Digital mapping of soil physical and mechanical properties using machine learning at the watershed scale

Knowledge about the spatial distribution of the soil physical and mechanical properties is crucial for soil management,water yield,and sustainability at the watershed scale;however,the lack of soil data hinders the application of this tool,thus urging the need to estimate soil properties and consequently,to perform the spatial distribution.This research attempted to examine the proficiency of three machine learning methods(RF:Random Forest;Cubist:Regression Tree;and SVM:Support Vector Machine)to predict soil physical and mechanical properties,saturated hydraulic conductivity(Ks),Cohesion measured by fall-cone at the saturated(Psat)and dry(Pdry)states,hardness index(HI)and dry shear strength(SS)by integrating environmental variables and soil features in the Zayandeh-Rood dam watershed,central Iran.To determine the best combination of input variables,three scenarios were examined as follows:scenarioⅠ,terrain attributes derivative from a digital elevation model(DEM)+remotely sensed data;scenarioⅡ,covariates of scenarioⅠ+selected climatic data and some thematic maps;scenarioⅢ,covariates in scenarioⅡ+intrinsic soil properties(Clay,Silt,Sand,bulk density(BD),soil organic matter(SOM),calcium carbonate equivalent(CCE),mean weight diameter(MWD)and geometric weight diameter(GWD)).The results showed that for Ks,Psat Pdry and SS,the best performance was found by the RF model in the third scenario,with R2=0.53,0.32,0.31 and 0.41,respectively,while for soil hardness index(HI),Cubist model in the third scenario with R2=0.25 showed the highest performance.For predicting Ks and Psat,soil characteristics(i.e.clay and soil SOM and BD),and land use were the most important variables.For predicting Pdry,HI,and SS,some topographical characteristics(Valley depth,catchment area,mltiresolution of ridge top flatness index),and some soil characteristics(i.e.clay,SOM and MWD)were the most important input variables.The results of this research present moderate accuracy,however,the methodology employed provides quick and costeffective information serving as the scientific basis for decision-making goals.

Soil Organic Carbon Pools in Particle-Size Fractions as Affected by Slope Gradient and Land Use Change in Hilly Regions,Western Iran

This study was conducted to explore the effects of topography and land use changes on particulate organic carbon(POC),particulate total nitrogen(PTN),organic carbon(OC) and total nitrogen(TN) associated with different size primary particle fractions in hilly regions of western Iran.Three popular land uses in the selected site including natural forest(NF),disturbed forest(DF) and cultivated land(CL) and three slope gradients(0-10 %,S1,10-30 %,S2,and 30-50%,S3) were employed as the basis of soil sampling.A total of 99 soil samples were taken from the 0-10 cm surface layer in the whole studied hilly region studied.The results showed that the POC in the forest land use in all slope gradients was considerably more than the deforested and cultivated lands and the highest value was observed at NF-S1 treatment with 9.13%.The values of PTN were significantly higher in the forest land use and in the down slopes(0.5%) than in the deforested and cultivated counterparts and steep slopes(0.09%) except for the CL land use.The C:N ratios in POC fraction were around 17-18 in the forest land and around 23 in the cultivated land.In forest land,the silt-associated OC was highest among the primary particles.The enrichment factor of SOC,EC,was the highest for POC.For the primary particles,EC of both primary fractions of silt and clay showed following trend for selected land uses and slope gradients:CL> DF> NF and S3 > S2> S1.Slope gradient of landscape significantly affected the OC and TN contents associated with the silt and clay particles,whereas higher OC and TN contents were observed in lower positions and the lowest value was measured in the steep slopes.Overall,the results showed that native forest land improves soil organic carbon storage and can reduce the carbon emission and soil erosion especially in the mountainous regions with high rainfall in west of Iran.

An attempt to find a suitable place for soil moisture sensor in a drip irrigation system

Determination of an appropriate location for monitoring soil water content (SWC) is a keyfactor in efficient use of water in precision agriculture, however, the main challenge is thedynamic movement of water and root development in the soil profile. The objective of thisstudy was investigating how SWC distribution in a loam soil profile at two growth stages ofmaize may impact the suitable location for SWC monitoring in a drip-tape irrigation system.A new concept, Average Moisture Representative Surface (AMRS) was proposed to determine the surface of the soil profile, which represent the average soil moisture of the wettedvolume. SWC samples were taken during two irrigation intervals (48~52 days after planting(DAP) and 68~73 DAP) and root growth pattern was studied through root length density (RLD)at 50 and 100 DAP. The results revealed that a non-uniform wetting pattern after irrigationlimits the appropriate locations for SWC monitoring to point measurements and with time,SWC depletion resulted in enlarging AMRS. At the end of growing season, an increase of rootgrowth around the drippers increased the variation of root water uptake in different soil layers, and thus optimal place for soil sensors was limited to the upper layers, where the maximum root water uptake occurred. Overall, it is recommended to install soil sensors, such astensiometers and TDRs at a horizontal distance of 5~20 cm from the crop and a depth of10~20 cm from the soil surface while drip-tape is aligned close to the maize row.