Land Health Surveillance for identifying land constraints and targeting sustainable intensification in smallholder’s agriculture in Cameroon.

Takoutsing Bertin1, Ayenkulu Ermias2, Tchoundjeu Zacharie1, Edmundo Barrios2, Richard Coe2, Keith Shepherd2

Food production will have to increase substantially to overcome the challenges of the fast growing demand for food, widespread hunger and malnutrition. Intensification to increase agricultural productivity is seen as one of the solutions and entails among other things the enhancement of the capacity of soil to increase yields per hectare, increasing cropping intensity per unit of land, and changing land use from low value crops or commodities to those that receive higher market prices. There is therefore a need to establish a link between soil health and agricultural intensification. The objectives of this study were therefore to 1) establish baseline measurements to monitor land management impact over time, 2) describe land health patterns and associated degradation and 3) deduce implications for sustainable agricultural intensification

2) Materials and methods

Figure 1: Map of Cameroon showing the study site

2.2 Sampling framework The framework used Land Health Surveillance and the Land Degradation Surveillance Framework (LDSF) was used for the field implementation. The LDSF is a spatially stratified, random sampling design framework built around a hierarchical field survey and sampling protocol using the concept of sentinel site of 100 km2 each (10km × 10km) (figure 2a).

Property (n = 32) r-squared rmse

Clay 0.51 0.14

Silt 0.44 0.51

Sand 0.29 1.06

ExCa 0.76 1.03

ExK 0.43 0.78

ExMg 0.75 0.87

Al 0.58 0.11

Mn 0.60 0.90

P 0.53 0.55

Zn 0.12 0.49

pH 0.52 0.07

Total.Nitrogen 0.88 0.30

Total.Carbon 0.90 0.26

Acidified.Nitrogen 0.87 0.29

Acidified.Carbon 0.87 0.25

2.6 Statistical analysis Linear mixed models were used to compare differences in soil and vegetation results by land use/land cover, elevation and slope classification using cluster as a random effect. All statistical analysis were done using the open source software R version 3.0.2

3) Results and discussion

Our results showed that large parts of the cultivated lands have soils with good physical (texture) and chemical (soil carbon) compositions as potential to support agricultural intensification in the study area. • Textural analysis revealed that the site is dominated by clay soils

that drained well with acceptable infiltration rate. • Only 20% of plots have SOC below the critical level of 2% (Figure 7)

2.3 Field Data collection

Data were collected at the plot level and at the sub plot levels. Three overlapping subsets of indicators were used to assess three attributes of the site: soil and site stability, hydrologic function, and biotic integrity (Figure 3).

Figure 7: Soil organic carbon (SOC) and pH variation across the LDSF plots sampled

We found some limitations that require appropriate land management interventions if intensification has to be successful and sustainable. • Large part of the areas is dominated by sloping lands and cultivated

plots were found even on steep slopes (>20%). • High soil acidity (pH = 5 – 6) was observed in the area (Figure 8)

2.1 Study site

Figure 2: a) Illustration of the sentinel site of 100 km2 block divided into 16 clusters. b) Illustration of the 1 km2 clusters with 10 plots of 1000 m2 each. c) Sub-plots (area 100 m2)

*Corresponding author, b.takoutsing@cgiar.org 1World Agroforestry Centre, BP 16317 Yaounde, Cameroon, 2World Agroforestry Centre, PO Box 30677-00100, Nairobi, Kenya

1) Introduction

The study site is located in the West and North West regions of Cameroon, that make up the “Western Highlands” (Figure 1)

The best correlations (R2 > 50) obtained for clay, pH, Ca, Mg, Mn, Al, P, C and N suggest that there is quite a strong relationship between the results of the laboratory analysis and the MIR analysis procedures (Table 1)

3.1 Prediction of soil properties

A total of 320 samples were collected, processed and 10% analysed in the laboratory for selected chemical properties. The samples were scanned using a Bruker Alpha Drift FT MIR Spectrometer (Figure 4) and were analysed by MIR diffuse reflectance spectroscopy using the OPUS Laboratory software 6.5 version. For each selected soil property, a calibration model was developed and the known properties used to predict the values for the independent spectra in the remaining samples under investigation. (Table 1)

Figure 6: SOC concentration in cultivated (1) and uncultivated (0) plots Table 1: Calibration results for selected soil

chemical properties

4) Conclusion These limitations could be overcome by targeted land management intervention such as liming to reduce acidity and soil conservation measures to reduce erosion) for the successful implementations of agricultural intensification programs in smallholder’s agriculture in Cameroon.

Figure 8: Possible pH ranges under natural soil conditions

Figure 3: summary of data collected using the framework

Figure 4: Bruker Alpha Drift FT MIR Spectrometer Figure 5: A soil spectra obtained from the spectrometer

2.5 Soil sampling and spectral reflectance analysis

3.2 Potential for agricultural intensification

3.3 Limitation to agricultural intensification

Map of Cameroon showing the study site