1980) who confirmed that one of the best parameters 282
determine the leaf biomass for each of them
for quantifying biomass is canopy projection and 283
Species Methods
coverage. 284
Adelaide Dimensional Branch Only A. berlandieri (r = 0.26), A. farnesiana 285
Acacia berlandieri X (r = 0.34), A. texana (r = 0.29), C. mexicana 286
Acacia farnesiana X (r = 0.24) and C. pallida (r = 0.38) showed a poor 287
Acacia rigidula X X X
correlation between the total foliar biomass of the 288
Amyris texana X X
Bernardia myricaefolia X plant and its total height. This was due to the fact that 289
Caesalpinia mexicana X their branches tend to grow towards the horizontal 290

OF
Celtis pallida X X periphery of the canopy, forming a very extended 291
Cordia boissieri X crown. With the exception of H. parvifolia (r = 0.56) 292
Helietta parvifolia X
and C. mexicana (r = 0.43) all the other species 293
Leucophyllum frutescens X
Parkinsonia aculeate X showed a high correlation coefficient (r = 0.69–0.99) 294
Ebenopsis ebano X between leaf biomass and canopy projection. 295

RO
Prosopis leavigata X Table 6 shows the most suitable methods to 296
Viguiera stenoloba X estimate the leaf biomass for each species under 297
Zanthoxylum fagara X
study. 298

254 important role in the distribution and production of

255 foliar biomass. On the bases of these criterion, the DP 299
256 Adelaide method and secondly, the dimensional 6. Conclusions
257 methods have the necessary requirements to be 300
258 applied in the estimation of foliage biomass on site The method used in this study showed close 301
259 for most of the woody plant species in northeastern relationships between foliar biomass and some of the 302
260 Mexico. Nevertheless, there always exist the problems measured characteristics of the plants. These relation- 303
261 of morphological characteristics of the plants which ships allowed us to use simple and quick procedures to 304
TE

262 make the measurements quite difficult. generate reliable and valuable information to estimate 305
leaf biomass of shrub species in northeastern Mexico. 306
263 5.2. Structural analysis The non-destructive Adelaide and dimensional meth- 307
ods were the easiest to use and, at the same time, the 308
EC

264 On the basis of this analysis we observed that A. most precise and practical methods best situated to 309
265 berlandieri, A. rigidula, B. myricaefolia, C. pallida quantify foliar biomass for most of the woody species 310
266 and Z. fagara showed the highest coefficient of under study. 311
267 variation due to the existing variability associated The shape of the plants affected considerably the 312
268 with the foliar density within the canopies of these precision of the methods used, so the methods must be 313
RR

269 species. adjusted according to the morphological character- 314

270 Regarding the canopy projection, A. rigidula istics of the species determined by their responses to 315
271 (29.6%), B. myricaefolia (20.5%), C. pallida the environment. 316
272 (20.2%), L. frutescens (23.7%) and Z. fagara For other plant species under similar study 317
273 (26.4%) showed the highest degree of variation due conditions, the application of an estimation method 318
274 to the diversity in the shape of the canopies. A high must be in concordance with the architectural and
CO

319
275 value of the coefficient of variation is an important organic structures of the particular species as a result 320
276 indicator of intra as well as interspecific plant of its ecological efficiency under the environmental 321
277 heterogeneity. conditions. 322
278 The results of the correlation analysis (Table 5) It is recommended that for each plant species in 323
279 showed that the foliar biomass in most of the species different ecological sites, an appropriate method based 324
UN

280 was correlated, logically, with the growth parameters. on the plant morphology should be applied in order to 325

FORECO 9357 1–8

 DTD 5

8 R. Foroughbakhch et al. / Forest Ecology and Management xxx (2005) xxx–xxx

325 determine the foliar biomass production more simply, Foroughbakhch, R., 1992. Establishment and growth potential of 358
fuelwood species in northeastern Mexico. Agroforestry Syst. 19, 359
326 precisely and reliably.
95–108. 360
Foroughbakhch, R., Háuad, L.A., Cespedes, A.E., Ponce, E.E., 361
González, N., 2001. Evaluation of 15 indigenous and introduced 362
327 species for reforestation and agroforestry in northeastern Mex- 363
References ico. Agroforestry Syst. 51, 213–221. 364
328 Grigal, D.F., Ohmann, L.F., 1977. Biomass estimation for some 365
329 Andrew, M.H., Noble, I.R., Lange, R.T., 1979. A non-destructive shrubs from northeastern Minnesota. USDA Forest Service, 366
330 method for estimating the weight of forage on shrubs. Aust. Washington, DC, USA, pp. 1–3. 367
331 Rangeland J. 1 (3), 225–231. Ludwig, J.A., Reynold, J.F., Whitson, P.D., 1975. Size biomass 368
332 Andrew, M.H., Noble, I.R., Lange, R.T., Johnson, A.W., 1981. The relationships of several Chihuahuan desert shrubs. Am. Midl. 369

OF
333 measurement of shrub forage weight: three methods compared. Nat. 94 (2), 451–461. 370
334 Aust. Rangeland J. 3 (1), 74–82. von Maydel, H.J., 1996. Appraisal of practices to manage woody 371
335 Battey, N.H., 2000. Aspects of seasonality. J. Exp. Bot. 51 (352), plants in semiarid environment. In: Bruns, S., Furberg, J., 372
336 1769–1780. Luukanen, O., Woods, P. (Eds.), Dryland Forestry Research. 373
337 Blair, G.J., 1990. The diversity and potential value of shrubs and tree International Foundation for Science, Stockholm, Sweden, pp. 374
338 fodder. In: Devendra, C. (Ed.), Shrubs and Tree Fodder for Farm 47–64. 375

RO
339 Animals. Proceedings of a Workshop in Denpasar, Indonesia, Melgoza, A., Fierro, L.C., 1980. Manual de Métodos de Muestreo de 376
340 International Development Research Center (IDRC), Ottawa, la Vegetación., vol. 1. Departamento de Manejo de Pastizales, 377
341 Canada, p. 276. INIP-SARH. Serie Técnico Cientı́fica, pp. 46–79. 378
342 Cabral, D.R., West, N.E., 1986. Reference unit-based estimates Mueller-Dombois, D., Ellenberg, H., 1974. Aims and Methods of 379
343 of winterfat browse weights. J. Range Manage. 39 (2), 187– Vegetation Ecology. John Wiley & Sons Inc., USA, p. 547. 380
344 189. Pieper, R.D., 1973. Técnicas de Medición para la Vegetación Herbá- 381
Clark, T., Messina, F., 1998. Foraging behavior of lacewing larvae cea y Arbustiva. Department of Animal Range and Wildlife 382
345
346
347
on plants with divergent architectures. J. Insect Behav. 11, 303–
317.
DP
Sciences. New México State University, Las Cruces, NM, USA.
Reid, N., Marroquı́n, J., Beyer-Münzel, P., 1988. Intensidad de
383
384
348 Clutter, J.L., Fortson, J.C., Pienaar, L.V., Brister, G.H., Bailey, R.L., ramoneo en especies del matorral tamaulipeco del noreste de 385
349 1983. Timber Management: A Quantitative Approach. John México. Manejo de Pastizales 1, 25–28. 386
350 Wiley and Sons, New York, USA, pp. 312–329. Reid, N., Stafford-Smith, D.M., Beyer-Münzel, P., Marroquı́n, J., 387
351 Dieguez, U., Barrio, M., Castedo, F., Ruı́z, A., Álvarez, M., Álvarez, 1990. Floristic and structural variation in the tamaulipan thorn- 388
352 J., Rojo, A., 2003. Dendrometrı́a. Ediciones Mundi-Prensa, scrub, northeastern Mexico. J. Veget. Sci. 1, 529–538. 389
TE

353 Madrid, España, p. 327. Sanchez, R., Febles, I., 1999. Behavior of grazing Holstein cows in 390
354 Eviner, V.T., 2003. Functional matrix: a conceptual framework for natural shade. Cuban J. Agric. Sci. 33, 241–246. 391
355 predicting multiple plant effects on ecosystem processes. Annu. Whelan, C., 2001. Foliage structure influences foraging of insectivor- 392
356 Rev. Ecol. Evol. Syst. 34, 455–485. ous forest birds: an experimental study. Ecology 82, 219–231. 393
357 Fahring, L., 2003. Effects of the habitat fragmentation on biodi- Zar, J.H., 1996. Biostatistical Analysis, third ed. Prentice-Hall Inc., 394
versity. Annu. Rev. Ecol. Evol. Syst. 34, 487–515. New Jersey, USA. 395
EC

358
396
RR
CO
UN

FORECO 9357 1–8