3.1. Bulk Density, Soil Water Retention and Plant Available Water
The
U value measured at the considered |
h| values increased when BC was added to the three soils, with the differences being significantly different (
p < 0.05), except for the
U values measured for the clay soil at |
h| ≥ 1020 cm. The average gravimetric water contents,
U (g/g), corresponding to the applied pressure head values, |
h|, for the soils loamy sand, loamy sand + BC, loam, loam + BC, clay and clay + BC has been evaluated (
Table 2).
For the three soils, BC significantly increased both the water content at saturation (
Us) and the residual water content (
Ur), also increasing
α and
n. The Mualem-van Genucthen parameters obtained by fitting Equation (6) to the (
U, |
h|) pairs of the three BC-amended and non-amended soils were considered to explain the BC activities (
Table 3).
PAW
U (g/g) significantly increased from 0.033 to 0.092 for the loamy sand, from 0.0705 to 0.109 for the loam and from 0.0573 to 0.133 for the clay, after BC application (
Table 4). Differences (%) between the PAW measured in the BC amended soils and in the original soils were equal to 178%, 54.7% and 131.6% for the loamy sand, loam and clay respectively, showing that the effect of BC was in the order loamy sand > clay > loam. The bulk density,
ρb obtained for the non-amended and for the BC amended soils shows has the values of bulk density is a multivariate parameter affect by the mineral and organic material in which no linear definition is considerable (
Table 4). The lower bulk densities in clay + BC originate from mixing the lower density material in the clayey soil, is consistent with a no significant changes in aggregate structure. The phenomenon has been previously reported in a field study conducted by Sonnie et al. [
48] on clay soil also by using forest biomass BC. Therefore, appear more considerable from the reported laboratory experiment result that demonstrate as changes in the soil’s physical properties due to BC addition depend on soil type, instead of BC’s capacity to store water in its internal structure is immutable parameter.
The
ρb measured for the BC, equal to 0.173 g/cm
3 (
Table 1) was considerably lower than that measured in the non-amended soils, equal to 1.38 g/cm
3, 1.11 g/cm
3 and 1.20 g/cm
3 for the loamy sand, for the loam and for the clay, respectively. As a consequence of BC addition,
ρβ (g/cm
3) significantly decreased from 1.38 to 0.828 in the loamy sand + BC, from 1.11 to 0.726 in the loam + BC, and from 1.208 to 0.757 in the clay + BC, with ratios equal to 1.67, 1.53 and 1.60, respectively, between the
ρβ of the non-amended soils and the
ρβ of the BC-amended soils.
The decrease in
ρβ due to BC application [
47] agrees with the results of previous investigations [
31,
49,
50,
51], and it is due to a mixing or dilution effect [
51] determined by the difference between the
ρβ of the non-amended soils and of BC.
The soil water retention curves, expressed as |
h| vs. the gravimetric water content
U (g/g), evidencing that the considered standard soils were characterized by a very poor water retention before BC application (
Figure 1a–c).
Soil water retention curves represented as |
h| vs.
θ were also represented (
Figure 1d–f). Since the measured soil shrinkage was negligible both in the BC amended soils and in the soils without BC, constant
ρβ values (
Table 4) were used to convert
U into
θ (Equation (4)). The
θ values of the loamy sand + BC retention curve were systematically higher than those obtained for the loamy sand, with significant differences in the
θ values that increased at the lowest |
h| and at saturation (
Figure 1d). This indicated that BC improved the soil water retention at all the considered |
h|.
Instead, the
θ values of the loam + BC retention curve (
Figure 1e) were lower than those obtained for the loam, except for the
θ values measured at |
h| > 180 cm as well as for
θs, (
Table 4), indicating an increase in the amount of macropores, and a decrease in the amount of micropores. This result indicated a partial improvement in the retention properties of the loam after BC addition.
The
θ values of the clay + BC (
Figure 1f) were significantly lower than those obtained for the clay, indicating a non-positive effect of BC, due to formation of a larger number of aggregates or particles leading to smaller inter-aggregate pores than the original ones. Also, the saturated water content,
θs obtained for the clay + BC was significantly lower than that obtained for the clay, indicating a decrease in the amount of macropores.
Comparing the
θ and the
U values measured at the fixed |
h|, (
Figure 1) it appears evident that for the clay, the
U values measured after BC addition were higher than those measured before BC addition (
Figure 1c), but the opposite was found in terms of
θ values (
Figure 1f).
The loam shows the same trend, except for the
θ values at |
h| ≤ 180 cm (
Figure 1b,e). The loamy sand, instead, the same behaviour can be observed between the
U(
h) and the
θ(
h) functions before and after BC addition (
Figure 1a,d), with both the pairs (
U,
h) and (
θ,
h) increasing after BC addition, mainly near the low matric potentials (|
h|).
Reductions in the dry bulk density, ρβ, of the soil + BC mixtures, obtained for the loamy sand, for the loam and for the clay, expressed as ratios ρb,soil/ρb,soil+BC, were equal to 1.67, 1.53 and 1.60, respectively.
However, the ρb reduction did not determine an increase the θ values in the BC-amended loam (except for the θ values at |h| ≤ 180 cm) and in the BC-amended clay, compared to the increase observed on the U values, because for these two soils the ratios Usoil+BC/Usoil (average ratio Usoil+BC/Usoil = 1.33 for the loam and 1.11 for the clay) were lower than the ratios ρb,soil/ρb,soil+BC. (1.53 and 1.60). Instead, for the loamy sand, the ratio Usoil+BC/Usoil was higher (average ratio = 2.73) than the ratio ρb,soil/ρb,soil+BC, (1.67), and therefore the θ values obtained after BC addition were higher than those measured in the non-amended loamy sand. This explains why BC had the effect of increasing the θ values in the loamy sand, whereas θ values lower than those obtained before BC addition were obtained for the clay and, partially, for the loam.
This suggests that in clay soils higher BC amounts might improve the soil water retention [
29] by determining
Usoil+BC/U
soil ratios higher than the
ρb,soil/
ρb,soil+BC, and positive differences between the
θ values of the BC-amended soils and the
θ of the soils without BC.
Herath et al. [
38] also reported more evident effects of BC on the total soil porosity when a higher difference occurred between the
ρb of BC and the
ρb of the soil. This suggests that an excessive reduction in the
ρb of the soil mixtures is not always beneficial in terms of water retention, confirming that positive effects of BC application are to be expected in soils having high initial
ρb values, such as the coarse textures soils, and not in soils with lower initial
ρb, such as the fine textured ones.
In conclusion, BC certainly improved water retention of the loamy sand, confirming the results of previous investigations on coarse textured soils [
31,
33,
34] partially improved water retention of the loam, increasing macroporosity, as also reported by Ouyang et al. [
27], but did not determine positive effects on water retention of the clay soil, as also found in previous investigations [
22,
37,
38].
The differences obtained in our soil water retention curves obtained before and after BC addition by expressing the soil water content in terms of
U or in terms of
θ values, also explain why some contrasting results have been found when the effect of BC on fine textured soils has been investigated. Of the previously mentioned papers, those reporting positive effect of BC based their conclusions considering the measured
U values [
19,
28], whereas those reporting no effects, or negative BC effects, based their analyses and conclusions considering the
θ values [
36,
37,
38,
39].
Both
θfc and
θwp significantly increased in the loamy sand, but significantly decreased in the loam and in the clay after BC addition. Values of the saturated water content,
θs, the measured field capacity,
θfc, the measured wilting point,
θwp, obtained for the three soils amended and not amended with BC, are reported in
Table 4.
D values (μm) pore diameter corresponding to the |
h| values at which
θfc and
θwp were measured, i.e., |
h| = 330 cm and 15,300 cm, respectively, were equal to 9.09 μm and 0.2 μm. The amount of pores with
D = 9.09 μm was classified as storage pores (0.5–50 μm), and the amount of pores with
D = 0.2 μm, called residual pores (
D < 0.5 μm) [
52] increased after BC application in the loamy sand, but decreased in the loam + BC as well as in the clay + BC. The amount of pores corresponding to
θs, indicated as macropores, significantly increased in the loamy sand + BC and in the loam + BC, but decreased in the clay + BC.
Plant available water, PAW (cm
3/cm
3), calculated on the basis of the
θ values reported in significantly increased from 0.046 to 0.076 for the loamy sand + BC, non-significantly increased from 0.078 to 0.079 in the loam + BC, and significantly increased from 0.069 to 0.104 in the clay + BC (
Table 4). The increase in PAW was equal to 66.1% and to 45.7% for the loamy sand and for the clay, respectively. These results indicated that BC positively affected the PAW of the loamy sand and of the clay but did not determine any significant increase in the PAW of the loam.
PAW values, calculated based on the
θ values, were lower than those obtained on the basis of the
U values, PAW
U (
Table 4), as a consequence of the reduced
ρb in the soils + BC mixtures, but the trend was the same, with the highest increment in the PAW values obtained for loamy sand + BC, followed by the increment observed on the clay + BC. Instead, in the loam + BC, there was a non-significant increase in the PAW, expressed based on
θ, compared to the previously found significant increase obtained by considering the
U values.
This means that in some cases analysis of the effect of BC on PAW might lead to different conclusions if carried out using the U or the θ values and could not be in agreement with the results obtained in terms of water retention curve. For the clay soil, our results indicated an increase in the PAW, even if the θ(h) functions obtained on the clay + BC mixture showed a decrease in the θ values, at fixed h, a decrease in the pore-size, as well as a reduction in the macroporosity, expressed by θs.
Considering applications related to irrigation scheduling, in the loamy sand and in the clay soil the increased PAW may be important to the enhancement of plant productivity as well as to the reduction of irrigation amounts and/or frequency, while no improvements are to be expected for the loam in terms of irrigation scheduling. However, applications carried out with physically based models considering soil water retention and hydraulic conductivity [
52,
53] should be carried out to evaluate how the hydraulic characteristics of the three soils after BC addition would affect water flow and crop water uptake, with effects on irrigation scheduling that could be quantified by simulating management scenarios [
54,
55].
3.2. Aggregate Stability Index
Volume of drainable pores, Δ
Ug, the modal suction, |
τd|, and the SI values obtained for the three BC amended and non-amended soils. Δ
Ug significantly increased in the loamy sand + BC and in loam + BC compared to the soils without BC but decreased in the clay + BC compared to the clay (
Table 5).
The modal suction, |τd|, describing the water potential corresponding to the liquid constrained in the most common pore size diameter, significantly decreased in the loamy sand + BC and in the loam + BC soil, indicating an increase in the most frequent pore diameter, D, and thus a pores’ enlargement, but significantly increased in the clay + BC, indicating a reduction in the most frequent D.
Compared to the original soils, the SI values increased by about 60 times in both the loamy sand + BC and the loam + BC, but decreased by 0.48 times for the clay + BC, indicating an improvement in the aggregate stability condition for the loamy sand and for the loam, but not for the clay.
The diameter,
D (μm), corresponding to |
τd|, increased from 25 to 200 for the loamy sand, and from 20 to 433 for the loam and, according to the Greenland classification, shifted from the range of storage pores (0.5 μm <
D < 50 μm) to the range of transmission pores (
D ≥ 50 μm), showing that BC induced formation of water stable macro-aggregates, which stored more water than small aggregates, as also shown by the increased Δ
Ug. Similar results were reported by Baiamonte et al. and by Herath et al. [
31,
38], for a desert sandy soil. Instead, for the clay,
D (μm) decreased from 39 to 24, falling in the range of storage pores, and indicating that BC induced formation of aggregates characterized by smaller inter-aggregate pores than in the original soil, as also shown by the decreased Δ
Ug.
These results indicated that BC did not improve the aggregate stability condition of the clay soil, appearing consistent with those obtained by analysing the θ(h) functions obtained before and after BC addition.