Table 3 Characteristics of land degradation data used for modelling the agricultural land multi-degradation in Europe

From: A unifying modelling of multiple land degradation pathways in Europe

No.

Land degradation data

Original resolutiona

Time period

Metric/Unit of measure

Critical threshold/classb

Data sourcec

1

Water erosion

250 × 250 m

2012

t ha–1 yr–1

> 2 t ha–1 yr–1

11

2

Wind erosion

500 × 500 m

2001–2021

t ha–1 yr–1

> 2 t ha–1 yr–1

13

3

Soil organic carbon loss

1 × 1 km

2001–2015

t C km2 yr–1

<−0.1 t C km2 yr–1d

19

4

Soil salinization

1 × 1 km

2008

%

> 50 %e

30

5

Soil acidification

500 × 500 m

2019

pH units

< 5.5

31

6

Soil compaction

1 × 1 km

2008

Susceptibility

H and VH SC

32

7

Soil nutrient imbalances

1 × 1 kmf

100 ×100 mg

2010–2019

kg/ha (N) mg/kg (P)

> 50 kg/ha/NUE > 0.9h

> 50 mg/kg/< 25 mg/kgi

31,33,34,35,89

8

Soil pollution via pesticides

10 × 10 km

2015

Risk score

H and VH RSCj

29

9

Soil pollution via heavy metals

1 × 1 kmk

500 × 500 ml

250 × 250 mm

2009

mg/kg

> 5 (As), > 1 (Cd), > 100 (Cr), > 20 (Co), > 60 (Pb), > 2 (Sb), > 50 (Ni), > 100 (Cu), > 0.5 (Hg)n

38,39,40

10

Vegetation degradation

500 × 500 m

2000–2015

NDVI units

<−0.001 units yr–1o

43

11

Groundwater decline

1 × 1 km

2004–2013

GTD (m yr–1)

<−0.001 m yr–1p

49

12

Aridity

1 × 1 km

1981–2018

AI (mm/mm)

< 0.65 mm/mm

51,52

  1. m meter, km kilometer, ha hectare, t ton, C carbon, H high, VH very high, SC susceptibility classes, N nitrogen, P phosphorous, NUE Nitrogen Use Efficiency, RSC risk score classes, NDVI Normalized Difference Vegetation Index, GTD Groundwater Table Depth, AI Aridity Index. aspatial resolution of the originally collected data, which were processed in this study at a common resolution of 500 × 500 m. bcritical thresholds over/under which each land degradation process triggers the reduction or loss of agricultural land productivity. These thresholds used for modelling agricultural land multi-degradation were documented and set for each process based on scientific literature: water and wind erosion10,61,90, soil organic carbon loss2,6, soil salinization30, soil acidification4,72, soil compaction32, soil nutrient imbalances33,36,37,89, soil pollution via pesticides29, soil pollution via heavy metals38,42 and aridity2,53. As no concrete thresholds were found in the literature for vegetation degradation and groundwater decline, in these two cases some critical thresholds/classes were set in accordance with the reasoning explained in o and p. csource for databases that were already available and directly collected (processes with no. 1, 3, 4, 5, 6 and 8) or for pre-existing data used for refining/modelling/obtaining the other processes in this study (2, 7, 9, 10, 11 and 12). dnegative statistically significant trends of soil organic carbon stock, detected during 2001–2015 using the MK test and Sen’s slope estimator. epercentage of areas affected by saline and sodic soils (mainly Solonchaks and Solonetz). ffor N data. gfor P data. hfor highlighting the N excess (>50 kg/ha) and deficit (NUE > 0.9) in soil. ifor highlighting the P excess (>50 mg/kg) and deficit (<25 mg/kg) in soil. jhigh (3 < RS ≤ 4) and very high (RS > 4) risk of soil pollution (with dozens of pesticides in Europe), according to the data source. kfor Arsenic (As), Cadmium (Cd), Chrome (Cr), Cobalt (Co), Lead (Pb), Antimony (Sb), and Nickel (Ni) data. lfor Copper (Cu) data. m – for Mercury (Hg) data. nconcentrations of nine heavy metals, above the standard guideline of safe limits. onegative statistically significant trends of NDVI, detected during 2000–2015 using the MK test and Sen’s slope estimator. pnegative statistically significant trends of GTD, detected during 2004–2013 using the MK test and Sen’s slope estimator. For trend-based data of soil organic carbon loss, vegetation degradation and groundwater decline, the confidence level of the MK (two-tailed) test was set to a p-value threshold ≤ 0.1, which includes both highly statistically significant trends (for p-values ≤ 0.05) and trends with lower statistical significance (p-values between 0.05 and 0.1).
Back to article page