abstract - university of edinburgh · web viewtable 3. semivariogram parameter estimates for...
TRANSCRIPT
Article reference: FORECO_2017_1302To be published in: Forest Ecology and ManagementOriginal title: Aboveground carbon storage in tropical dry forest landscapes: A multi-scale analysis
Aboveground carbon storage in tropical dry forest plots
in Oaxaca, Mexico
Rogelio O. Corona-Núñeza,c,*, Julio Campob, Mathew Williamsc
a Procesos y Sistemas de Información en Geomática, SA de CV. Calle 5 Viveros de Petén, No. 18,
Col. Viveros del Valle, Tlalnepantla, Estado de México, Mexico. CP 53240;
b Instituto de Ecología, Universidad Nacional Autónoma de México, Mexico City, Mexico;
c School of GeoSciences, Global Change Research Institute, University of Edinburgh, EH9 3FF,
UK, Edinburgh, UK.; [email protected]
* Corresponding author
1234
5
6
7
8
9
10
11
12
13
14
15
16
17
ABSTRACT
Tropical forests are subject to increasing pressures due to global change. Globally, tropical dry
forests (TDFs) have been heavily impacted and these impacts have been poorly quantified. Despite
its large coverage in tropical regions, and its important influence on the global C cycle, little is
known of spatial variations in aboveground biomass (AGB) distribution of TDFs. The
understanding of TDF aboveground biomass has been biased towards the secondary forests, with
few studies of mature forests. The aim of this study is to quantify, allocate and understand the
natural factors responsible for driving the AGB distribution and, consequently, on C stocks, in
natural TDFs in Mexico. The study region represents ~14% of the total TDF area in Oaxaca.
Remote sensing time series across field sites were used to identify suitable sampling locations for
mature forests in Oaxaca. Aboveground biomass was normally distributed with a mean of 117±5
Mg ha-1 (±1 SE). Large trees (diameter at breast height, DBH ≥ 30 cm) were found at similar
frequencies in small (300-400 m2) and large (4 ha) plots. Depending on the selected allometric
equation, at least 60% of the AGB is held in trees with DBH < 28.7cm. At local scale, large trees
(DBH ≥ 30 cm) did not show spatial autocorrelation and, in the landscape, AGB showed a spatial
correlation in distances < 250 m. Because of low densities of very large trees (DBH ≥ 75 cm), the
mean AGB estimates across different allometric equations only resulted in differences of ~10%.
State factors including climate (mean annual precipitation, temperature and solar radiation) and
topography (altitude and distance to streams) modulate the TDF structure and its potential for
aboveground storage C across the landscape. Soil texture and pH were the most important soil
properties in explaining variations in AGB, with stronger effects than soil nutrients. Across
different scales of analysis, higher biomass estimates were related to water availability. This
information can support spatial estimates of biomass storage capacity for Mexican TDF, crucial for
land and C management.
2
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
Keywords:
Biomass; Landscape patterns; spatial scale; Climatic and Geophysical controls on AGC storage;
allometries, uncertainty; Mexico
3
43
44
45
1. Introduction
Forest land plays a major role in the global carbon (C) cycle, with nearly 50% of aboveground C
stored in tropical forests (Houghton, 2005; Pan et al., 2011), and significant land use and land cover
change leading to large C losses (Houghton and Nassikas, 2017). A majority of the research effort
has focused on tropical rain forests, neglecting the C storage in tropical dry forests (TDFs) (Skutsch
et al., 2009; Jaramillo et al., 2011; Gei and Powers, 2013), despite evidence indicates this
ecosystem has broad extent and appreciable C pools in vegetation and soils (Read and Lawrence,
2003; Roa-Fuente et al., 2012; Campo and Merino, 2016). TDFs accounted over 40% of all tropical
forest biome (Murphy and Lugo, 1986a; Cao et al., 2016), and important C densities in vegetation
(39 to 57 MgC ha-1) and soils (40 to 80 MgC ha-1) (Houghton and Nassikas, 2017).
Information on forest losses for TDFs are sparse and uncertain. According to Chomitz et al. (2007)
nearly 78% of the original area that was once covered by TDFs had been modified, but other
researchers suggest 48% (Miles et al., 2006; Dirzo et al., 2011). In Mexico about 25 to 36% of the
original 33.5 million ha of primary TDFs remains (INEGI, 2003a), with important extension of
secondary forest (Rzedowski, 1998; Dirzo et al., 2011; Tobón et al., 2017). This loss explains why
TDF is considered as one of the most endangered ecosystem in the tropics (Janzen, 1998; Dirzo et
al., 2011). Moreover, under the global change scenarios, land-use and land-cover and in climate
changes, intensive studies are needed for a better understanding of long-term behavior and drivers
of C sequestration in tropical forest.
Biomass and C stock estimations depends on forest inventories, which should be able to provide a
full representation of the forest type. However, there are different sources of uncertainty. Chave et
al. (2004) suggest that primary sources of uncertainty are due to tree measurement and plot size. In
this context, size and number of sample plots become fundamental issues to be considered during
4
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
the inventory. Biomass is not normally distributed in small sampling plots - rare large trees can
contribute to most of the biomass in the landscape - and it is recognized that larger plots could
reduce this bias (Keller et al., 2001; Chave et al., 2004; Chave et al., 2005). However, there is not a
clear understanding whether a small or large sampling design drive divergent conclusions regarding
the TDF structure and C stocks. For example, in the last decade, studies on TDF have been
collected over diverse sampling sizes, plots ≤ 100 m2 (Cairns et al., 2003; Gallardo-Cruz et al.,
2005; García et al., 2005), 100-400 m2 (Sagar and Singh, 2006; Bijalwan et al., 2010; Burquez and
Martínez-Yrízar, 2010) and 500 m2 (Eaton, 2005; Sagar and Singh, 2006; Bijalwan et al., 2010;
Burquez and Martínez-Yrízar, 2010; Návar, 2010). There are very few examples of plots with sizes
near 10,000 m2 or bigger (Jaramillo et al., 2003; Gasparri et al., 2010). Another source of
uncertainty in biomass estimates comes from the proper selection of the allometric equation (Chave
et al., 2004). Thus, it is important to identify the best sampling approach and allometric equations
for reducing the uncertainty in aboveground biomass (AGB) estimates and consequently on C
stocks.
Mean annual temperature (MAT) and mean annual precipitation (MAP), solar irradiation and soil
nutrient availability are recognized as primarily responsible for ecosystem development (Holmgren
et al., 1997; Vitousek, 2004; Ordoñez et al., 2009; Berdanier and Klein, 2011; Medeiros and
Drezner, 2012; Peterson, 2012). However, depending on the scale of analysis and the ecosystem
under study water availability and soil nutrients perform differently (Allen and Hoekstra, 1990;
Turner, 2005; Currie, 2011). On the one hand, at global scales, climatic variables are main factors to
explaining AGB (Becknell et al., 2012) and ecological processes such as nutrient cycling (Snyder
and Tartowski, 2006). Different authors found that MAP explains over 50% of the variation in AGB
with an inverse correlation (Brown and Lugo, 1982; Eaton and Lawrence, 2009; Becknell et al.,
2012). On the other hand, at landscape scale solar irradiation, slope and slope expose, soil texture
and terrain concavity are recognized to be the major influences on water availability for plants
5
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
(Leitner, 1987) and biomass allocation (Berdanier and Klein, 2011; Peterson, 2012). Moreover, the
availability of a suitable microclimate is critical for the response of species distribution (Bennie et
al., 2008), by driving the individual development of trees (Holmgren et al., 1997; Berdanier and
Klein, 2011). Nevertheless, precipitation regimen and soil properties are the main factors that
module structural changes in TDF biomass at regional scale (Powers et al., 2009; Medeiros and
Drezner, 2012; Roa-Fuentes et al., 2012). However, little is known about the influence of nitrogen
(N) and phosphorus (P) availabilities on aboveground C storage in mature TDFs (Gei and Powers,
2013; Campo, 2016), despite TDFs productivity could be limited by both nutrients (Campo and
Vázquez-Yanes, 2004).
This lack of integrity across scales complicates the understanding of the drivers of AGB in mature
forests, making it even more complex in a heterogeneous natural landscape. Thus, the critical
questions to be addressed by this study are, what is the typical biomass of undisturbed Mexican
TDF, and what are the main factors that control biomass accumulation, and consequently C storage,
in a natural landscape, across landscape scales? To be capable of answering these questions and in
light of current knowledge the following hypotheses were developed:
H1. The ideal minimum sampling size to reach an AGB normally distributed in TDF
ecosystems should be ≥ 2,500 m2, as suggested by Chave et al. (2004).
H2. Large trees dominate AGB storage, therefore the AGB spatial autocorrelation should
show similar ranges to large tree locations.
H3. Over different spatial scales, water availability is the major limiting factor for AGB in
mature TDFs.
2. Methods
6
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
2.1. Study region
The study region is located in Southern Oaxaca, Mexico (Fig. 1) with a total area of 215,687 ha
within the boundaries of five municipalities (San Miguel del Puerto Sta. Ma. Colotepec, Sta. Ma.
Huatulco, Sta. Ma. Ozolotepec, and Sta. Ma. Tonameca). The area represents 14.4% of the total
TDF surface in Oaxaca (INEGI, 2008). The region shows an important extension of TDF in mature
stage (Durán et al., 2007; Corona, 2012) with deforestation rates below the national and state
averages (Velazquez et al., 2003; Corona et al., 2016). The landscape consists of low hills, followed
by pre-mountain landforms with low dissection (altitudinal range is 0 to 1,200 m asl). The climate is
hot and sub-humid, with a small thermal oscillation during the year (García, 2004). Mean annual
temperature ranges from 19 to 33 °C, and MAP is ~1,600 mm, most of which (>75% of the total)
falls between June and September (rainy season) (Hijmans et al., 2005; SMN, 2014). The
predominant soil is Regosol with a lithic phase, followed by Cambisol and Lithosol (Corona, 2009;
INEGI, 2009). The predominant vegetation in the area is the TDF (Corona, 2012; Corona et al.,
2016).
7
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
Fig. 1. The left-hand panel shows the sampling location of 83 plots of tropical dry forests over an
altitude grid map. The top right-hand panel highlights in red the study site in the State of Oaxaca,
Mexico. On the bottom right-hand panel is the location of the State of Oaxaca in the country. The
icons refers the location of clusters of plots according to each sampling strategy. The location of the
plots is approximate to the central coordinate of the clusters. The number of plots per cluster are
presented in color coding.
2.2. Data collection
To capture the spatial heterogeneity of AGB densities in mature TDFs over the landscape, field
campaigns were designed to record the landscape variability by sampling the most common
biophysical regions and the rare elements of the landscape. We established 60plots (10-m x 30-m).
The information was complimented with the National Forest Inventory (NFI) (CONAFOR, 2007;
CONAFOR, 2012), in which 128 sampling plots (10-m x 40-m) were systematically allocated. The
campaign plots (n=188) were filtered to select forest plots of probably at least 50 years old (our
field campaigns n=42 and the NFI campaigns n=41). Complementary, two plots of 4 ha each were
collected at two different altitudes (50 and 250 m asl) within a distance of ~7 km between them
(Figure 1). Our field campaigns were carried out during summer 2012 and spring 2013, and for the
NFI in the year 2007. In spring 2014, about 70% of the NFI plots were visited to identify the site
conditions.
Finally, the sample plots followed the next filtering process: (1) with an absence of human tracks,
invasive species, logging or any kind of wood extraction, burned areas and evidence of cattle inside
or around the plot; (2) located over 2 km from agricultural fields; and (3) with no experience human
disturbance in the past. We analyzed remote sensing time series from aerial photos (1985 and
1995), and Google Earth-Digital Globe imagery (2004 to 2014). Aerial photographs were
8
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
complemented with Landsat 8 (2013 and 2014 years). By visually analyzing aerial photographs
(Shoshany, 2000; Shoshany, 2002) since 1985 they had to show mature forest surrounded by a
natural forest matrix. The maximum distance between two plots was 80 km with an average of
26.6±2.1 km (mean±1 SE). About 80% of the plots have at least another neighbor plot in a distance
~320 m and ~50% of the plots have a neighbor plot at less than 100 m.
2.2.1. Field data collection
We considered stems ≥1cm in diameter at breast height (DBH). The NFI plots measured trees ≥7.5
cm in DBH (DBH and height). Complementarily, the NFI registered and classified small trees
(DBH <7.5 cm and over 0.25 m in height) in a sub-plot of 3.54-m x 3.54-m. Each tree was
registered in height classes (CONAFOR, 2007). We develop a linear regression to estimate DBH
for trees with <7.5 cm based on their height. In each 4 ha plot, trees with DBH ≥30 cm were
sampled and geotagged (n=365) and the DBH and height recorded. We carried out a pseudo-
sampling (multiple virtual plots) to identify the minimum sampling plot for a normal distribution.
For each sampling design the mean, median, and normality test were calculated. In all cases (our
field campaign and NFI) tree species were classified in situ by botanist and/or foresters, and
corroborated in herbariums (CONAFOR, 2007). Our samples were evaluated in the MEXU
herbarium, while the NFI samples were tested in different herbariums (CONAFOR, 2007). The
species classification was based on the sampled leaves, fruits and/or flowers, and images of the
bark.
We collected litter (fine and coarse) in each of our campaign plots (10-m x 30-m) at the end of the
dry season (May), when litter standing crop in TDFs reaches its maximum (Martínez-Yrízar, 1995;
Campo et al., 2001). The litter samples consisted of all dead plant material lying on the forest floor,
including the freshly fallen litter and the more finely decomposed litter fraction. Fine litter ≤2 cm in
9
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
diameter, leaves, and twigs, was collected in two sub-samplings. In plots of 10-m x 10-m three sub-
samplings of 30-cm x 30-cm were collected (n=120). We aggregated the information to represent
the fine litter in plots of 10-m x 30-m (n=40). We analyzed coarse litter ≥2 cm in diameter, and
woody debris <2 cm in diameter. Coarse litter and woody debris were collected in 19 sub-samples
(2-m x 30-m) within our campaign plots. Manually each woody sample was classified. We excluded
the section of the woody sample when it fell outside of the sub-sample plot. All samples were dried
in the oven to a constant weight at 70°C and weighed on a digital scale with a SE of 1 mg.
We investigated the top soil (0-5 cm in depth) by composing a mixture of six sub-samples (n=38) in
plots of 10-m x 10-m, turning into 19 mean values for 10-m x 30-m plots (May). The top soil is
where most of the microbial activity, and soil nutrients are concentrated in TDF ecosystems
(Campo et al., 1998; Roa-Fuentes et al., 2012).
2.2.2. Biomass estimation
To estimate the AGB, we used the DBH, height, wood density and a set of seven allometric
equations (Table 1). Due to the elevated number of endemic and rare species (Trejo and Dirzo,
2002; Castillo-Campos et al., 2008; Dirzo et al., 2011) we used a stand-level wood density (WD)
(Baker et al., 2004). Wood density was constructed based on a collection of 28 plots during the dry
season (May). The wood cores measured 0.515 cm in diameter with variable length (15.5±0.1 cm).
All samples were dried in the oven to a constant weight at 70°C and weighed on a digital scale with
a SE of 1 mg.
Table 1. Allometric equations used in this study
Reference Ecosystem Location MAP Equation for aboveground biomass
10
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
(mm) (in kg per tree)
Brown et al. (1989) eq 1
Brown et al. (1989) eq 2
Brown (1997)
Chave et al. (2005) eq 1
Chave et al. (2005) eq 2
Martínez-Yrízar et al.
(1992)
Návar (2009)
TMF
TDF
TDF
TDF
TDF
TDF
TDF
Pantropic
Pantropic
Pantropic
Pantropic
Pantropic
Mexico
Mexico
1500-4000
<1500
>900
<1500
<1500
<900
<700
e^(-2.409+0.9522*ln(DBH^2*H*D))
34.4703 - (8.0671*DBH) + (0.6589*DBH^2)
e^(-1.996+2.32*ln(DBH))
0.112*(D * DBH^2*H)^0.916
D*e^(-0.667+(1.784*ln(DBH)) +
(0.207*ln(DBH)^2)-(0.0281*ln(DBH)^3))
10^(-0.535 + LOG(BA))
0.0841*(DBH^2.41)
________________________________________________________________________________
TMF: tropical moist forests; TDF: tropical dry forests; MAP: mean annual precipitation.
e: Euler’s constant; DBH: diameter at breast height (cm); H: height (m); D: wood density (g cm-3); BA: basal
area (cm2) per stem.
The set of allometric equations include two for Mexican TDFs (Martínez-Yrízar et al., 1992; Návar,
2009), and five for the Pantropic calibrated under different precipitation regimes (Brown, 1997;
Brown et al., 1989; Chave et al., 2005) (Table 1). We filtered the allometric equations after
processing all the plots. We eliminated the allometries that produced the highest and the lowest
AGB estimates and keept the remnant five to calculate the mean AGB, and its standard error (±1
SE) as a measure of uncertainty. All the estimates were converted into Mg ha-1. Tree diametric
classes were evaluated for each field plot and the cumulative AGB was estimated at stand level.
2.2.3. Characterization of the landscape and digital data acquisition
11
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
All topographic variables were built in the QGIS (QGIS-2.6.0, 2014) platform. From digital contour
lines every 10 m (INEGI, 2003b) we constructed a Digital Elevation Model with a resolution of 20-
m x 20-m per pixel. From the DEM we derived altitude, slope, terrain curvature, topographic
indexes and hydrological network. We further calculated the distance to the hydrological network
(streams) and distance to the coast.
We built the terrain shape or exposure from estimating the difference between mean altitude
neighborhood pixels and the elevation at the central pixel (Zimmermann, 2000). The compound
topographic index (CTI) summarizes soil moisture by considering water flow direction and
accumulation (Beven, 1977; Burrough et al., 1998) and higher scores are associated with moist sites
(Spadavecchia et al., 2008).
We used the potential mean solar irradiance (PmI) in the dry season (October to May), period when
water availability is limited (Giraldo and Holbrook, 2011; Maass and Burgos, 2011). The PmI
indicates the maximum heat that potentially strikes, and it works as a limiting factor on soil
moisture and plant productivity (Wright and van Schaik, 1994). We downscaled the monthly and
hourly solar irradiance reported for the region (IG-UNAM, 2017) by calculating the proportion of
the potential incoming shortwave radiation. The proportion of the potential incoming shortwave
radiation was estimated by changing the solar geometry (between 11:00 and 16:00 h, local time),
the azimuth and altitude depending on the month and hour (Kumar et al., 1997; NOAA, 2015), and
its figures fall between 0 (absent solar transmission) and 100% (complete solar transmission).
Climatic variables have been suggested that correlate strongly to AGB in TDFs (Murphy and Lugo,
1986b; Martínez-Yrízar et al., 1992; Read and Lawrence, 2003). We analyzed the importance of
climatic variables by taking the available information from WorldClim (version 1, 1960-1990)
(Hijmans et al., 2005). The variables included were: (1) MAT (BIO1); (2) maximum temperature in
12
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
the warmest month (BIO5); (3) minimum temperature in the coldest month (BIO6); (4) temperature
seasonality (BIO4); (5) MAP (BIO12); and (6) total precipitation in the warmest quarter (BIO18).
Because it is recognized that the dry season module the TDF function other climatic variables were
considered such as (7) maximum, (8) minimum and (9) mean temperatures in both dry and rainy
seasons; (10) the seasonal climate amplitude, calculated as the maximum-to-minimum temperature
ratio; and (11) the MAP to MAT ratio.
We used time series of leaf area index (LAI) to evaluate the effect of the leaf span expression (Ryser
and Urbas, 2000) and as indicator of primary productivity in TDFs (Gower et al., 1999). Time
series of LAI were acquired by the Terra/MODIS satellite (Moderate Resolution Imaging
Spectroradiometer), in an eight-day mean composite from 2002 to 2014.
2.2.4. Laboratory analyses of wood, litter and mineral soil
For determination of C concentration in wood, we collected six mixed samples from different stems
in each plot. Coarse litter, and woody debris ≥2 cm in diameter, were weighed in the field on a
digital scale with an error of ±1 g. For dry weight estimation, about 30% of the biomass per plot
was dried in the oven to a constant weight. All dried subsamples were mixed and sent for chemical
analysis for C and nitrogen (N) concentration. For each composite soil (0-5 cm in depth) organic C,
total N, and available phosphorus (P) concentrations, pH (H2O), bulk density and texture were
determined. Phosphorus extraction technique was selected based on soil pH. Olsen P-extraction was
implemented when soil reaction was ≥7; otherwise Bray P-extraction was used (Bray and Kurtz,
1945; Olsen et al., 1954). Wood, litter and SOC samples were analyzed in an automated C-analyzer
(SHIMADZU 5005A). The concentration of N was determined from acid digestion in H2SO4
concentrated using a NP elemental analyzer (Technicon Autoanalyzer II). Concentrations of organic
13
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
C, total N and available P were transformed into area units (Mg ha-1, in the case of C and N; kg ha-1,
in the case of P) according to the bulk densities for each plot.
2.3. Analysis of data
Litter and soil data from samples collected on a fine scale (10-m x10-m) were treated independently
for local spatial autocorrelation analysis and grouped in 10-m x 30-m plots for landscape analysis.
All descriptive statistical analyses were performed in the R software version x64 2.15.1 (R-Core-
Team, 2014). We use the Mann-Whitney-Wilcox test to establish that two samples came from the
same population, and a Jarque Bera test (JB) to evaluate the normality of the data. Then, we
developed a linear regression (ordinary-least-squares) and boxplots to examine the relationships
between each measure of environmental heterogeneity. To meet the assumptions for the linear
regressions, and to avoid any further transformations, we tested for normality and when needed, we
excluded the outliers from the analysis. We choose the standard error as a measure of uncertainty.
Spatial dependence of a location on neighboring sites was measured by spatial autocorrelation
(SAC) (nugget, partial sill and range). The SAC was estimated by using the library geoR (Ribeiro
and Diggle, 2001). The semivariograms were fitted to a spherical model as inversely proportional to
approximate variance (weighted Cressie). This approach was chosen because there is a small to
moderate sample size (Cressie, 1985; 1993). Spatial dependence in the data was estimated from the
nugget-to-sill ratio (Rossi et al., 2009).
3. Results
3.1. Community structure, tree size distribution and basal area
14
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
In this study 58 genera and 229 species of trees were recorded (supplementary material). We were
unable to classify 87 potential species (i.e., around 24% of the total species). About 50% of the total
species were represented by six families (Fabaceae, Euphorbiaceae, Bignoniaceae, Clusiaceae,
Burseraceae and Bombacaceae); 23% of these families were represented by five genera (Tabebuia,
Calophyllum, Bursera, Hura and Acacia) and five species (T. chrysantha, C. brasiliense, H.
polyandra, C. vitifolium and Guazuma ulmifolia).
Tree size distribution follows an inverse J-shape. On average there are ~6,450 trees ha-1 (DBH ≥1
cm) (Fig. 2). About 90% of trees had a DBH <10 cm and trees with a DBH ≥30 cm represented
<1% of the total stems (between 45 and up to 56 stems per hectare). The two sampling approaches
(300-400 m2 and 4 ha plots) for tree densities with DBH ≥30 cm, did not differed statistically
(p=0.13).
The basal area showed a median of 27.2 m2 ha-1 and a mean of 27.6±1.0 m2 ha-1 with a normal
distribution when excluding six outliers (p=0.34, n=77; Fig. 3a). There are not statistical differences
between the NFI and our sampling strategies (p=0.55). Wood density showed a normal distribution
(p=0.78, n=121) with a mean of 0.51±0.01 g cm-3 and a range between 0.09 and 0.93 g cm-3. There
was no relationship between the tree DBH and the wood density (r2 =0.001, p=0.74, n=121).
15
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
Fig. 2. Tree size distribution for two sampling designs (small and large sampling plots) in Oaxaca,
Mexico. Black bars apply to small sampling plots (300-400 m2) and grey bars apply to large
sampling plots (200-m x 200-m).
3.2. Aboveground biomass and carbon stocks estimates and its uncertainty
The allometries developed by Martínez-Yrízar et al. (1992) and Chave et al. (2005; eq 2) produced
the lowest and the highest AGB estimates, respectively. The former equation returns the lowest
AGB estimates per tree with DBH >20 cm, resulting also in the lowest AGB estimates per plot. The
latter equation derived in an exponential AGB estimation per tree in relation to the DBH promoting
the highest AGB estimates per plot.
Large (4 ha) and small (300-400 m2) sampling approaches resulted in AGB being normally
distributed. From a pseudo-sampling of large plots (4 ha) it can be inferred that the minimum
16
324
325
326
327
328
329
330
331
332
333
334
335
336
337
sampling plot size for a normal distribution in AGB for trees with DBH ≥30 cm is 2,500 m2
(p=0.24, n=32). The median and mean AGB derived from small sampling plots was 107.9 Mg ha-1
and 117.5±5.0 Mg ha-1, respectively, with a normal distribution when four outliers were excluded
(Table 2, Fig. 3b; p=0.18, n=79). The outlier plots resulted from the clustering of three or more
large trees (DBH≥35 cm with a height between 10-15 m, n=3 plots) or two colossal trees (DBH≥45
cm with height over 15 m, n =1 plot). There are not statistical differences between the NFI and our
sampling strategy (p=0.57). Aboveground biomass correlated strongly to the basal area (Fig. 4a; r2
=0.88, p<0.001, n=83).
Fig. 3. Frequency graphs for the basal area (a), and aboveground biomass (b) in tropical dry forests
at Oaxaca, Mexico. Data for basal area come from 300-400m2 plots. Red bars refer to the outlier
observations (n=6 for basal area and n=4 for AGB) and solid line refers to the expected normal
distribution basal area (mean=27.6 m2 ha-1 and STD=8.9 m2 ha-1) and aboveground biomass (mean=
117.5 Mg ha-1 and STD= 45.0 Mg ha-1). **STD stands for standard deviation.
Mean AGB estimates varied depending on the selected equation. The proportional distribution of
the AGB in diametric classes differs depending on the selected allometric equation. For example, to
measure at least 60% of the AGB in average the DBH was restricted to trees under 28.7±0.8 cm
(25.3 cm using the equation developed by Martínez-Yrízar et al. (1992), and up to 31.5 cm when
using the equation developed by Návar (2009)) (Table 2, Figs. 5a and 5b). However, to explain 80%
17
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
of the AGB the range increases. The DBH could be as low as 40.4 cm and up to 63.0 cm (Table 2
and Fig. 5b) with an average of 50.7±3.3 cm and a median of 48.4 cm.
Table 2. Mean aboveground biomass for each allometric equation and minimum diameter at breast
height to reach 60 and 80% of the total aboveground biomass in tropical dry forests at Oaxaca,
Mexico. The mean and the standard error (±1SE) were calculated by the integration of the sample
plots excluding the outliers (*) and derived from the seven allometric equations (**).
Reference of equation used AGB (Mg ha-1) DBH60%AGB (cm) DBH80%AGB (cm)
Brown et al. (1989) eq 1 105.8 29.1 49.5
Brown et al. (1989) eq 2 124.1 27.5 48.4
Brown (1997) 124.8 30.3 60.5
Chave et al. (2005) eq 1 98.4 28.1 45.3
Chave et al. (2005) eq 2 145.4 29.0 47.5
Martínez-Yrízar et al. (1992) 77.6 25.3 40.4
Návar (2009) 103.6 31.5 63.0
Mean±1 SE 117.5± 5* 28.7±0.8** 50.7±3.3**
n=79 n=83 n=83
Mean AGB only relates to number of trees with DHB ≥30 cm (Figs. 4b and 4c; r2 =0.46, p<0.001,
n=79). However, trees <40 cm in DBH represent a large proportion of the observed AGB. For
example, all the AGB within plots <50 Mg ha-1 was restricted in trees with DBH <30 cm, and in
sites with AGB densities <100 Mg ha-1 ~95% of it is represented by trees with DBH <40 cm; and up
to 70% of the AGB when the sites showed an AGB density ≥100 Mg ha-1.
18
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
Fig. 4. Relationships of aboveground biomass with basal area (a), tree density (b) and tree density
(stems with a diameter at breast height, DBH≥30 cm) (c), and relationships of leaf area index (LAI)
with the tree density (d) and tree density (stems with DBH≥30 cm) (e) in tropical dry forests at
Oaxaca, Mexico. When the regression was significant (p<0.05), the equation was included. The
outliers are reported in red circles and they were excluded from the statistical analysis.
When contrasting different allometric equations, the uncertainty in biomass estimates per tree
increased with DBH, mainly occurring in the largest class (>75 cm in DBH). However, because of
their low densities large trees (>75 cm in DBH) accounted for 7.7-12.4 Mg ha-1, which represents
~10% of the total AGB (Fig. 5b). Additionally, the primary source of uncertainty was related to
more abundant trees (stem with DBH <35 cm; they include about 66.3% of the AGB estimates),
with a range of 65.1-81.9 Mg ha-1 (Figs. 5a and 5b).
19
373
374
375
376
377
378
379
380
381
382
383
384
385
Fig. 5. Cumulative aboveground biomass in Mg ha-1 (a) and in percentage (b) per diametric class
for tropical dry forests at Oaxaca, Mexico. In all cases the seven allometric equations were used.
Bars refer to the upper and lower quartile; solid black line refers to the median, white dots refer to
outliers while the greatest and the lowest values, excluding outliers, are expressed with whiskers.
20
386
387
388
389
390
391
Wood and litter C concentration (47.2±0.4% and 36.3±1.3% , respectively) showed normal
distribution (p>0.2). The total (live and dead) aboveground C (AGC) pool was of 63.8±2.6 MgC ha-
1. The majority of C density (87%) (55.5±2.4 MgC ha-1) was in the live fraction, while 4% (2.5±0.01
MgC ha-1) was in the dead fraction of forest floor litter.
3.3. Properties of the mineral soil
Soils textures were characterized by a sandy loam, with a few examples of sandy clay loam. The
mean composition of the topsoil (0-5 cm) was 62.6±2.1%, 19.7±1.2%, and 17.7±1.1% for sand, silt
and clay, respectively (n=19). The pH values were classified as acidic to neutral (6.6±0.2) (JB
p=0.45, n=19; Wilcox test p=0.97).
3.4. Spatial autocorrelation analysis
The spatial dependence among variables and scales showed that the landscape is heterogeneous
(Figs. 6 and 7). At the fine scale, individual large sized-trees (DBH ≥30 cm, n=365) basal area and
AGB, and soil properties (n=38) showed high semivariance and did not exhibit spatial
autocorrelation. At a landscape scale, the range of spatial autocorrelation showed two contrasting
patterns (Table 3). On the one hand, basal area (n=83), AGB (n=83) and SOC (n=19) depicted a
range of <250 m, with a moderate spatial dependence. The pool of C in fine litter (n=40) had a
range of ~1.5 km with a strong spatial dependence (Table 3). On the other hand, soils N and P
stocks (n=19), were driven by nugget effect, suggesting there is no spatial autocorrelation (Table 3
and supplementary material).
21
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
Table 3. Semivariogram parameter estimates for measured variables in tropical dry forests at
Oaxaca, Mexico. Data were determined at a coarse spatial scale (10-m x 30-m) derived from a fitted
spherical model.
Range (m) Sill Nugget Nugget/Sill
ratio
Basal area 87 120 42 0.35M
Live aboveground
biomass
229 1,431 598 0.42M
Fine litter 1,506 4.7 0.4 0.08S
Soil organic carbon 248 27.2 8.8 0.32M
Altitude 1,638 916 66 0.07S
Distance to streams 292 1,687 53 0.03S
3.5. The landscape and biophysical influences on the aboveground biomass in tropical dry
forests
There is a strong correlation between the AGB and variables related to water availability, and less
correlation with the abundance of soil nutrients (Figs. 6 and 8). The mean annual LAI was the
strongest factor (r2=0.19, p <0.001, n=79) and it explained the spatial variation in AGB densities
(Fig. 6a). The LAI time series suggested that differences across AGB occurred in the dry season
(Fig. 7). In the period of the leaf senescence low to high AGB (<200 Mg ha-1) showed the lowest
mean LAI values (1.62±0.1 m2 m-2, n=74), whereas the largest AGB (≥200 Mg ha-1) reached greater
LAI (3.7±0.02 m2 m-2, n=9) (p<0.001). In contrast, when the renewal of leaves dominates the
canopy (rainy season) the LAI is more homogeneous within the sites, and did not varied between
low and medium AGB groups (<100 Mg ha-1) (4.5±0.1 m2 m-2, n=35) and for high and very high
22
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
AGB group (≥100 Mg ha-1) (4.9±0.1 m2 m-2, n=48) (p>0.05). The total number of trees did not
correlate with LAI (Fig. 4d). The best correlation was identified for the number of trees with
DBH≥30 cm (r2=0.05, p=0.040, n=79; Fig. 4e).
The AGB rose with the increase in altitude (r2=0.12, p<0.001, n=79; Fig. 6b). At higher altitudes
(>500 m asl) AGB was 185±20 Mg ha-1 (n=9), while at lower altitudes (<200 m asl) the AGB
decreased to 116±10 Mg ha-1 (n=49) (p=0.002). The AGB decreased with higher solar irradiance
(r2=0.09, p<0.01, n=79; Fig. 6c); places with low potential solar irradiance during the dry season
(<440 W m-2) reached the highest AGB (263±53 Mg ha-1, n=2), and when solar irradiance was ≥500
W m-2 the AGB densities decreased (121±6 Mg ha-1, n=64) (p=0.029). The terrain orientation
influenced the AGB accumulation showing larger values in West and North aspects (175±14 Mg ha-
1, n=19), than in East and South (123±8 Mg ha-1, n=56) (Fig. 6d; p<0.001). Also, the terrain
curvature influenced the AGB in the driest sites where they receive less than 850 mm of total
precipitation in the year; and the AGB was greater in concave terrains (142±11 Mg ha-1, n=10) than
in convex (88±7 Mg ha-1, n=5) (Fig. 6e; p<0.001). The AGB increased with closeness to streams in
the first 140 m of distance (r2=0.07, p<0.05, n=79; Fig. 6f). Compound topographic index (soil
wetness and nutrients accumulation) influenced the AGB accumulation (from AGB of 106±11 Mg
ha-1 (n=12) to 157±23 Mg ha-1 (n=4), for sites with values <5.0 in comparison to sites with values
7.5-10; p=0.03). Variables such as slope and the distance to the coast did not show significant
relationships with the AGB (p>0.05).
The influence of climatic variables showed that AGB decreased while increasing the MAT
(r2=0.13, p<0.001, n=79; Fig. 6g). The AGB in sites with MAT <25°C (210±33 Mg ha-1, n=5)
closely doubled than in the warmer sites (>25 °C) (110±6 Mg ha-1, n=78) (p=0.003). Moreover,
higher AGB related to the increment of the seasonal climate amplitude during the dry season
(r2=0.09, p<0.001, n=79; Fig. 6h). For example, with a climate amplitude of <1.9ºC the mean AGB
23
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
was smaller (116±8 Mg ha-1, n=57) than in those with larger amplitudes (153±12 Mg ha-1, n=26)
(p<0.001).
Fig. 6. Significant biophysical variables tested against aboveground biomass in tropical dry forests
at Oaxaca, Mexico. (a) Mean annual leaf area index (LAI) comes from a MODIS/Terra; (b) altitude,
(c) solar irradiance, (d) terrain orientation, (e) terrain curvature and (f) distances to streams are
derived from a digital elevation model; (g) mean annual temperature, (h) ratio of extreme
temperatures during dry season, (i) mean annual precipitation and (j) mean annual precipitation to
mean annual temperature ratio, come from WorldClim (Hijmans et al., 2005). Linear regressions
come from ordinary least squares. The regressions report adjusted r2 and their significance p-value
(p). The outliers are reported in red circles and they were excluded from the regression analysis.
24
458
459
460
461
462
463
464
465
466
467
468
469
As expected for these ecosystems, AGB increased with MAP (r2=0.14, p<0.001, n=79; Fig. 6i).
The AGB in drier sites (MAP<1,000 mm) were lower (115±7 Mg ha-1, n=62) than in wetter sites
(MAP ≥1000 mm) (160±12 Mg ha-1, n=21) (p<0.001). We found that AGB increased with the
increment in the ratio MAP:MAT (Fig. 6j) (from 118±6 Mg ha-1 (n=73) in sites with ratios <50 mm
°C-1 to 185±24 Mg ha-1 (n=10) in sites with ratios ≥50 mm °C-1) (p=0.002).
a)
Fig. 7. Time series of mean leaf area index (2002-2014) and aboveground biomass in tropical dry
forests at Oaxaca, Mexico. Leaf area index time series are expressed in m2 m-2 per day.
Soil texture was an important factor in explaining the AGB. The AGB declined with the increase in
clay and silt contents in the soil (r2=0.38, p=0.003, n=19 and r2=0.18, p=0.041, n=19, respectively;
Figs. 8a and 8b); and enlarged with increase in sand content (r2=0.30, p=0.009, n=19; Fig. 8c).
Moreover, it was found an effect of soil pH on the AGB in which the AGB increased from 57±6 Mg
ha-1 (n=4) in acidic soils (pH<5.5) to 120±7.2 Mg ha-1 (n=9) in those slightly acid to neutral soils
(pH=5.5-7.3) (p=0.025). Besides, AGB in TDFs in sites with slightly alkaline soils (pH=7.3-7.8)
(113±24 Mg ha-1, n=6; Fig. 8d) did not differ from the previous two groups (p>0.05).
25
0.00
1.00
2.00
3.00
4.00
5.00
6.00
7.00
0 100 200 300
Leaf
Are
a In
dex
Day of year
0-50 Mg Ha50-100 Mg ha100-200 Mg ha>200 Mg ha
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
Fig. 8. Variations in aboveground biomass in the tropical dry forests at Oaxaca, Mexico, in function
of the top soil (0-5 cm depth) clay (a), silt (b) and sand (c) contents, and pH (d).
The SOC and total N concentrations in the top soil were 4.28±0.4% (n=19), and 0.33±0.03%
(n=19), and did not show significant relationships to AGB (p>0.5). The soil organic matter quality
in forests was high as indicated by a low soil C:N ratio averaging 11.4±0.6 (n=19), and it did not
show significant relationship to AGB (p=0.33). Phosphorus availability in the soil was low
4.31±0.44 µg g-1 (JB test, p=0.83, n=19).
The SOC, N and P pools in the top 5 cm of the soil was not related to AGB (p=0.98, p=0.28, n=19,
and p=0.88, n=19, respectively). Soil organic C pool has a mean stock of 21.6±1.2 MgC ha-1, the N
pool showed a mean of 1.53±0.1 MgN ha-1 and the P pool exhibit a mean of 35.0±0.5 kgP ha-1.
Finally, the AGB did not show significant relationships to soil C:N or N:P ratios (p>0.05, n=19).
26
493
494
495
496
497
498
499
500
501
502
503
504
505
506
4. Discussion
The tree-size distribution suggests that the observations come from a mature TDF with continuous
regeneration and constant replacement (Hall and Bawa, 1993; Lykke, 1998; El-Sheick, 2013). Tree
density is similar to others TDF's in Latin America (3,270-7,770 trees per hectare; Trejo, 1998;
Jaramillo et al., 2003; Gallardo-Cruz et al., 2005; Marín et al., 2005; Powers et al., 2009), with
similar proportional tree-size-distribution to other tropical forests (Chave et al., 2003; Jaramillo et
al., 2003; Chave et al., 2004; DeWalt and Chave, 2004; Marín et al., 2005).
4.1. Carbon storage in aboveground biomass
Mean wood density and C concentrations are comparable to those other dry forests. The wood
density falls in between the reported range (0.59 to 0.73 g cm3: Brown et al., 1989; Brown, 1997;
Fearnside, 1997; Jaramillo et al., 2003; Read and Lawrence, 2003; Baker et al., 2004; Ribeiro et al.,
2011). Some authors have suggested that local wood C concentration can reduce the uncertainties in
5-7% (Martin and Thomas, 2011; Saner et al., 2012; Thomas and Malczewski, 2007) and up to 10%
(Elias and Potvin, 2003). In our study we account for 5.5%.
Basal area distribution across diametric classes is similar to others’ TDFs where most of it (67%)
was measured in trees with DBH<30 cm (Gillespie et al., 2000; Jaramillo et al., 2003; Marín et al.,
2005). However, trees with a DBH≥50 cm represent almost double the amount of basal area
reported by Marín et al. (2005) and Jaramillo et al. (2003). The AGB fell within the reported range
(28-292 Mg ha-1: Brown and Lugo, 1982; Murphy and Lugo, 1986a; Murphy and Lugo, 1986b;
Brown and Lugo, 1990; Martínez-Yrízar et al., 1992; Roa-Fuentes et al., 2012). Finally, total dry
mass of the standing fine litter is higher than other mature TDF in Mexico (Campo et al., 2014), and
27
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
its C concentration is relatively smaller (Anaya et al., 2007; Firdaus et al., 2010; Petit-Aldana et al.,
2011).
Forest inventories and allometric equations may drive significant changes to understand factors that
controls forest structure and AGB densities. On the one hand, small stems are considered important
in young forests, and most of the studies focused on classes with DBH≥10cm (Brown, 1997; Keller
et al., 2001). On the other hand, allometric equations produce different AGB estimates per tree.
These can lead to differences in the understanding of the community assembly and AGB
estimations at the landscape scales, similar observation to Corona et al. (2017). At tree level,
different allometric equations have shown that large trees were the most uncertain AGB estimates
(Baker et al., 2004; Chave et al., 2004). However, large trees (DBH≥60 cm) are scarce in the TDFs
which has limited impact on mean AGB estimates at the landscape scale (20%). Small to medium-
sized trees (DBH<30 cm) are important because they contribute to the greatest biomass (>60% of
the total) (Baker et al., 2004) and up to ~80% (Jaramillo et al., 2003; Chaturvedi et al., 2011), and
in its uncertainty (Keller et al., 2001). This has led to the conclusion that there are little differences
across various allometric equations to estimate AGB densities for small to medium-sized trees.
Finally, the under sampling of small-sized trees (DBH<10 cm) represent an important loss in AGB
estimates (Keller et al., 2001); in our study this loss was around 15%.
We identified that in the landscape large stems are not the major contributors to AGB estimates.
Because most of the TDF biomass is concentrated in trees with DBH<30 cm. To reduce the
uncertainty on AGB estimations at landscape scale, forest inventories should implement a full
sampling collection, and allometric equations should be improved for small to medium-sized trees
(<50 cm in DBH). Moreover, our data indicate that collecting trees with a DBH≥1 cm in plots of
300-400 m2 can produce an AGB normally distributed. However, when there is the interest of
28
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
understanding the influence of trees DBH≥30 cm a minimum sampling of 50-m x 50-m is necessary
to reach the normality. Therefore, based on these results our first hypothesis (H1) turned to be false.
Large trees spatial distribution did not show similar ranges of spatial autocorrelation to AGB
densities, in contrast to our hypothesis (H2). Ours results suggest that while higher AGB densities
correlate with large trees, the spatial autocorrelation is driven by other environmental factors, such
topography, soil properties and climate. The location of trees depends on a multi-variable inter-
correlation over different spatial scales (climate, topography, species competence, light or soil
nutrients) (Hubbell, 1980; Gonzalez and Zak, 1994; Condit et al., 2000). In our study, we found that
soil properties and tree location (AGB per tree DBH ≥30 cm) resulted from random phenomena.
This result implies that at tree level the community was spatially heterogeneous and the tree-spatial
distribution did not show an assembly of patches of similar environmental characteristics (Hubbell,
1979; Levings, 1983; Murphy and Lugo, 1986b; Condit et al., 2000). However, at the landscape
scale, AGB was influenced by water availability showing a similar range of spatial autocorrelation
to distance to streams. Based on these results we conclude that there is no spatial autocorrelation in
AGB beyond the local scale, because AGB and soil’ properties in the landscape are heterogeneous.
4.2. Mature forest biomass is weakly linked to soil properties
Soil properties were similar to other TDFs; for example, pH (Lugo and Murphy, 1986; García-Oliva
et al., 1994; Wick et al., 2000; García-Oliva et al., 2006), clay content (García-Oliva et al., 1994)
and, N and P concentrations (Lugo and Murphy, 1986; Campo et al., 1998). However, in contrast to
Jaramillo et al. (2011) we identified a higher C concentration for the topsoil (1.63-3.05%) and in the
upper limit of the N concentration (0.12-0.39%) for another TDF in Mexico. These may be the
result of the sampling collection that took place during the dry season when all nutrients accumulate
in the top soil (Campo et al, 1998; Saynes et al., 2005; Jaramillo et al., 2011). The P concentration
29
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
suggests a limitation of the Oaxacan’ TDF function. The C:N ratio was smaller than the reported
mean of 20:1 for a mature TDF in Mexico (Saynes et al., 2005) but close to the global scale ratios
(14:1) (Cleveland and Liptzin, 2007).
This study suggests that mature TDF is not as highly nutrient demanding as a secondary forest (see
Christensen and Peet, 1984; de Castilho et al., 2006; Mirmanto et al., 1999; and Wright et al., 2011)
and there is no evidence that make us believe that soil C, N, P, C:N and N:P play an important role
in explaining AGB in vegetation (Mirmanto et al., 1999; Campo and Vázquez-Yanes, 2004; Elser
et al., 2007; LeBauer and Treseder, 2008; Slik et al., 2010; Wright et al., 2011; Alvarez-Clare et al.,
2013, Campo, 2016). However, we found that the most conclusive soil properties to explain AGB
was pH and soil texture. The pH constrained the plant productivity by decreasing soil nutrients
availability, mainly of P, and possibly bacterial activity (Salcedo et al., 1997; Gallardo and
Gonzáles, 2004), and neutral to slightly alkaline soils pH's are just the right condition to promote C
sequestration in wood.
There are limited examples to explain how soil nutrients drives AGB accumulation and C storage in
mature TDF's and the framework to understand the influence of nutrient availability on AGB is
complex and it is inconclusive (Wall et al., 2011). Wood debris is a major component of litter input
to soils in TDFs, and its decomposition rates varies within the ecosystem by different processes
(Gonzales and Seastedt, 2000 and 2001; Torres and Gonzales, 2005; Eaton and Lawrance, 2006;
Bejarano et al., 2014) with impacts on the C sequestration and nutrient cycling in the soil (Sight,
1991; Höfer et al., 2001, Decaëns et al., 2004; Barrios, 2007). Future studies should experiment on
mature TDF's, and to build a multivariable integration of different factors to evaluate the co-
localization of the biological processes and their influence to reproduce AGB accumulation and
ecosystem productivity in mature TDF's.
30
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
4.3. Water availability is the main driver of carbon storage
At different spatial scales, water availability was an important predictor of the aboveground C
storage in Oaxacan TDFs, as occurs in other arid (Snyder and Tartowski, 2006) and semi-arid
regions (Jaramillo et al., 2011; Hulshof et al., 2013). At the local scale, soil texture actively
mediates water availability (Sperry et al., 1998; Sperry and Hacke, 2002). In particular, TDFs
developed on soils with sandy loam textures have good drainage, good water-holding capacity and
the right amount of movement of air in the soil, helping the development of plants during the rainy
season, and to overcome the dry season when water uptake is hydraulically limited (Hultine et al.,
2005). Trees in these textures are highly sensitive to small precipitation pulses (Campo et al., 1998),
with effects on soil C mineralization (Saynes et al., 2005), and C sequestration in biomass
(Fravolini et al., 2005). Geophysical variables also played an important role in explaining water
availability at landscape scale. For example, at higher altitudes less potential evaporation happened
due to a reduction in temperature and increase in precipitation amount (SMN, 2014). This turns in
greater AGC stocks at higher altitudes, as a result of an increase of water availability and not
because of higher restrictions for anthropogenic pressures, as could be suggested from disturbed
forests (de Castilho et al., 2006; Alves et al., 2010; Woollen et al., 2012). Some other topographical
variables correlated indirectly to bring shelter to plants against water stress, such as terrain
curvature (Cusack et al., 1997; Gessler et al., 2000), slope expose (Galicia et al., 1999; Bijalwan,
2012), and distance to streams promoting accumulation of AGC stocks; this can be related to
differences in root architecture and water table in upper soil depths (Canadell et al., 1996; Rango et
al., 2006).
At the landscape scale, climatic factors were related to AGC storage. On the one hand, AGC storage
decreased with increase in MAT and in the range of variations in intra-seasonal temperatures. These
relationships suggest that in sites where temperature does not vary significantly during the rainy
31
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
season plants tend to be more water stressed than in those areas with more variability, among which
are great differences during day and night when in the latter, plants re-hydrate themselves (Reich
and Borchert, 1984). On the other hand, AGC storage is higher with the increase in the amount of
precipitation and in the ratio MAP:MAT, as expected for a water-limited ecosystem (Murphy and
Lugo, 1986b; Martínez-Yrízar et al., 1992; Jaramillo et al., 2003; Read and Lawrence, 2003). Water
stress is a key factor that drives AGC and soil C storage in TDFs due to its effect in nutrient
availability for ecosystem function, by restrictions in soil nutrient mineralization, mobility and plant
uptake (see Campo et al., 1998; Hulshof et al., 2013).
Finally, mean and seasonal LAI were capable of capturing the seasonal water availability by leaf
expression and senescence (see Yingduan et al., 2013; Iio et al., 2014). Higher mean LAI and the
senescence over the year were related with water availability (less solar radiation, changes in
precipitation regime and reduction of MAT), and showed higher AGC storage. A high LAI favor C
uptake by photosynthesis in areas where direct environmental controls (water, temperature and soil
nutrients) are not limiting factors (Fatichi et al., 2014). Therefore, our study suggests that in mature
forest with limited seasonality in precipitation, the rainfall variability promotes an increment of the
growing period (see also Kenzo et al., 2010).
4.4. A multi-scale conceptual integration
Over different spatial scales, water availability turned to be the major limiting factor for AGC
storage at the local and landscape scales, as we hypothesized (see H3). However, in our study was
possible to see that AGB estimates >300 Mg ha-1 are difficult to understand by a single factor
analysis. In those sites at least two very large trees with DBH≥45 cm were recorded. The limitation
in reproducing very high AGB estimates could be the result of the spatial resolution of our data sets
32
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
(Castilho et al., 2006), or the need of multivariable analysis (Corona et al., 2017). For example,
some authors argue that large-sized tree allocation is the result of biotic interactions which modify
the competitive ability of plants (Yeaton et al., 1977; Hubbell, 1979 and 1980; Gonzalez and Zak,
1994; Condit et al., 2000; Lucero et al., 2006). While there are contrasting results to understand the
influence of local topography, soil and air temperature, which creates different types of light
microhabitats (Korner, 2007), these elements not always are conclusive to understand patchy
distribution of perennial plants (Arriaga et al., 1993). In any case, such information is difficult to
acquire on a landscape scale with such a degree of detail, and the most conclusive studies were done
at continental or global scales (Slik et al., 2013), where climatic variables explain patterns in AGB
and density of large-sized trees.
4.5. Forest management implications
This study has shown that missing small trees may drive important consequences in the
underestimation of large C stocks and sequestration. Particularly, TDFs are dominated by thinner
stems than forests in the humid counterpart. This information takes importance when accounting
carbon stocks for REDD+ projects, especially when there is a need to estimate C emissions due to
deforestation and/or degradation.
Many regions that encompasses TDF's are predicted to experience more frequent and extended
period of droughts, rise of temperatures and reduction of precipitation (Meir and Pennington, 2011).
However, it is not entirely understood what will be the effects of climate change on C storage
(Campo and Merino, 2016), complementarily to the impacts of land-use and land cover change due
to the increase of the extension of agricultural practices to overcome the reduction of production
yields (Mendoza, 2015).
33
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
5. Conclusion
For a proper understanding of C storage in TDFs of Oaxaca we evaluated the climatic and
geophysical controls on AGB. From a finer scale, it was possible to identify that the main source of
uncertainty in estimations of aboveground biomass is related to small-medium sized trees (DBH≤30
cm). State factors (sensu Chapin et al., 2011) as climate (precipitation, temperature and solar
radiation), topography (altitude and distance to streams), and soil resources (pH and soil texture)
constraints the ecosystem structure and its potential for storage C in TDFs of Oaxaca.
Tropical dry forests are highly heterogeneous not only in landscape structure but also in tree
species, resulting in variations of C storage in aboveground biomass. By excluding the outliers, the
mean AGB estimates followed a normal distribution and fell between the reported ranges for similar
vegetation type and rainfall, with no statistical differences between the sampling designs. The
community structure was mainly driven by small trees, which account for much of the above-
ground C estimates, contrasting with results reported for other tropical forests. In a mature TDF,
water availability plays the major role in explaining above-ground C in a local, sub-regional and
landscape scale. We consider that there are other underlying controls that could not be captured,
thus misleading the above-ground C estimates, mainly where large-sized trees were present. Soil
nutrients play an important role in regeneration processes, but how they link into above-ground C
estimates in a mature forest is not well documented. Therefore, future studies should look into how
fine-scale patch dynamics may be coupled on broader scales to make a better representation of
above-ground C and its potential implications for climate change risks. Moreover, more research is
necessary to understand how nutrients are distributed and how they may correlate to environmental
34
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
factors and to above-ground C in mature forest, rather than focus the studies under regeneration
processes, which may influence the understanding of the ecosystem dynamics.
Acknowledgements
This research was supported by Procesos y Sistemas de Información en Geomática. We want
to thank the CONACYT for the scholarship 187019. We appreciate the collaboration of the MsC
Jorge Calónico Soto for doing all the botanical specimens collection and species classification
during the fieldwork campaigns and for his fruitful comments. Also, we thank to the MEXU-
UNAM herbarium for giving access to their specimens collection. Finally, we want to thank the
anonymous reviewer for its constructive comments, which helped us to improve the manuscript.
References
Allen, T.F.J., Hoekstra, T.W., 1990. The confusion between scale-defined levels and conventional
levels of organization in ecology. J. Veg. Sci. 1, 5-12.
Alvarez-Clare, S., Mack, M.C., Brooks, M., 2013. A direct test of nitrogen and phosphorus
limitation to net primary productivity in a lowland tropical wet forest. Ecology 94, 1540-
1551.
Alves, L.F., Vieira, S.A., Scaranello, M.A., Camargo, P.B., Santos, F.A.M., Joly, C.A., Martinelli,
L.A., 2010. Forest structure and live aboveground biomass variation along an elevational
gradient of tropical Atlantic moist forest (Brazil). For. Ecol. Manage. 260, 679-691.
Anaya, C.A., García-Oliva, F., Jaramillo, V.J., 2007. Rainfall and labile carbon availability control
litter nitrogen dynamics in a tropical dry forest. Oecologia 150, 602-610.
Arriaga, L., Maya, Y., Diaz, S., Cancino, J., 1993. Association between cacti and nurse perennials
in a heterogeneous tropical dry forest in northwestern Mexico. J. Veg. Sci. 4, 349-356.
Baker, T.R., Phillips, O.L., Malhi, Y., Almeida, S., Arroyo, L., Di Fiore, A., Erwin, T., Killeen,
T.J., Laurance, S.G., Laurance, W.F., Lewis, S.L., Lloyd, J., Monteagudo, A., Neill, D.A.,
35
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
Patiño, S., Pitman, N.C.A., M. Silva, J.N., Vásquez Martínez, R., 2004. Variation in wood
density determines spatial patterns in Amazonian forest biomass. Glob. Change Biol. 10,
545-562.
Barrios, E. 2007. Soil biota, ecosystem services and land productivity. Ecol. Econom. 64, 269-285.
Becknell, J.M., Kissing Kucek, L., Powers, J.S., 2012. Aboveground biomass in mature and
secondary seasonally dry tropical forests: A literature review and global synthesis. For. Ecol.
Manage. 276, 88-95.
Bejarano, M., Crosby, M.M., Parra, V., Etchevers, J.D., Campo, J., 2014. Precipitation regime and
nitrogen addition effects on leaf litter decomposition in tropical dry forests Biotropica. 46,
415-424.
Bennie, J., Huntley, B., Wiltshire, A., Hill, M.O., Baxter, R., 2008. Slope, aspect and climate:
Spatially explicit and implicit models of topographic microclimate in chalk grassland. Ecol.
Model. 216, 47-59.
Berdanier, A.B., Klein, J.A., 2011. Growing season length and soil moisture interactively constrain
high elevation aboveground net primary production. Ecosystems. 14, 963-974.
Beven, K.J., 1977. Distributed Hydrological Modelling: Applications of the TOPMODEL Concept.
Wiley & Sons, Chichester, 356 pp.
Bijalwan, A., 2012. Land Use and Vegetation Analysis of Dry Tropical Forest Using Remote
Sensing & GIS. Academic Publishing, UK, 188 pp.
Bijalwan, A., Swamy, S., Sharma, C., Sharma, N., Tiwari, A., 2010. Land-use, biomass and carbon
estimation in dry tropical forest of Chhattisgarh region in India using satellite remote sensing
and GIS. J. For. Res. 21, 161-170.
Bray, R., Kurtz, L., 1945. Determination of total, organic and available forms of phosphorus in soil.
Soil Sci. 59, 39-45.
Brown, S., 1997. Estimating Biomass and Biomass Change of Tropical Forests: A Primer. FAO
Forestry Paper FAO - Food and Agriculture Organization of the United Nations, Rome.
36
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
Brown, S., Gillespie, A.J.R., Lugo, A.E., 1989. Biomass estimation methods for tropical forests
with applications to forest inventory data. For. Sci. 35, 881-902.
Brown, S., Lugo, A.E., 1982. The storage and production of organic matter in tropical forests and
their role in the global carbon cycle. Biotropica. 14, 161-187.
Brown, S., Lugo, A.E., 1990. Tropical secondary forests. Journal of Tropical Ecology. 6, 1-32.
Burquez, A., Martínez-Yrizar, A., 2010. Accuracy and bias on the estimation of aboveground
biomass in the woody vegetation of the Sonoran Desert. Botany. 89, 625-633.
Burrough, P.A., McDonnell, R., Burrough, P.A., 1998. Principles of Geographical Information
Systems. Oxford University Press, Oxford, 352 pp.
Cairns, M., Olmsted, I., Granados, J., Argaez, J., 2003. Composition and aboveground tree biomass
of a dry semi-evergreen forest on Mexico’s Yucatan Peninsula. For. Ecol. Manage. 186, 125-
132.
Campo, J. 2016. Shift from ecosystem P to N limitation at precipitation gradient in dry tropical
forests at Yucatan, Mexico. Environ. Res. Lett. 11, DOI:1088/1748-9326/11/9/095006.
Campo, J., Gallardo, J.F., Hernández, G., 2014. Leaf and litter nitrogen and phosphorus in three
forests with low P supply. Eur. J. For. Res. 133, 121-129.
Campo, J., Jaramillo, V., and Maass, J.M., 1998. Pulses of soil phosphorus availability in a Mexican
tropical dry forest: effects of seasonality and level of wetting. Oecologia 115, 167-172.
Campo, J., Maass, M., Jaramillo, V., Martínez-Yrízar, A., Sarukhán, J., 2001. Phosphorus cycling
in a Mexican tropical dry forest ecosystem. Biogeochemistry. 53, 161-179.
Campo, J., Merino A. 2016. Variations in soil carbon sequestration and their determinants along a
precipitation gradient in seasonally dry tropical forests. Glob. Change Biol. 22,1942-1956.
Campo, J., Vázquez-Yanes, C. 2004. Effects of nutrient limitation on aboveground carbon
dynamics during tropical dry forest regeneration in Yucatán, Mexico. Ecosystems. 7, 311-
319.
37
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
Canadell, J., Jackson, R.B., Ehleringer, J.B., Mooney, H.A., Sala, O., Schulze, E.D. 1996.
Maximum rooting depth of vegetation types at the global scale. Oecologia. 108, 583-595.
Cao, S., Sánchez-Azofeifa, G.A., Duran, S.M., Calvo-Rodriguez, S., 2016. Estimation of
aboveground net primary productivity in secondary tropical dry forests using the Carnegie–
Ames–Stanford approach (CASA) model. Environ. Res. Lett. 11, 075004, doi:10.1088/1748-
9326/11/7/075004
Castillo-Campos, G., Halffter, G., Moreno, C.E., 2008. Primary and secondary vegetation patches
as contributors to floristic diversity in a tropical deciduous forest landscape. Biodiv. Conserv.
17, 1701-1714.
CCSP, 2003. Strategic Plan for the U.S. Climate Change Science Program. A Report by the Climate
Change Science Program and the Subcommittee on Global Change Research. In: Office,
C.C.S.P. (Ed.). U.S. Climate Change Science Program, Washington, DC, USA, p. 211.
Chapin, F.S., Matson, P.A., Vitousek, P.M., 2011. Principles of Terrestrial Ecosystem Ecology.
Springer, New York, 529 pp.
Chaturvedi, R.K., Raghubanshi, A.S., Singh, J.S., 2011. Carbon density and accumulation in woody
species of tropical dry forest in India. For. Ecol. Manage. 262, 1576-1588.
Chave, J., Andalo, C., Brown, S., Cairns, M.A., Chambers, J.Q., Eamus, D., Fölster, H., Fromard,
F., Higuchi, N., Kira, T., Lescure, J.P., Nelson, B.W., Ogawa, H., Puig, H., Riéra, B.,
Yamakura, T., 2005. Tree allometry and improved estimation of carbon stocks and balance in
tropical forests. Oecologia. 145, 87-99.
Chave, J., Condit, R., Aguilar, S., Hernandez, A., Lao, S., Perez, R., 2004. Error propagation and
scaling for tropical forest biomass estimates. Phil. Trans. R. Soc. Lond. B 359, 409–420.
Chave, J., Condit, R., Lao, S., Caspersen, J.P., Foster, R.B., Hubbell, S.P., 2003. Spatial and
temporal variation of biomass in a tropical forest: results from a large census plot in Panama.
J. Ecol. 91, 240-252.
38
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
Chomitz, K.M., Buys, P., De Luca, G., Thomas, T.S., Wertz-Kanounnikoff, S., 2007. ¿Realidades
antagónicas? Expansión agrícola, reducción de la pobreza y medio ambiente en los bosques
tropicales. Mayol Ediciones S.A., Mexico City.
Christensen, N.L., Peet, R.K., 1984. Convergence during secondary forest sucession. J. Ecol. 72,
25-36.
Cleveland, C., Liptzin, D., 2007. C:N:P stoichiometry in soil: is there a “Redfield ratio” for the
microbial biomass? Biogeochemistry. 85, 235-252.
CONAFOR, 2007. Manual y procedimientos para el muestreo de campo del Inventario Nacional
Forestal y de Suelos 2004-2009. In: Forestal, C.N. (Ed.). CONAFOR, Zapopan, Jalisco,
Mexico.
CONAFOR, 2012. Inventario Nacional Forestal y de Suelos. Informe 2004-2009. In: Geomática,
G.d.I.F.y. (Ed.). Coordinación General de Planeación e Información, Zapopan, Jalisco,
Mexico.
Condit, R., Ashton, P.S., Baker, P., Bunyavejchewin, S., Gunatilleke, S., Gunatilleke, N., Hubbell,
S.P., Foster, R.B., Itoh, A., LaFrankie, J.V., Lee, H.S., Losos, E., Manokaran, N., Sukumar,
R., Yamakura, T., 2000. Spatial patterns in the distribution of tropical tree species. Science,
288, 1414-1418.
Corona, R., 2009. Programa de Manejo Forestal Sostenible y Aseguramiento de los Servicios
Ambientales en Bahías de Huatulco, Oaxaca. FONATUR-PSIG, Mexico City.
Corona, R., 2012. Conductores de la deforestación: Estudio de Caso en el Bosque Tropical
Caducifolio en Oaxaca. Editorial Académica Española, 160 pp.
Corona, R., Galicia, L., Palacio, J., Bürgi, M., Hersperger, A.M., 2016. Local deforestation patterns
and their driving forces of tropical dry forest in two municipalities in Southern Oaxaca,
Mexico (1985-2006). Investigaciones Geográficas, Boletín del Instituto de Geografía,
UNAM. 91, 86-104, dx.doi.org/10.14350/rig.50918.
39
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
Corona-Núñez, R.O., Mendoza-Ponce, A., López-Martínez, R., 2017. Model selection changes the
spatial heterogeneity and total potential carbon in a tropical dry forest. Forest Ecology and
Management, 405:69-80 doi:https://doi.org/10.1016/j.foreco.2017.09.018.
Cressie, N.A.C., 1985. Fitting variogram models by weighted least squares. J. Inter. Assoc.
Mathematical Geol. 17, 563-586.
Cressie, N.A.C., 1993. Statistics for Spatial Data, Revised Edition. Wiley, New York.
Currie, W.S., 2011. Units of nature or processes across scales? The ecosystem concept at age 75.
New Phytol. 190, 21-34.
Cusack, G.A., Hutchinson, M.P., Kalma, J.D., 1997. Relating biomass and land surface reflectance
to primary terrain attributes in a small catchment. In: Zerger, A., Argent, R.M. (Eds.),
MODSIM 1997. International Congress on Modelling and Simulation. Modelling and
Simulation Society of Australia and New Zealand, University of Tasmania, Hobart.
de Castilho, C.V., Magnusson, W.E., de Araújo, R.N.O., Luizão, R.C.C., Luizão, F.J., Lima, A.P.,
Higuchi, N., 2006. Variation in aboveground tree live biomass in a central Amazonian forest:
Effects of soil and topography. For. Ecol. Manage. 234, 85-96.
Decaëns, T., Jiménez, J.J., Barros, E., Chauvel, A., Blanchart, E., Fragoso, C., Lavelle, P., 2004.
Soil macrofaunal composition in permanent pastures derived from tropical forests or savanna.
Agric. Ecosys. Environ. 103, 301-312.
DeWalt, S.J., Chave, J., 2004. Structure and biomass of four lowland Neotropical forests.
Biotropica. 36, 7-19.
Dirzo, R., Young, H., Mooney, H., Ceballos, G., 2011. Seasonally Dry Tropical Forest. Ecology
and Conservation. Island Press, Washington.
Durán, E., Merino, G.L., Velázquez, A., López, F., Larrazábal, A., Medina, C. 2007. Análisis del
cambio en la cobertura de vegetación y usos del suelo en Oaxaca. Paper presented at the II
Simposio Biodiversidad, Oaxaca, Mexico, April 2007.
40
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
Eaton, J.M., 2005. Woody Debris and the Carbon Budget of Secondary Forests in the Southern
Yucatán Peninsular Region. Thesis in Masters of Science University of Virginia, Virginia,
USA.
Eaton, J.M., Lawrence, D., 2009. Loss of carbon sequestration potential after several decades of
shifting cultivation in the southern Yucatan. For. Ecol. Manage. 258, 949-958.
Elias, M., Potvin, C., 2003. Assessing inter- and intra-specific variation in trunk carbon
concentration for 32 neotropical tree species. Can. J. For. Res. 33, 1039-1045.
Elser, J.J., Bracken, M.E.S., Cleland, E.E., Gruner, D.S., Harpole, W.S., Hillebrand, H., Ngai, J.T.,
Seabloom, E.W., Shurin, J.B., Smith, J.E., 2007. Global analysis of nitrogen and phosphorus
limitation of primary producers in freshwater, marine and terrestrial ecosystems. Ecol. Lett.
10, 1135-1142.
El-Sheick, M.A., 2013. Population structure of woody plants in the arid cloud forests of Dhofar,
southern Oman. Acta Bot. Croat. 72, 97-111.
Fatichi, S., Leuzinger, S., Körner, C. 2014. Moving beyond photosynthesis: from carbon source to
sinl-driven vegetation modeling. New Phytol. 201, 1086-1095.
Fearnside, P.M., 1997. Wood density for estimating forest biomass in Brazilian Amazonia. For.
Ecol. Manage. 90, 59-87.
Firdaus, M.S., Hanif, A.H.M., Safiee, A.S., Ismail, M.R., 2010. Carbon sequestration potential in
soil and biomass of Jatropha curcas. World Congress of Soil Science, Soil Solutions for a
Changing World, Brisbane, Australia, pp. 62-65.
Fravolini, A., Hultine, K.R., Bugnoli, E., Gazal, R., English, N.B., Williams, D.G., 2005.
Precipitation pulse use by an invasive woody legume: the role of soil texture and pulse size.
Oecologia. 144, 618-627.
Galicia, L., López-Blanco, J., Zarco-Arista, A.E., Filips, V., García-Oliva, F., 1999. The
relationship between solar radiation interception and soil water content in a tropical
deciduous forest in Mexico. Catena. 36, 153-164.
41
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
Gallardo, F.F., Gonzáles, M.I., 2004. Sequestration of carbon in Spanish deciduous oak forest. Adv.
Geo. Ecol. 37, 341-351.
Gallardo-Cruz, J.A., Meave, J.A., Pérez-García, E.A., 2005. Estructura, composición y diversidad
de la selva baja caducifolia del Cerro Verde, Nizanda (Oaxaca), México. Bol. Soc. Bot.
Mexico. 76, 19-35.
García, E., 2004. Modificaciones al Sistema de Clasificación Climatica de Köppen. Instituto de
Geografía - UNAM, Mexico City.
García, R., Mendoza, I., Galicia, L., 2005. Evaluation of tropical deciduous forest landscapes, lower
Papagayo river basin (Guerrero), Mexico. Inv. Geogr. 56, 77-100.
García-Oliva, F., Casar, I., Morales, P., Maass, J.M., 1994. Forest-to-pasture conversion influences
on soil organic carbon dynamics in a tropical deciduous forest. Oecologia. 99, 392-396.
García-Oliva, F., Gallardo, J.F., Montaño, N.M., Islas, P., 2006. Soil carbon and nitrogen dynamics
followed by a forest-to-pasture conversion in western Mexico. Agrofor. Syst. 66, 93-100.
Gasparri, N., Parmuchi, M., Bono, J., Karszenbaum, H., Montenegro, C., 2010. Assessing multi-
temporal Landsat 7 ETM+ images for estimating above-ground biomass in subtropical dry
forests of Argentina. J. Arid Environ. 74, 1262-1270.
Gei, M.G., Powers, J.S., 2013. Nutrient cycling in tropical dry forests. Tropical Dry Forests in the
Americas. Ecology, Conservation, and Management. Sánchez-Azofeifa, A., Powers, J.S.,
Fernandes, G.W., Quesada, M. (Eds.), CRC Press, Boca Raton. pp. 141-155.
Gessler, P.E., Chadwick, O.A., Chamran, F., Althouse, L., Holmes, K., 2000. Modeling soil–
landscape and ecosystem properties using terrain attributes. Soil Sci. Soc. Amer. J. 64, 2046-
2056.
Gillespie, T.W., Grijalva, A., Farris, C.N., 2000. Diversity, composition, and structure of tropical
dry forests in Central America. Plant Ecol. 147, 37-47.
Giraldo, J.P., Holbrook, N.M. 2011. Physiological mechanisms underlying the seasonality of leaf
senescence and renewal in seasonally dry tropical forests. Seasonally Dry Tropical Forests.
42
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
Ecology and Conservation, Dirzo, R., Young, H.S., Mooney, H.A., Ceballos, G. (Eds.),
Island Press, Washington. pp 129-40.
Gonzáles, G.R., Seastedt, T.R., 2000. Comparison of the abundance and composition of litter fauna
in tropical subalpine forests. Pedobiologia. 44, 549-456.
Gonzales, G.R., Ley, R., Schmidt, S.K., Zou, X., Seastedt, T.R., 2001 Soil ecological integrations:
Comparisons between tropical and subalpine forests. Oecologia. 128, 549-556.
Gonzalez, O.J., Zak, D.R., 1994. Geostatistical analysis of soil properties in a secondary tropical
dry forest, St. Lucia, West Indies. Plant Soil. 163, 45-54.
Gower, S.T., Kucharik, C.J., Norman, J.M., 1999. Direct and indirect eestimation of leaf area index,
fAPAR, and net primary production of terrestrial ecosystems. Remote Sens. Environ. 70, 29-
51.
Hall, P., Bawa, K., 1993. Methods to assess the impact of extraction of non-timber tropical forest
products on plant populations. Economic Bot. 47, 234-247.
Hijmans, R.J., Cameron, S.E., Parra, J.L., Jones, P.G., Jarvis, A., 2005. Very high resolution
interpolated climate surfaces for global land areas. Int. J. Climatol. 25, 1965-1978.
Höfer, H., Hanagarth, W., Garcia, M., Martinus, C., Franklin, E., Rombke, J., Beck, L., 2001.
Structure and function of soil fauna communities in Amazonia anthropogenic and natural
ecosystems. Eur. J. Soil Biol. 37, 229-235.
Holmgren, M., Scheffer, M., Huston, M., 1997. The interplay of facilitation and competition in
plant communities. Ecology. 78, 1966-1975.
Houghton, R.A., 2005. Aboveground forest biomass and the global carbon balance. Glob. Change
Biol. 11, 945-958.
Houghton, R.A., Nassikas, A.A., 2017. Global and regional fluxes of carbon from land use and land
cover change 1850–2015. Global Biogeochem. Cycl. 31, 456-472.
Hubbell, S.P., 1979. Tree dispersion, abundance, and diversity in a tropical dry forest. Science. 203,
1299-1309.
43
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
Hubbell, S.P., 1980. Seed predation and the coexistence of tree species in tropical forests. Oikos.
35, 214-229.
Hulshof, M.C., Martínez-Yrízar, A., Burquez, A., Boyle, B., Enquist, B.J., 2013. Plant functional
trait variation in tropical dry forests. Ecology, Conservation, and Management. Sánchez-
Azofeifa, A., Powers, J.S., Fernandes, G.W., Quesada, M. (Eds.), CRC Press, Boca Raton.
Press, pp. 129-140.
Hultine, K.R., Koepke, D.F., Pockman, W.T., Fravolini, A., Sperry, J.S., Williams, D.G., 2005.
Influence of soil texture on hydraulic properties and water relations of a dominant warm-
desert phreatophyte. Tree Physiol. 26, 313-323.
IG-UNAM (Instituto de Geofísica, Universidad Nacional Autónoma de México). 2017. Solar
radiation, (In: http://www.geofisica.unam.mx/radiacion_solar/datos.html).
INEGI. 2008. Conjunto de Datos Vectoriales de Uso del suelo y Vegetación, escala 1:250,000 (serie
IV). Instituto Nacional de Estadística, Geografía e Informática, Aguascalientes, Mexico.
Iio, A., Hikosaka, K., Anten, N.P.R., Nakagawa, Y., Ito, A., 2014. Global dependence of field-
observed leaf area index in woody species on climate: a systematic review. Global Ecol.
Biogeogr. 23, 274-285.
INEGI, 2003a. Conjunto de datos vectoriales de la carta de vegetacion primaria 1:1,000,000. In:
Informática, I.N.d.E.G.e. (Ed.). INEGI, Aguascalientes.
INEGI, 2003b. Digital topographic vectors (1995, 1999, 2000, 2001, 2002 and 2003). In:
Informática, I.N.d.E.G.e. (Ed.). INEGI, Aguascalientes.
INEGI, 2009. Prontuario de Información Geográfica municipal de los Estados Unidos Mexicanos.
Santa María Huatulco, Oaxaca. INEGI, Aguascalientes.
Janzen, D.H., 1998. Tropical dry forests: the most endangered major tropical ecosystem. In: Wilson,
E.O. (Ed.), Biodiversity. Natural Academy, Washington DC, USA, pp. 130-137.
Jaramillo, V., Martínez-Yrízar, A., Sanford, R.L., 2011. Primary productivity and biogeochemistry
of seasonally dry tropical forests. In: Dirzo, R., Young, H., Mooney, H., Ceballos, G. (Eds.),
44
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
Seasonally Dry Tropical Forests: Ecology and Conservation. Island Press, Washington, pp.
109-128.
Jaramillo, V.J., Kauffman, J.B., Rentería-Rodríguez, L., Cummings, D.L., Ellingson, L.J., 2003.
Biomass, carbon, and nitrogen pools in Mexican tropical dry forest landscapes. Ecosystems.
6, 609-629.
Keller, M., Palace, M., Hurtt, G., 2001. Biomass estimation in the Tapajos National Forest, Brazil:
Examination of sampling and allometric uncertainties. For. Ecol. Manage. 154, 371-382.
Kenzo, T., Ichie, T., Hattori, D., Jawa Kendawang, J., Sakurai, K., Ninomiya, I., 2010. Changes in
above- and belowground biomass in early successional tropical secondary forests after
shifting cultivation in Sarawak, Malaysia. For. Ecol. Manage. 260, 875-882.
Korner, C., 2007. The use of “altitude” in ecological research. Trend Ecol. Evol. 22, 569–574.
Kumar, L., Skidmore, A.K., Knowles, E., 1997. Modelling topographic variation in solar radiation
in a GIS environment. Inter. J. Geogr. Inform. Sci. 11, 475-497.
LeBauer, D.S., Treseder, K.K., 2008. Nitrogen limitation of net primary productivity in terrestrial
ecosystems is globally distributed. Ecology. 89, 371-379.
Leitner, L.A., 1987. Plant cmmunities of a large arroyo at Punta Cirio, Sonora. Southwestern Natur.
32, 21-28.
Levings, S.C., 1983. Seasonal, annual, and among-site variation in the ground ant community of a
deciduous tropical forest: Some causes of patchy species distributions. Ecological Monogr.
53, 435-455.
Lucero, M.E., Barrow, J.R., Osuna, P., Reyes, I., 2006. Plant–fungal interactions in arid and semi-
arid ecosystems: Large-scale impacts from microscale processes. J. Arid Environ. 65, 276-
284.
Lugo, A.E., Murphy, P.G., 1986. Nutrient dynamics of a Puerto Rican subtropical dry forest. J.
Trop. Ecol. 2, 55-72.
45
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
Lykke, A.M., 1998. Assessment of species composition change in savanna vegetation by means of
woody plant's size class distributions and local information. Biodiv. Conserv. Biol. 7, 1261-
1275.
Maass, J.M., Burgos, A. 2011. Water dynamics at the ecosystem level in seasonally dry tropical
forests. Seasonally Dry Tropical Forests. Ecology and Conservation, In: Dirzo, R., Young,
H., Mooney, H., Ceballos, G. (Eds.), Seasonally Dry Tropical Forest: Ecology and
Conservation. Island Press, Washington. pp 141-56.
Marín, G.C., Nygård, R., Rivas, B.G., Oden, P.C., 2005. Stand dynamics and basal area change in a
tropical dry forest reserve in Nicaragua. For. Ecol. Manage. 208, 63-75.
Martin, A.R., Thomas, S.C., 2011. A reassessment of carbon content in tropical trees. PLoS One 6,
8, e23533:1–e23533:9. (https://doi.org/10.1371/journal.pone.0023533).
Martínez-Yrízar, A., 1995. Biomass distribution and primary productivity of tropical dry forests. In:
Bullock, S.H., Mooney, H.A., Medina, E. (Eds.), Seasonally Dry Tropical Forests.
Cambridge University Press, Cambridge, pp. 326-345.
Martínez-Yrízar, A., Sarukhan, J., Perez-Jimenez, A., Rincon, E., Maass, J.M., Solis-Magallanes,
J.A., Cervantes, L., 1992. Above-ground phytomass of a tropical deciduous forest on the
coast of Jalisco, Mexico. Journal of Tropical Ecology. 8, 87-96.
Medeiros, A.S., Drezner, T.D., 2012. Vegetation, climate, and soil relationships across the Sonoran
Desert. Écoscience. 19, 148-160.
Meir, P., Pennington, T., 2011. Climatic change and seasonally dry tropical forest. In: Dirzo, R.,
Young, H., Mooney, H., Ceballos, G. (Eds.), Seasonally Dry Tropical Forest: Ecology and
Conservation. Island Press, USA, pp. 279-299.
Mendoza, A., 2015. Vulnerability of biodiversity to land use change and climate change in Mexico.
Physical Geography. University of Edinburgh, Edinburgh.
Miles, L., Newton, A., DeFries, R., Ravilious, C., May, I., Blyth, S., Kapos, V., Gordon, J., 2006. A
global overview of the conservation status of tropical dry forests. J. Biogeogr. 33, 491-505.
46
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
Mirmanto, E., Proctor, J., Green, J., Nagy, L., Suriantata, 1999. Effects of nitrogen and phosphorus
fertilization in a lowland evergreen rainforest. Phil. Trans. Royal Soc. B: Biol. Sci. 354,
1825-1829.
Murphy, P.G., Lugo, A.E., 1986a. Ecology of tropical dry forest. Annu. Rev. Ecol. Syst. 17, 67-88.
Murphy, P.G., Lugo, A.E., 1986b. Structure and biomass of a subtropical dry forest in Puerto Rico.
Biotropica. 18, 89-96.
Nascimento, H.E.M., Laurance, W.F., 2002. Total aboveground biomass in central Amazonian
rainforests: a landscape-scale study. For. Ecol. Manage. 168, 311-321.
Návar, J., 2009. Allometric equations and expansion factors for tropical dry forests trees of Eastern
Sinaloa, Mexico. Trop. Subtrop. Agroecosys. 10, 45-52.
Návar, J., 2010. Biomass allometry for tree species of Northwestern Mexico. Trop. Subtrop.
Agroecosys. 12, 507-519.
NOAA, 2015. NOAA Solar Calculator. Earth System Research Laboratory, Global Monitoring
Division. (https://www.esrl.noaa.gov/gmd/grad/solcalc/).
Olsen, S.R., Cole, C.V., Watanabe, F.S., Dean, L.A., 1954. Estimation of available phosphorus in
soil by extraction with sodium bicarbonate.No. 939. Department of Agriculture, Washington
D.C.
Ordoñez, J.C., van Bodegom, P.M., Witte, J.P.M., Wright, I.J., Reich, P.B., Aerts, R., 2009. A
global study of relationships between leaf traits, climate and soil measures of nutrient
fertility. Global Ecol. Biogeogr. 18, 137-149.
Pan, Y.D., Birdsey, R.A., Fang, J.Y., Houghton, R., Kauppi, P.E., Kurz, W.A., Phillips, O.L.,
Shvidenko, A., Lewis, S.L., Canadell, J.G., Ciais, P., Jackson, R.B., Pacala, S.W., McGuire,
A.D., Piao, S.L., Rautiainen, A., Sitch, S., Hayes, D., 2011. A large and persistent carbon
sink in the World's forests. Science. 333, 988-993.
47
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
Peterson, F.S., 2012. Post-Harvest Establishment Influences ANPP, Soil C and DOC Export in
Complex Mountainous Terrain. PhD Thesis in Forest Science. Oregon State University,
Oregon.
Petit-Aldana, J., Casanova-Lugo, F., Solorio-Sánchez, J., Ramírez-Avilés, L., 2011. Litterfall
production and quality in pure and mixed fodder banks in Yucatan, Mexico. Rev. Chapingo
Ser. Ciencias Forest. Amb. 17, 165-178.
Powers, J.S., Becknell, J.M., Irving, J., Pèrez-Aviles, D., 2009. Diversity and structure of
regenerating tropical dry forests in Costa Rica: Geographic patterns and environmental
drivers. For. Ecol. Manage. 258, 959-970.
QGIS-2.6.0, 2014. Quantum GIS Development Team, Quantum GIS Geographic Information
System. Open Source Geospatial Foundation Project. (http://qgis.osgeo.org).
Rango, A., Tartowski, S.L., Laliberte, A., Wainwright, J., Parsons, A., 2006. Islands of
hydrologically enhanced biotic productivity in natural and managed arid ecosystems. J. Arid
Environ. 65, 235-252.
R-Core-Team, 2014. R: A language and environment for statistical computing. Foundation for
Statistical Computing, Vienna.
Read, L., Lawrence, D., 2003a. Recovery of Biomass following shifting cultivation in dry tropical
forests of the Yucatan. Ecological Appl. 13, 85-97
Reich, P.B., Borchert, R., 1984. Water stress and tree phenology in a tropical dry forest in the
lowlands of Costa Rica. J. Ecol. 72, 61-74.
Ribeiro, J., Diggle, P.J., 2001. geoR: A package for geostatistical analysis. R-NEWS Vol 1, No 2.
ISSN 1609-3631, (http://cran.r-project.org/doc/Rnews).
Ribeiro, S.C., Fehrmann, L., Soares, C.P.B., Jacovine, L.A.G., Kleinn, C., de Oliveira Gaspar, R.,
2011. Above- and belowground biomass in a Brazilian Cerrado. For. Ecol. Manage. 262,
491-499.
48
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
Roa-Fuentes, L.L., Campo, J., Parra, V., 2012. Plant biomass allocation across a precipitation
gradient: an approach to seasonally dry tropical forest at Yucatán, Mexico. Ecosystems. 15,
1234-1244.
Rossi, J., Govaerts, A., De Vos, B., Verbist, B., Vervoort, A., Poesen, J., Muys, B., Deckers, J.,
2009. Spatial structures of soil organic carbon in tropical forests - a case study of
Southeastern Tanzania. Catena. 77, 19-27.
Ryser, P., Urbas, P., 2000. Ecological significance of leaf life span among Central European grass
species. Oikos. 91, 41-50.
Rzedowski, J., 1998. Diversidad y orígenes de la flora fanerogámica de México. In: T.P.
Ramamoorthy, R.B., Fa, A.L.y.J. (Eds.), Diversidad biologica de Mexico: origenes y
distribucion. Instituto de Biología-UNAM, Mexico, DF, pp. 129-145.
Sagar, R., Singh, J., 2006. Tree density. basal area and species diversity in a disturbed dry tropical
forest of northern India: implications for conservation. Environ. Conserv. 33, 256-262.
Salcedo, I.H., Tiessen, H., Sampaio, E.V.S.B., 1997. Nutrient availability in soil samples from
shifting cultivation sites in the semi-arid Caatinga on NE Brazil. Agric. Ecosys. Environ. 65,
177-186.
Saner, P., Loh, Y.Y., Ong, R.C., Hector, A., 2012. Carbon stocks and fluxes in tropical lowland
Dipterocarp rainforests in Sabah, Malaysian Borneo. PLoS ONE. 7, e29642:29641–
e29642:29611.
Saynes, V., Hildago, C., Etchevers, J.D., Campo, J.E., 2005. Soil C and N dynamics in primary and
secondary seasonally dry tropical forests in Mexico. Appl. Soil Ecol. 29, 282–289.
Shoshany, M., 2000. Detection and analysis of soil erodibility patterns using air photographs of the
Avisur Highlands, Israel. In: Hassan, M., Slymaker, O., S.M, B. (Eds.), The Hydrology-
Geomorphology Interface: Rainfall, Floods, Sedimentation. Landuse. Int. Assoc. Hidrol. Sci.
Publ, pp. 127-138.
49
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
Shoshany, M., 2002. Landscape fragmentation and soil cover changes on south- and north-facing
slopes during ecosystems recovery: an analysis from multi-date air photographs.
Geomorphology. 45, 3-20.
Sight, J.S., Sight, L., Pandey, C.B., 1991. Savannization of dry tropical forest increases carbon flux
relative to storage. Cur. Sci. 61, 477-480.
Skutsch, M., McCall, K., Lovett, J., 2009. Carbon emissions: dry forests may be easier to manage.
Nature. 7273, 462. doi: 10.1038/462567b.
Slik, J.W.F., Aiba, S.I., Brearley, F.Q., Cannon, C.H., Forshed, O., Kitayama, K., Nagamasu, H.,
Nilus, R., Payne, J., Paoli, G., Poulsen, A.D., Raes, N., Sheil, D., Sidiyasa, K., Suzuki, E.,
van Valkenburg, J.L.C.H., 2010. Environmental correlates of tree biomass, basal area, wood
specific gravity and stem density gradients in Borneo's tropical forests. Global Ecol.
Biogeogr. 19, 50-60.
Slik, J.W.F., Paoli, G., McGuire, K., Amaral, I., Barroso, J., Bastian, M., Blanc, L., Bongers, F.,
Boundja, P., Clark, C., Collins, M., Dauby, G., Ding, Y., Doucet, J.-L., Eler, E., Ferreira, L.,
Forshed, O., Fredriksson, G., Gillet, J.-F., Harris, D., 2013. Large trees drive forest
aboveground biomass variation in moist lowland forests across the tropics. Global Ecol.
Biogeogr. 22, 1261-1271.
SMN, 2014. Servicio Metereologico Nacional. CONAGUA, Mexico.
Snyder, K.A., Tartowski, S.L., 2006. Multi-scale temporal variation in water availability:
Implications for vegetation dynamics in arid and semi-arid ecosystems. J. Arid Environ. 65,
219-234.
Spadavecchia, L., Williams, M., Bell, R., Stoy, P.C., Huntley, B., van Wijk, M.T., 2008.
Topographic controls on the leaf area index and plant functional type of a tundra ecosystem.
J. Ecol. 96, 1238-1251.
Sperry, J.S., Adler, F.R., Campbell, G.S., Comstock, J.P., 1998. Limitation of plant water use by
rhizosphere and xylem conductance: results from a model. Plant Cell Environ. 21, 347-359.
50
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
Sperry, J.S., Hacke, U.G., 2002. Desert shrub water relations with respect to soil characteristics and
plant functional type. Funct. Ecol. 16, 367-378.
Thomas, S.C., Malczewski, G., 2007. Wood carbon content of tree species in eastern China:
Interspecific variability and the importance of the volatile fraction. J. Env. Manag. 85, 659-
662.
Tobón, W., Urquiza-Haass, T., Koleff, P., Schroeter, M., Ortega-Álvarez, R., Campo, J., Lindig
Cisneros, R., Sarukhán, J., Bonn, A., 2017. Restoration planning using a multi-criteria
approach to guide Aichi targets in a megadiverse country. Conserv. Biol. 31, 1086-1097.
Torres, J.A., Gonzáles, G., 2005. Wood decomposition of Cyrilla racemiflora (Cyrillaceae) in
Puerto Rican dry and wet forests: A 13-year case study. Biotropica 37, 452-456.
Trejo, I., 1998. Distribución y diversidad de selvas bajas de México: relaciones con el clima y el
suelo. PhD thesis, Universidad Nacional Autónoma de México, Mexico City, p. 210.
Trejo, I., Dirzo, R., 2002. Floristic diversity of Mexican seasonally dry tropical forests. Biodiv.
Conserv. 11, 2063- 2084.
Turner, M.J., 2005. Landscape ecology: What is the state of the science? Annu. Rev. Ecol., Evol.,
Syst. 36, 319-344.
Velázquez, A., Durand, E., Ramírez, I., Mas, J., Bocco, G., Ramírez, G., Palacio, J., 2003. Land
use-cover change processes in highly biodiverse areas: the case of Oaxaca, México Global
Environmental Change 13:175-184.
Vitousek, P., 2004. Nutrient Cycling and Limitation. Princeton University Press, Princeton. 223 pp.
Walker, S.M., Desanker, P.V., 2004. The impact of land use on soil carbon in miombo woodlands
of Malawi. For. Ecol. Manage. 203, 345-360.
Wall, D.H., González, G., Simmons, B.L., 2011. Seasonally dry tropical forest soil diversity and
functioning. In: Dirzo, R., Young, H., Mooney, H., Ceballos, G. (Eds.), Seasonally Dry
Tropical Forest: Ecology and Conservation. Island Press, USA, pp 61-70.
51
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
Wick, B., Tiessen, H., Menezes, R.S.C., 2000. Land quality changes following the conversion of the
natural vegetation into silvo-pastoral systems in semi-arid NE Brazil. Plant Soil. 222, 59-70.
Woollen, E., Ryan, C., Williams, M., 2012. Carbon stocks in an African Woodland landscape:
Spatial distributions and scales of variation. Ecosystems. 15, 804-818.
Wright, J., van Schaik, C.P., 1994. Light and the phenology of tropical trees. Amer. Naturalist. 143,
192-199.
Wright, J.S., Yavitt, J.B., Wurzburger, N., Turner, B.L., Tanner, E.V., Sayer, E., Santiago, L.S.,
Kaspari, M., Hedin, L.O., Harms, K.E., Garcia, M.N., Corre, M.D., 2011. Potassium,
phosphorus, or nitrogen limit root allocation, tree growth, or litter production in a lowland
tropical forest. Ecology. 92, 1616-1625.
Yamada., A., Inoue, A., Wiwatwitaya, D., Ohkuma, M., Kudo, T., Sugimoto, A., 2006. Nitrogen
fixation by termites in tropical forest, Thailand. Ecosystems 9, 75-83.
Yeaton, R.I., Travis, J., Gilinsky, E., 1977. Competition and spacing in plant communites. The
Arizona Upland association. J. Ecol. 65, 587-595.
Yingduan, H., Arturo, S.-A., Benoit, R., 2013. Linkages among ecosystem structure, composition
and leaf area index along a tropical dry forest chronosequence in Mexico. In: Quesada, M.,
(Ed.), Tropical Dry Forests in the Americas. CRC Press, pp. 267-280.
Zimmermann, N.E., 2000. toposcale.aml: An ArcInfo-script for computation of multi-scale
topographic position. Swiss Federal Research Institute (WSL), Birmensdorf.
http:www.wsl.ch/staff/niklaus.zimmermann/programs/aml4_1.html, (Accessed: 19/10/2016).
52
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162