2. MATERIALS AND METHODS
This research was developed in three phases: i) selection of sampling
units: data collection and analysis, and identification of main land
uses; ii) collection and analysis of soil samples: sampling design,
sample extraction and laboratory analysis; and iii) data analysis:
descriptive statistics and statistical hypothesis testing.
2.1 Study area
We studied the Berlin dry páramo located in the Eastern Cordillera of
Colombia at 07°06’15”N latitude and 72°56’35”W longitude, inside the
Santurbán páramo complex. This area is a strategic water source for
1,400 water-use rights, a hydropower generation central, and a
centralized supply system that provides water to around 1 million people
(C. Sarmiento & Ungar, 2014). The study area is located in the Parra
Juan Rodriguez village, spans between 3,420 and 3,630 m a.s.l., has an
average annual temperature of 8°C, a relative humidity of 86%, an
average annual rainfall of 702 mm, and multiannual monthly rainfall
between 40 mm and 702 mm during the dry and the wet seasons,
respectively (Avellaneda, 2012).
Figure 1 shows the behavior of the multiannual monthly rainfall reported
by the Berlin meteorological station (07°11’24”N latitude and
72°52’12”W longitude) for the period 1973–2018, obtained from data
provided by the Instituto de Hidrología, Meteorología y Estudios
Ambientales (IDEAM, 2020).
[Insert Figure 1]
The soils in the study area are classified as Typic Dystropepts in the
Inceptisols order, according to the United States Agricultural
Department taxonomy (USDA & NRCS, 2014), in association with a wavy
steep slope. These soils are acidic, isomesic and udic. Morphologically,
the soil profile has a distribution of horizons A-B-Cr: horizon A is
dark brown, with a clay loam texture; horizon B is yellow with a texture
similar to the upper horizon; horizon Cr is quartz monzonite. This is a
well-drained, deep and strongly acidic soil, with low contents of
calcium, magnesium and phosphorous, high in exchangeable aluminum and
low in fertility (IGAC, 2003).
2.2 Phase 1: Selection of sampling units
The study area comprises two hydrological units draining to the Jordan
river basin and was selected based on a review of cartographic
information from secondary sources, field visits and stakeholder
consultations, together with photointerpretation (Patiño, 2020). Two
field visits with detailed walking surveys were carried out to identify
the dominant land uses and to define sampling units (SUs). We selected
SUs that were adjacent, had the same type of soils and similar climate,
rainfall distribution, and topography to minimize confounding factors.
This consideration allows to reduce uncertainty related to random crop
locations in order to attribute observed changes on soil hydrophysical
properties to land-use types.
Six SUs were identified, three under natural páramo vegetation and three
under agricultural use. The SUs were named: páramo grasslands – low
slope (Ls), páramo grasslands – steep slow (Ss), páramo – dense
shrublands (Ds), agricultural use: fallow (F), spring onion crops (Oc),
and potato crops (Pc). Fallow, in particular, is the consequence of a
cultural practice in the area which consists of the removal of natural
vegetation, mechanical ploughing at 30 cm depth, followed by a cycle of
potato and barley cultivation, and then onion cultivation and livestock
grazing. This practice has been performed during a period of
approximately 30 years. This area is still grazed with an estimated
animal density of 1.3 cows per hectare.
2.3 Phase 2: Sample collection and analysis
Ten monitoring campaigns were carried out, five in each season, dry and
wet, over seven months (September 2019 – March 2020). A stratified
discrete random sampling at different depths was implemented (0–5,
10–15 and 20–25 cm). Samples and duplicates were randomly collected at
each SU, resulting in ten points in each SU for different seasons. Thus,
60 samples were obtained in each SU, accounting for 360 collected soil
specimens (180 in each season). Specimen extraction was performed using
a hand-operated bucket auger and thin Kopecky rings of 100
cm3. Four soil hydrophysical properties were then
analyzed: Soil Organic Matter (SOM), Bulk density (Bd), pH, and Electric
Conductivity (Ec). Table 1 shows details regarding sampling and
analytical methods.
[Insert Table 1]
Samples were transported in individual metallic rings with plastic lids
in labeled airtight bags, preserved in a plastic cooler box protected
from vibration using foam board layers, according to the recommendations
from ASTM D4220-95 (ASTM, 2007). The analyses were carried out up to a
maximum of two days after sample collection at the laboratory of the
Grupo de Investigación en Recursos Hídricos y Saneamiento Ambiental at
Universidad Industrial de Santander.
2.4 Phase 3: Data analysis
The assessment of differences on the soil hydrophysical properties was
carried out at two levels: i) the influence of the rainfall regime (dry
seasons and wet seasons) was established for each SU; and ii) the soil
hydrophysical properties in the SUs under natural vegetation were
compared with those in the SUs under anthropic land uses. Samples during
each rainfall regime were classified according to the behavior of the
multiannual monthly rainfall in the study area (Figure 1).
The soil hydrophysical properties for the two levels of analysis were
assessed considering the three depths in which the soil specimens were
extracted (0–5, 10–15 and 20–25 cm) in order to avoid confounding
effects of phenomena such as soil compaction, water content and organic
matter distribution (EPA, 2014). In the two levels of analysis, data
were assessed using descriptive statistics, completely randomized
analysis of variance (ANOVA; normality test with Shapiro-Wilk method)
and multiple comparisons (Fisher-LSD test). To analyze the influence of
land use on the soil hydrophysical properties, boxplots were used to
identify differences and data variability within the SU. All the
statistical tests were carried out at a significance level of 5%. In
both levels of analysis, data were contrasted with current scientific
literature on the influence of anthropic activities on the soils and
hydrology of páramo ecosystems to evidence distinct patterns particular
to those of dry páramos.