Grazing treatment influences recovery of mesic grassland from seasonal drought: An assessment using unmanned aerial vehicle-enabled remote sensing

Authors: Polley, Herbert; Kolodziejczyk, Chris; Jones, Katherine; Smith, Douglas

Submitted to: Rangeland Ecology and Management
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 1/27/2022
Publication Date: 5/18/2022
Citation: Polley, H.W., Kolodziejczyk, C.A., Jones, K.A., Smith, D.R. 2022. Grazing treatment influences recovery of mesic grassland from seasonal drought: An assessment using unmanned aerial vehicle-enabled remote sensing. Rangeland Ecology and Management. 82:12-19. https://doi.org/10.1016/j.rama.2022.01.008.
DOI: https://doi.org/10.1016/j.rama.2022.01.008

Interpretive Summary: Grasslands experience frequent periods of drought that reduce coverage of soil with living (green) vegetation. Vegetation recovery from drought likely depends on the spatial extent and sizes of patches in which plant coverage is most greatly reduced by water limitation. However, it remains a challenge to measure plant cover over both relatively large areas and at the sufficiently small spatial resolution (grain) required to adequately characterize drought effects on grasslands. We used airborne (unmanned aerial vehicle; UAV) remote sensing to measure green plant cover at 8 cm (approximately 3 inches) spatial resolution and at a spatial scale (tens of hectares) relevant to grassland management in relatively high precipitation (mesic) regions. From these measurements, we determined the aerial extent and patch sizes of mesic grassland in central Texas, USA in which green plant cover was eliminated during drought in each of two cattle grazing treatments (rotational, continuous). Cattle are moved among pastures in the rotational system such that each rotationally grazed pasture was grazed for 1 month annually. Cattle had year-round access to pastures included in the continuous treatment. During the 2 years following drought, we remotely estimated plant biomass to determine grazing treatment effects on plant recovery in patches that lacked vs. retained green plant cover during drought (not green vs. green). Drought eliminated green plant cover from 5-12% of grassland area. Greater than 40% of the not-green area during drought occurred in large (>30 m2) patches across the two grazing treatments combined. Plant biomass recovered quickly from drought. Within two years, remote estimates of plant biomass were similar (continuous grazing) or greater (rotational grazing) in previously not-green than green patches. Vegetation in previously not-green patches recovered rapidly because relative abundances of early-season annual plants species increased in patches in which green cover was eliminated during drought. Biomass recovery was smaller in large compared to small (<1 m2) patches that lacked green cover during drought and in the continuous vs. rotational grazing treatment. Our results imply that managers can speed drought recovery in mesic grassland by 1) managing cattle so as to reduce the patch sizes of grassland that suffer complete loss of green plant cover during drought, and 2) employing rotational or similar grazing systems that eliminate grazing pressure for several consecutive months each year following drought.

Technical Abstract: Drought is frequent on grasslands, often creating patches with greatly reduced coverage of living (green) vegetation that, in turn, influence the trajectory of vegetation recovery following drought. Limiting the extent and sizes of low-cover patches thus should be a target of adaptive management on grazed grasslands, but quantifying patch sizes requires pasture (paddock)-level measurements at small spatial resolution (grain). We used airborne (unmanned aerial vehicle; UAV) remote sensing to measure green plant cover at a spatial grain of 8 cm ground sample distance. From these measurements, we quantified the aerial extent and patch sizes of mesic grassland in central Texas, USA in which green plant cover was eliminated during drought in each of two grazing treatments (rotational, continuous). Remote measurements of the normalized difference vegetation index (NDVI), indicative of plant biomass, then were used to compare post-drought recovery of vegetation in patches that lacked vs. retained green plant cover during drought (not green vs. green). Seasonal drought eliminated green plant cover from 5-12% of grassland area. Across grazing treatments, 43% of the not-green area during drought occurred in large (>30 m2) patches. NDVI recovered quickly from drought, however. Within two years, seasonal mean values of NDVI were similar or greater in previously not-green than green patches. Vegetation in previously not-green patches recovered rapidly because of a marked increase in relative abundances of early-season annual species. Recovery in NDVI was reduced in large compared to small (<1 m2) patches that lacked green cover during drought and in the continuous vs. rotational grazing treatment. Our results imply that managers can facilitate drought recovery in mesic grassland by 1) adaptively managing grazing so as to reduce the patch sizes of grassland that suffer complete loss of green plant cover, and 2) eliminating grazing pressure for several consecutive months annually following drought.